From geoff at iconia.com Fri Jan 3 08:28:17 2025 From: geoff at iconia.com (the keyboard of geoff goodfellow) Date: Fri, 3 Jan 2025 09:28:17 -0700 Subject: [ih] NATO Plans an Orbital Backup Internet Using Satellite Broadband (IEEE Spectrum) In-Reply-To: References: Message-ID: *An undersea cable breach would reroute to satellites* EXCERPT: ON 18 FEBRUARY 2024, a missile attack from the Houthi militants in Yemen hit the cargo ship Rubymar in the Red Sea. With the crew evacuated, the disabled ship would take weeks to finally sink, becoming an symbol for the security of the global Internet in the process. Before it went down, the ship dragged its anchor behind it over an estimated 70 kilometers. The meandering anchor wound up severing three fiber-optic cables across the Red Sea floor, which carried about a quarter of all the Internet traffic between Europe and Asia. Data transmissions had to be rerouted as system engineers realized the cables had been damaged. So this year, NATO, the North Atlantic Treaty Organization, will begin testing a plan to fix the vulnerability that the Rubymar?s sinking so vividly illustrated. The world?s submarine fiber-optic lines carry more than 95 percent of intercontinental Internet communications. These tiny, drawn-out strands of glass fiber stretch some 1.2 million km around the planet, each line with the potential to become its own delicate choke point. Between 500 and 600 cables crisscross ocean floors worldwide. ?They?re not buried when they cross an ocean,? says Tim Stronge, vice president of research at the telecommunications consulting firm TeleGeography. ?They?re sitting right on the seafloor, and at oceanic depths, at deep-sea depths, they?re about this thick??he makes a circle with his fingers??less than a garden hose. They?re fragile.? Undersea fiber-optic cables, by some estimates, are used for more than US $10 trillion in financial transactions every day, as well as encrypted defense communications and other digital communications. If one sinking ship could accidentally take out a portion of global data transmission, what could happen in an organized attack by a determined government? Enter NATO, which has now launched a pilot project to figure out how best to protect global Internet traffic and redirect it when there?s trouble. The project is called HEIST, short for hybrid space-submarine architecture ensuring infosec of telecommunications. (?Infosec? is short for ?information security.?) [...] https://spectrum.ieee.org/undersea-internet-cables-nato -- Geoff.Goodfellow at iconia.com living as The Truth is True From venture37 at geeklan.co.uk Fri Jan 3 09:23:51 2025 From: venture37 at geeklan.co.uk (Sevan Janiyan) Date: Fri, 3 Jan 2025 17:23:51 +0000 Subject: [ih] NATO Plans an Orbital Backup Internet Using Satellite Broadband (IEEE Spectrum) In-Reply-To: References: Message-ID: <5cc265e8-b265-49ab-936e-83fcef153c0f@geeklan.co.uk> On 03/01/2025 16:28, the keyboard of geoff goodfellow via Internet- history wrote: > So this year, NATO, the North Atlantic Treaty Organization, will > begin testing a plan to fix the vulnerability that the Rubymar?s > sinking so vividly illustrated. By coincidence, mention of John Perry Barlow's A Declaration of the Independence of Cyberspace showed up in my social bubble yesterday. The irony of NATO being involved. :) Sevan From touch at strayalpha.com Fri Jan 3 09:30:20 2025 From: touch at strayalpha.com (touch at strayalpha.com) Date: Fri, 3 Jan 2025 09:30:20 -0800 Subject: [ih] NATO Plans an Orbital Backup Internet Using Satellite Broadband (IEEE Spectrum) In-Reply-To: References: Message-ID: A reminder from your list manager: This list is for postings related to Internet *history*. There are other venues for news and news-related discussions. Joe ? Dr. Joe Touch, temporal epistemologist www.strayalpha.com From mgrant at grant.org Tue Jan 7 05:47:25 2025 From: mgrant at grant.org (Michael Grant) Date: Tue, 07 Jan 2025 13:47:25 +0000 Subject: [ih] The netmask Message-ID: Before CIDR sub-netting there were fixed subnets: A, B, C, D, & E. (and from memory D and E came later). What was the rational for this being represented as an actual bit-mask which could have been represented as a number of bits like we do today? I know that not many protocols send the mask over the wire, aside from perhaps routing protocols. Did any early protocols use say just 5 or even just 2 bits to represent classes before things went to CIDR? I never saw anything like ifconfig report "Class C", it was always represented as 255.255.255.0. I realize it's more efficient from a computing point of view to deal with bit-masks. But I'm curious, from a historic point of view, why it wasn't just a number of bits or even just a number representing the class (A, B, C)? In the old days when every byte of memory was sacred, it seems like it would have been thought of as wasteful. From craig at tereschau.net Tue Jan 7 06:01:53 2025 From: craig at tereschau.net (Craig Partridge) Date: Tue, 7 Jan 2025 07:01:53 -0700 Subject: [ih] The netmask In-Reply-To: References: Message-ID: ARPANET IMPs had 24-bit addresses, which perfectly fit with a class A address. ARCNET and the V2LNI Ring (which I think borrowed from an early ring network?) both had 8 bit LAN addresses. I don't know what ALOHA used but wouldn't be surprised if it was 8 bits too. Craig On Tue, Jan 7, 2025 at 6:47?AM Michael Grant via Internet-history < internet-history at elists.isoc.org> wrote: > Before CIDR sub-netting there were fixed subnets: A, B, C, D, & E. (and > from memory D and E came later). What was the rational for this being > represented as an actual bit-mask which could have been represented as a > number of bits like we do today? I know that not many protocols send > the mask over the wire, aside from perhaps routing protocols. Did any > early protocols use say just 5 or even just 2 bits to represent classes > before things went to CIDR? I never saw anything like ifconfig report > "Class C", it was always represented as 255.255.255.0. > > I realize it's more efficient from a computing point of view to deal > with bit-masks. But I'm curious, from a historic point of view, why it > wasn't just a number of bits or even just a number representing the > class (A, B, C)? In the old days when every byte of memory was sacred, > it seems like it would have been thought of as wasteful. > -- > Internet-history mailing list > Internet-history at elists.isoc.org > https://elists.isoc.org/mailman/listinfo/internet-history > -- ***** Craig Partridge's email account for professional society activities and mailing lists. From vint at google.com Tue Jan 7 06:06:05 2025 From: vint at google.com (Vint Cerf) Date: Tue, 7 Jan 2025 09:06:05 -0500 Subject: [ih] The netmask In-Reply-To: References: Message-ID: the original design just used 8 bits for network ID and 24 bits for host ID. When LANS proliferated we went to class A,B,C,D,E structure. When class B became exhausted we went to CIDR concept. v On Tue, Jan 7, 2025 at 8:47?AM Michael Grant via Internet-history < internet-history at elists.isoc.org> wrote: > Before CIDR sub-netting there were fixed subnets: A, B, C, D, & E. (and > from memory D and E came later). What was the rational for this being > represented as an actual bit-mask which could have been represented as a > number of bits like we do today? I know that not many protocols send > the mask over the wire, aside from perhaps routing protocols. Did any > early protocols use say just 5 or even just 2 bits to represent classes > before things went to CIDR? I never saw anything like ifconfig report > "Class C", it was always represented as 255.255.255.0. > > I realize it's more efficient from a computing point of view to deal > with bit-masks. But I'm curious, from a historic point of view, why it > wasn't just a number of bits or even just a number representing the > class (A, B, C)? In the old days when every byte of memory was sacred, > it seems like it would have been thought of as wasteful. > -- > Internet-history mailing list > Internet-history at elists.isoc.org > https://elists.isoc.org/mailman/listinfo/internet-history > -- Please send any postal/overnight deliveries to: Vint Cerf Google, LLC 1900 Reston Metro Plaza, 16th Floor Reston, VA 20190 +1 (571) 213 1346 until further notice From erey at ernw.de Tue Jan 7 06:06:25 2025 From: erey at ernw.de (Enno Rey) Date: Tue, 7 Jan 2025 15:06:25 +0100 Subject: [ih] The netmask In-Reply-To: References: Message-ID: Hi, you may find the answer in this post or one of the links referenced there: https://insinuator.net/2019/08/a-brief-history-of-the-ipv4-address-space/ cheers, Enno On Tue, Jan 07, 2025 at 01:47:25PM +0000, Michael Grant via Internet-history wrote: > Before CIDR sub-netting there were fixed subnets: A, B, C, D, & E. (and > from memory D and E came later). What was the rational for this being > represented as an actual bit-mask which could have been represented as a > number of bits like we do today? I know that not many protocols send the > mask over the wire, aside from perhaps routing protocols. Did any early > protocols use say just 5 or even just 2 bits to represent classes before > things went to CIDR? I never saw anything like ifconfig report "Class C", > it was always represented as 255.255.255.0. > > I realize it's more efficient from a computing point of view to deal with > bit-masks. But I'm curious, from a historic point of view, why it wasn't > just a number of bits or even just a number representing the class (A, B, > C)? In the old days when every byte of memory was sacred, it seems like it > would have been thought of as wasteful. > -- > Internet-history mailing list > Internet-history at elists.isoc.org > https://elists.isoc.org/mailman/listinfo/internet-history -- Enno Rey Cell: +49 173 6745902 Twitter: @Enno_Insinuator IPv6 Blog: https://theinternetprotocolblog.wordpress.com From awalding at gmail.com Tue Jan 7 06:06:51 2025 From: awalding at gmail.com (Andrew Walding) Date: Tue, 7 Jan 2025 08:06:51 -0600 Subject: [ih] The netmask In-Reply-To: References: Message-ID: Hi Michael, Before the classful system, the first octet was the network address, and the three remaining octets were the hosts. That's right, in the experiment we now call the internet, no one could see a need for more than 256 networks! That all changed of course, but the recognition that 256 networks was not going to be enough I think can be attributed to Internet Engineering Note 46 written by David Clark and Danny Cohen around 1976-1978. They basically concluded that there was going to be an exhaustion of the 256 network boundary. So in 1981 John Postel's famous RFC 790 introduced the classful system you refer to. From today's perspective it is a simple robbing of the first 4 bits of the 32 bit address to indicate what octets demarc the network address boundary: 0 = Class A - this left 1-127 for all the class A's 1 0 = Class B - this left 128-191 for the class B's 1 1 0 = Class C - this left 192-223 for class C's 1 1 1 0 = Class D - this used 224 and up for Multicast - well not really because Class E as you mention was also later specified, but simply reserved, not used. This system served the internet for 12 years or so. Not bad, not as you say, not efficient. Then as you mention we went to a more efficient use of the bit boundary with subnet and supernet masks, rather that the byte boundary that the first four bits specified in RFC 790. One minor thing I would like to correct is that the mask is only communicated in routing messages, no data packets ever carry the mask. Hope that helps. Andy On Tue, Jan 7, 2025 at 7:47?AM Michael Grant via Internet-history < internet-history at elists.isoc.org> wrote: > Before CIDR sub-netting there were fixed subnets: A, B, C, D, & E. (and > from memory D and E came later). What was the rational for this being > represented as an actual bit-mask which could have been represented as a > number of bits like we do today? I know that not many protocols send > the mask over the wire, aside from perhaps routing protocols. Did any > early protocols use say just 5 or even just 2 bits to represent classes > before things went to CIDR? I never saw anything like ifconfig report > "Class C", it was always represented as 255.255.255.0. > > I realize it's more efficient from a computing point of view to deal > with bit-masks. But I'm curious, from a historic point of view, why it > wasn't just a number of bits or even just a number representing the > class (A, B, C)? In the old days when every byte of memory was sacred, > it seems like it would have been thought of as wasteful. > -- > Internet-history mailing list > Internet-history at elists.isoc.org > https://elists.isoc.org/mailman/listinfo/internet-history > -- Best Regards, Andy Walding (cell: 214-405-3708) CONFIDENTIALITY NOTICE: The contents of this email message and any attachments are intended solely for the addressee(s) and may contain confidential and/or privileged information and may be legally protected from disclosure. If you are not the intended recipient of this message or their agent, or if this message has been addressed to you in error, please immediately alert the sender by reply email and then delete this message and any attachments. If you are not the intended recipient, you are hereby notified that any use, dissemination, copying, or storage of this message or its attachments is strictly prohibited. From jeanjour at comcast.net Tue Jan 7 06:11:52 2025 From: jeanjour at comcast.net (John Day) Date: Tue, 7 Jan 2025 09:11:52 -0500 Subject: [ih] The netmask In-Reply-To: References: Message-ID: Because networks come in all sorts of sizes and 255 is pretty small and it is along way to 65535 and lots of networks were in that range and it didn?t eat as much address space and of course the next gap was even larger. It also is a part of the move to make IP addresses be addresses, i.e., location-dependent and route-independent, and arresting router table growth. Prior to CIDR IP addresses were just flat identifiers. Take care, John Day > On Jan 7, 2025, at 08:47, Michael Grant via Internet-history wrote: > > Before CIDR sub-netting there were fixed subnets: A, B, C, D, & E. (and from memory D and E came later). What was the rational for this being represented as an actual bit-mask which could have been represented as a number of bits like we do today? I know that not many protocols send the mask over the wire, aside from perhaps routing protocols. Did any early protocols use say just 5 or even just 2 bits to represent classes before things went to CIDR? I never saw anything like ifconfig report "Class C", it was always represented as 255.255.255.0. > > I realize it's more efficient from a computing point of view to deal with bit-masks. But I'm curious, from a historic point of view, why it wasn't just a number of bits or even just a number representing the class (A, B, C)? In the old days when every byte of memory was sacred, it seems like it would have been thought of as wasteful. > -- > Internet-history mailing list > Internet-history at elists.isoc.org > https://elists.isoc.org/mailman/listinfo/internet-history From craig at tereschau.net Tue Jan 7 07:49:35 2025 From: craig at tereschau.net (Craig Partridge) Date: Tue, 7 Jan 2025 08:49:35 -0700 Subject: [ih] The netmask In-Reply-To: References: Message-ID: On Tue, Jan 7, 2025 at 7:12?AM John Day via Internet-history < internet-history at elists.isoc.org> wrote: > > Prior to CIDR IP addresses were just flat identifiers. > > We routed on the network part from the start. And while officially the host part may have been flat, in practice it wasn't. Everyone in network operations knew how to read net 10 addresses to figure out which IMP and port the host was on (and thus who to call if things went wrong). As I recall, BBN's (and I think MITs and others) class B addresses used one byte of the host part for a subnet identifier (my first machine was on 128.89.1.XX -- which was a coax Ethernet hung off the first internal port of the BBN router in Building 6). Craig -- ***** Craig Partridge's email account for professional society activities and mailing lists. From jeanjour at comcast.net Tue Jan 7 08:23:42 2025 From: jeanjour at comcast.net (John Day) Date: Tue, 7 Jan 2025 11:23:42 -0500 Subject: [ih] The netmask In-Reply-To: References: Message-ID: <45E5E84B-B7CD-4A69-A6E7-5A8AEA899E6A@comcast.net> Yes, the network part was, in essence, a flat address space. (Remember the Internet had already lost the Internet Layer.) The blocks of IP addresses were handed out more or less in order regardless of who was requesting the block. Every block allocated required another entry in the router table. The network part did not exhibit any concept of ?nearness? or locality. Until they went to assigning large blocks to Tier 1 providers and people had to get new blocks from their provider, the addresses did not exhibit locality and hence were not really addresses any more than MAC addresses were addresses. (Dalal in his paper on MAC addresses says that they knew the MAC address was really a device-id.) And of course by this time, the ?host? part wasn?t really just host addresses, i.e., enumerating the hosts, but if they were smart, it was the address space of the organization?s network that was assigned the block. And they assigned the addresses hierarchically to reflect nearness within their network to simplify routing. In effect, the ?host? part was the network address and the ?network? part was the internet address, instead they had have AS numbers for that. Bottom line: Addresses belong to layers, not protocols. Protocols just carry addresses. > On Jan 7, 2025, at 10:49, Craig Partridge wrote: > > > > On Tue, Jan 7, 2025 at 7:12?AM John Day via Internet-history > wrote: >> >> Prior to CIDR IP addresses were just flat identifiers. >> > > We routed on the network part from the start. And while officially the host part may have been flat, in practice it wasn't. Everyone in network operations knew how to read net 10 addresses to figure out which IMP and port the host was on (and thus who to call if things went wrong). As I recall, BBN's (and I think MITs and others) class B addresses used one byte of the host part for a subnet identifier (my first machine was on 128.89.1.XX -- which was a coax Ethernet hung off the first internal port of the BBN router in Building 6). > > Craig > > -- > ***** > Craig Partridge's email account for professional society activities and mailing lists. From agmalis at gmail.com Tue Jan 7 10:02:43 2025 From: agmalis at gmail.com (Andrew G. Malis) Date: Tue, 7 Jan 2025 13:02:43 -0500 Subject: [ih] The netmask In-Reply-To: References: Message-ID: Michael, RFC 1519 goes into a lot of detail regarding the history of IP addressing and the design decisions that led to CIDR. While the RFC refers to bit-masks, operationally people realized that since CIDR bit-masks were always all ones from the left and all zeros from the right, they could be equivalently represented by an integer count of one bits, usually following a "/" character. So then we ended up with tables like https://docs.netgate.com/pfsense/en/latest/network/cidr.html . Cheers Andy On Tue, Jan 7, 2025 at 8:47?AM Michael Grant via Internet-history < internet-history at elists.isoc.org> wrote: > Before CIDR sub-netting there were fixed subnets: A, B, C, D, & E. (and > from memory D and E came later). What was the rational for this being > represented as an actual bit-mask which could have been represented as a > number of bits like we do today? I know that not many protocols send > the mask over the wire, aside from perhaps routing protocols. Did any > early protocols use say just 5 or even just 2 bits to represent classes > before things went to CIDR? I never saw anything like ifconfig report > "Class C", it was always represented as 255.255.255.0. > > I realize it's more efficient from a computing point of view to deal > with bit-masks. But I'm curious, from a historic point of view, why it > wasn't just a number of bits or even just a number representing the > class (A, B, C)? In the old days when every byte of memory was sacred, > it seems like it would have been thought of as wasteful. > -- > Internet-history mailing list > Internet-history at elists.isoc.org > https://elists.isoc.org/mailman/listinfo/internet-history > From karl at iwl.com Tue Jan 7 12:49:55 2025 From: karl at iwl.com (Karl Auerbach) Date: Tue, 7 Jan 2025 12:49:55 -0800 Subject: [ih] The netmask In-Reply-To: References: Message-ID: As folks discuss this, let's not forget that interwoven with the subnet-mask and CIDR ideas was the question whether such masks had to be contiguous bits. (I'm rather glad that we went with contiguous bits - otherwise would have perhaps led to a lot of chaos.) (I remember at a San Diego IETF when Tony Li first informed me of CIDR ideas.) --karl-- On 1/7/25 5:47 AM, Michael Grant via Internet-history wrote: > Before CIDR sub-netting there were fixed subnets: A, B, C, D, & E.? (and > from memory D and E came later).? What was the rational for this being > represented as an actual bit-mask which could have been represented as a > number of bits like we do today?? I know that not many protocols send > the mask over the wire, aside from perhaps routing protocols.? Did any > early protocols use say just 5 or even just 2 bits to represent classes > before things went to CIDR?? I never saw anything like ifconfig report > "Class C", it was always represented as 255.255.255.0. > > I realize it's more efficient from a computing point of view to deal > with bit-masks.? But I'm curious, from a historic point of view, why it > wasn't just a number of bits or even just a number representing the > class (A, B, C)?? In the old days when every byte of memory was sacred, > it seems like it would have been thought of as wasteful. From gtaylor at tnetconsulting.net Tue Jan 7 13:19:44 2025 From: gtaylor at tnetconsulting.net (Grant Taylor) Date: Tue, 7 Jan 2025 15:19:44 -0600 Subject: [ih] The netmask In-Reply-To: References: Message-ID: <234bb038-f598-5b21-9074-1ed3f59475fc@tnetconsulting.net> On 1/7/25 8:06?AM, Andrew Walding via Internet-history wrote: > So in 1981 John Postel's famous RFC 790 introduced the classful system > you refer to. From today's perspective it is a simple robbing of > the first 4 bits of the 32 bit address to indicate what octets demarc > the network address boundary: Is it really robbing the first 4 bits for all classes? I thought it was robbing the: first 1 bit for class A, first 2 bits for class B, first 3 bits for class C, and first 4 bits for class D (and E). If it was really robbing 4 bits for all classes, then we could have 16 classes. ;-) > This system served the internet for 12 years or so. Not bad, not as > you say, not efficient. Didn't this system just use a "netmask"? class A had 8-bit netmask class B had 16-bit netmask class C had 24-bit netmask > Then as you mention we went to a more efficient use of the bit boundary > with subnet and supernet masks, rather that the byte boundary that > the first four bits specified in RFC 790. Wasn't the sub-netmask introduced with CIDR? As in the part that subdivided the network portion that was already divided based on class? I've run into extremely obtuse IP configurations on things designed very early on in CIDR's lifetime wherein the sub-net is the number of bits /after/ the *netmask*. E.g. 192.0.2.0/24 (Tnet-Net-1) has a netmask of 255.255.0.0 and a sub-netmask of 0.0.255.0. Yes, I am thinking about OS/390's TCP/IP configuration. In retrospect, I've seen hints of this in other systems too. I think that some Cisco IOS output hints at this if you know how to squint at it to see the pattern. > One minor thing I would like to correct is that the mask is only > communicated in routing messages, no data packets ever carry the mask. Arguably, routing (their protocols) and security devices / software are the only thing that /need/ to know anything about the (sub)netmask. -- Grant. . . . unix || die From gtaylor at tnetconsulting.net Tue Jan 7 13:24:01 2025 From: gtaylor at tnetconsulting.net (Grant Taylor) Date: Tue, 7 Jan 2025 15:24:01 -0600 Subject: [ih] The netmask In-Reply-To: References: Message-ID: <6e39d905-fc86-63af-36a9-68ca55f2be68@tnetconsulting.net> On 1/7/25 12:02?PM, Andrew G. Malis via Internet-history wrote: > While the RFC refers to bit-masks, operationally people realized that > since CIDR bit-masks were always all ones from the left and all zeros > from the right, they could be equivalently represented by an integer > count of one bits, usually following a "/" character. I don't know if CIDR itself codified (sub)netmasks being sequential ones or not. But I do know that some older TCP/IP implementations would work with netmasks that were non-contiguous ones. I remember some early discussions about being EXTREMELY creative with non-contiguous bit patterns in netmasks in the mid-to-late '90s. In retrospect, I'm glad that we settled on the simpler convention of contiguous ones. -- Grant. . . . unix || die From matt.mathis at gmail.com Tue Jan 7 15:37:14 2025 From: matt.mathis at gmail.com (Matt Mathis) Date: Tue, 7 Jan 2025 15:37:14 -0800 Subject: [ih] The netmask In-Reply-To: <6e39d905-fc86-63af-36a9-68ca55f2be68@tnetconsulting.net> References: <6e39d905-fc86-63af-36a9-68ca55f2be68@tnetconsulting.net> Message-ID: I believe Philip Almquist explicitly studied tradeoffs in mask designs, and that his wisdom is embodied in RFC 1380. However that document does not mention discarded designs. The point I remember from a hallway conversation was about the computational complexity of resolving "longest match" (or some other route selection algorithm) with non-contiguous masks. It may be that this point alone precluded all designs other than contiguous masks. Look for Archives of the IETF ROAD (Routing + Addressing) group. Thanks, --MM-- Evil is defined by mortals who think they know "The Truth" and use force to apply it to others. ------------------------------------------- Matt Mathis (Email is best) Home & mobile: 412-654-7529 please leave a message if you must call. On Tue, Jan 7, 2025 at 1:24?PM Grant Taylor via Internet-history < internet-history at elists.isoc.org> wrote: > On 1/7/25 12:02?PM, Andrew G. Malis via Internet-history wrote: > > While the RFC refers to bit-masks, operationally people realized that > > since CIDR bit-masks were always all ones from the left and all zeros > > from the right, they could be equivalently represented by an integer > > count of one bits, usually following a "/" character. > > I don't know if CIDR itself codified (sub)netmasks being sequential ones > or not. But I do know that some older TCP/IP implementations would work > with netmasks that were non-contiguous ones. > > I remember some early discussions about being EXTREMELY creative with > non-contiguous bit patterns in netmasks in the mid-to-late '90s. In > retrospect, I'm glad that we settled on the simpler convention of > contiguous ones. > > > > -- > Grant. . . . > unix || die > > -- > Internet-history mailing list > Internet-history at elists.isoc.org > https://elists.isoc.org/mailman/listinfo/internet-history > From lyndon at orthanc.ca Tue Jan 7 16:14:40 2025 From: lyndon at orthanc.ca (Lyndon Nerenberg (VE7TFX/VE6BBM)) Date: Tue, 07 Jan 2025 16:14:40 -0800 Subject: [ih] The netmask In-Reply-To: References: Message-ID: <164d79fbf62547ca@orthanc.ca> Michael Grant: > While the RFC refers to bit-masks, operationally people realized > that since CIDR bit-masks were always all ones from the left and > all zeros from the right, they could be equivalently represented > by an integer count of one bits, usually following a "/" character. > So then we ended up with tables [...] My admittedly broken memory recalls a period where the netmask did not have to be contiguous bits, thus 255.255.192.192 was a valid mask. This was then changed to add the restriction the mask bits had to be contiguous, after which the /nn representation became viable. If this is true, it explains why we have the dotted quad metmask format. So, am I hallucinating this? --lyndon From lyndon at orthanc.ca Tue Jan 7 16:27:32 2025 From: lyndon at orthanc.ca (Lyndon Nerenberg (VE7TFX/VE6BBM)) Date: Tue, 07 Jan 2025 16:27:32 -0800 Subject: [ih] The netmask In-Reply-To: <35712F85-A96F-450F-A957-A32AE8931B30@tony.li> References: <164d79fbf62547ca@orthanc.ca> <35712F85-A96F-450F-A957-A32AE8931B30@tony.li> Message-ID: <164d7a1787180917@orthanc.ca> Tony Li writes: > No, you are not. The original BSD implementation of subnetting did = > support discontiguous subnet masks. See the radix trie implementation = > in the BSD source code if you want a headache. ... and I remember reading about this in the (probably 3.5) SunOS documentation, so it all makes sense now. --lyndon From alejandroacostaalamo at gmail.com Tue Jan 7 18:33:31 2025 From: alejandroacostaalamo at gmail.com (Alejandro Acosta) Date: Tue, 7 Jan 2025 22:33:31 -0400 Subject: [ih] The netmask In-Reply-To: References: Message-ID: Hello all, ? I have been reading all the emails in this thread. So, I decided to try something "interesting". ? I have taken note of every RFC that have been mentioned in this discussion, I searched for the URLs. I also got every URL for the thread in the mailman; I added all these links to Google NotebookLM and it generated a quite interesting notebook (unfortunately I can not share it with everyone as google drive allows). ? Why am I saying this?. I'm thinking in publishing a blog post called something like: "the story behind the netmask" or something like that. ? I promise I will point to the mailing list and RFCs (if any of you want to join me in the blog post as author feel free to let me know) Thanks, Alejandro Acosta R+D Coordinator at LACNIC On 7/1/25 9:47 AM, Michael Grant via Internet-history wrote: > Before CIDR sub-netting there were fixed subnets: A, B, C, D, & E. > (and from memory D and E came later).? What was the rational for this > being represented as an actual bit-mask which could have been > represented as a number of bits like we do today?? I know that not > many protocols send the mask over the wire, aside from perhaps routing > protocols.? Did any early protocols use say just 5 or even just 2 bits > to represent classes before things went to CIDR?? I never saw anything > like ifconfig report "Class C", it was always represented as > 255.255.255.0. > > I realize it's more efficient from a computing point of view to deal > with bit-masks.? But I'm curious, from a historic point of view, why > it wasn't just a number of bits or even just a number representing the > class (A, B, C)?? In the old days when every byte of memory was > sacred, it seems like it would have been thought of as wasteful. From mgrant at grant.org Wed Jan 8 01:49:37 2025 From: mgrant at grant.org (Michael Grant) Date: Wed, 08 Jan 2025 09:49:37 +0000 Subject: [ih] The netmask In-Reply-To: <234bb038-f598-5b21-9074-1ed3f59475fc@tnetconsulting.net> References: <234bb038-f598-5b21-9074-1ed3f59475fc@tnetconsulting.net> Message-ID: >From "Grant Taylor via Internet-history" >I've run into extremely obtuse IP configurations on things designed very early on in CIDR's lifetime wherein the sub-net is the number of bits /after/ the *netmask*. E.g. 192.0.2.0/24 (Tnet-Net-1) has a netmask of 255.255.0.0 and a sub-netmask of 0.0.255.0. > >Yes, I am thinking about OS/390's TCP/IP configuration. > >In retrospect, I've seen hints of this in other systems too. I think that some Cisco IOS output hints at this if you know how to squint at it to see the pattern. Might this have been useful in cases where an internal network of machines was multi-homed on 2 separate networks? As in, it had 2 separate prefixes but for simplicity the internal network and host numbering was the same and the internal routing was the masked bits? You wouldn't need this today with either NAT or a globally unique IP address and routing. From gtaylor at tnetconsulting.net Wed Jan 8 06:49:22 2025 From: gtaylor at tnetconsulting.net (Grant Taylor) Date: Wed, 8 Jan 2025 08:49:22 -0600 Subject: [ih] The netmask In-Reply-To: References: <234bb038-f598-5b21-9074-1ed3f59475fc@tnetconsulting.net> Message-ID: <4cb98c46-9256-479c-8636-c1f39c2d87c5@spamtrap.tnetconsulting.net> On 1/8/25 03:49, Michael Grant via Internet-history wrote: > Might this have been useful in cases where an internal network of > machines was multi-homed on 2 separate networks?? As in, it had 2 > separate prefixes but for simplicity the internal network and host > numbering was the same and the internal routing was the masked bits? You > wouldn't need this today with either NAT or a globally unique IP address > and routing. Either I'm not understanding what you're suggesting or we're talking about two different things. The OS/390 bit was specifically about what we would consider to be the 24 contiguous bits for the /24 netmask. The 192.0.2.0/24 network is in the class B range, thus it has a network mask of 255.255.0.0. However we want it to be a subset of that network (mas), thus a sub-net(mask) therein. 11000000.00000000.00000010.00000000 192.0.2.0 NNNNNNNN.NNNNNNNN.xxxxxxxx.xxxxxxxx network mask xxxxxxxx.xxxxxxxx.SSSSSSSS.xxxxxxxx sub-network mask xxxxxxxx.xxxxxxxx.xxxxxxxx.HHHHHHHH host bits NNNNNNNN.NNNNNNNN.SSSSSSSS.HHHHHHHH combination of network mask, sub-network mask, and host bits Conversely for 10.0.0.0/24 00001010.00000000.00000000.00000000 10.0.0.0 NNNNNNNN.xxxxxxxx.xxxxxxxx.xxxxxxxx network mask xxxxxxxx.SSSSSSSS.SSSSSSSS.xxxxxxxx sub-network mask xxxxxxxx.xxxxxxxx.xxxxxxxx.HHHHHHHH host bits NNNNNNNN.SSSSSSSS.SSSSSSSS.HHHHHHHH combination of network mask, sub-network mask, and host bits In OS/390 (et al.) the sub-net(work) mask is a separate configuration from the net(work) mask which is based on the class of the IP address. At least this is, and has been, my working understanding of OS/390's TCP/IP configuration on my P/390-E. The point that I'm trying to emphasize is that the sub-net(work) mask is a separate configuration value than the net(work) mask. The net(work) mask and the sub-net(work) mask are combined to get what we now consider to be one aggregate thing and call sub-net / net-mask in common parlance. I suspect this harks back to an older version of the TCP/IP stack that only had support for the net(work) mask. I speculate that newer versions of the software wanted to be compatible with configuration files for older versions and as such added the sub-net(work) portion as an additional parameter. But I've seen reference to the difference in net(work) mask and sub-net(work) mask in other things. We can also see some hints of this separation of net(work) mask ans bub-net(work) mask in Cisco IOS wherein `show ip route` groups multiple sub-et(work) prefixes under their net(work) prefix. E.g. 17.0.0.0/8 is variably subnetted ... 17.43.58.0/24 [200/0] via ... ... 170.187.0.0/16 is variably subnetted ... 170.187.6.0/23 [200/0] via ... ... 216.152.65.0/24 [200/0] via ... 136.237.0.0/24 is subnetted ... 136.237.45.0 [200/0] via ... ... So the sub-net(work) being separate from the net(work) is not strictly isolated to OS/390. IMHO even the names suggest that they are, or at least were / used to be, separate things: "net(work) mask" and "sub-net(work) mask". The sub-net(work) mask wasn't anything like Cisco wildcards like I think you might be alluding to or what we might consider to be host-bits in various IPv6 network prefixes today. All of this is based on my observations and trying to understand them over the years and as such can be anywhere between not quite correct and completely wrong. -- Grant. . . . From gnu at toad.com Wed Jan 8 14:22:06 2025 From: gnu at toad.com (John Gilmore) Date: Wed, 08 Jan 2025 14:22:06 -0800 Subject: [ih] The netmask In-Reply-To: References: Message-ID: <13959.1736374926@hop.toad.com> Andrew G. Malis via Internet-history wrote: > people realized that since CIDR bit-masks were always all ones from > the left and all zeros from the right, they could be equivalently > represented by an integer count of one bits, usually following a "/" > character The Wikipedia page on CIDR says that Phil Karn invented the CIDR "/bitcount" notation in the 1980s, citing two NANOG messages from 2018 with the remembrances of Brian Kantor and Bill Simpson: https://seclists.org/nanog/2018/Dec/334 https://seclists.org/nanog/2018/Dec/368 https://en.wikipedia.org/wiki/Classless_Inter-Domain_Routing#CIDR_notation There's more CIDR history in that NANOG thread, though this week's thread has covered a lot of it. John From agmalis at gmail.com Thu Jan 9 03:29:58 2025 From: agmalis at gmail.com (Andrew G. Malis) Date: Thu, 9 Jan 2025 06:29:58 -0500 Subject: [ih] The netmask In-Reply-To: <13959.1736374926@hop.toad.com> References: <13959.1736374926@hop.toad.com> Message-ID: John, Thanks, I had forgotten it was Phil! Cheers, Andy On Wed, Jan 8, 2025 at 5:22?PM John Gilmore wrote: > Andrew G. Malis via Internet-history > wrote: > > people realized that since CIDR bit-masks were always all ones from > > the left and all zeros from the right, they could be equivalently > > represented by an integer count of one bits, usually following a "/" > > character > > The Wikipedia page on CIDR says that Phil Karn invented the CIDR > "/bitcount" notation in the 1980s, citing two NANOG messages from 2018 > with the remembrances of Brian Kantor and Bill Simpson: > > https://seclists.org/nanog/2018/Dec/334 > https://seclists.org/nanog/2018/Dec/368 > > > https://en.wikipedia.org/wiki/Classless_Inter-Domain_Routing#CIDR_notation > > There's more CIDR history in that NANOG thread, though this week's > thread has covered a lot of it. > > John > > From gregskinner0 at icloud.com Thu Jan 9 19:06:09 2025 From: gregskinner0 at icloud.com (Greg Skinner) Date: Thu, 9 Jan 2025 19:06:09 -0800 Subject: [ih] Fwd: Article about Peter Kirstein References: <1213214822.7832062.1736477904035@mail.yahoo.com> Message-ID: Forwarded for Barbara > ----- Forwarded Message ----- > From: Barbara Denny > To: Internet-history > Sent: Thursday, January 9, 2025 at 06:20:09 PM PST > Subject: Article about Peter Kirstein > > Link provided by Steve Berson. > > https://theconversation.com/how-britain-got-its-first-internet-connection-by-the-late-pioneer-who-created-the-first-password-on-the-internet-45404 > > barbara > From chonkn at gmail.com Thu Jan 9 21:48:15 2025 From: chonkn at gmail.com (kilnam chon) Date: Fri, 10 Jan 2025 14:48:15 +0900 Subject: [ih] Fwd: Article about Peter Kirstein In-Reply-To: References: <1213214822.7832062.1736477904035@mail.yahoo.com> Message-ID: you can also get the article by peter kirstein; Peter Kirstein, "The early history of packet switching in the UK," IEEE Communications Magazine, Feb. 2009, pp. 18-36. chon On Fri, Jan 10, 2025 at 12:06?PM Greg Skinner via Internet-history < internet-history at elists.isoc.org> wrote: > Forwarded for Barbara > > > ----- Forwarded Message ----- > > From: Barbara Denny > > To: Internet-history > > Sent: Thursday, January 9, 2025 at 06:20:09 PM PST > > Subject: Article about Peter Kirstein > > > > Link provided by Steve Berson. > > > > > https://theconversation.com/how-britain-got-its-first-internet-connection-by-the-late-pioneer-who-created-the-first-password-on-the-internet-45404 > > > > barbara > > > > -- > Internet-history mailing list > Internet-history at elists.isoc.org > https://elists.isoc.org/mailman/listinfo/internet-history > From mgrant at grant.org Fri Jan 10 09:33:07 2025 From: mgrant at grant.org (Michael Grant) Date: Fri, 10 Jan 2025 17:33:07 +0000 Subject: [ih] The netmask In-Reply-To: <4cb98c46-9256-479c-8636-c1f39c2d87c5@spamtrap.tnetconsulting.net> References: <234bb038-f598-5b21-9074-1ed3f59475fc@tnetconsulting.net> <4cb98c46-9256-479c-8636-c1f39c2d87c5@spamtrap.tnetconsulting.net> Message-ID: ------ Original Message ------ >From "Grant Taylor via Internet-history" To internet-history at elists.isoc.org Date 08/01/2025 14:49:22 Subject Re: [ih] The netmask >On 1/8/25 03:49, Michael Grant via Internet-history wrote: >>Might this have been useful in cases where an internal network of machines was multi-homed on 2 separate networks? As in, it had 2 separate prefixes but for simplicity the internal network and host numbering was the same and the internal routing was the masked bits? You wouldn't need this today with either NAT or a globally unique IP address and routing. > >Either I'm not understanding what you're suggesting or we're talking about two different things. > >The OS/390 bit was specifically about what we would consider to be the 24 contiguous bits for the /24 netmask. > >The 192.0.2.0/24 network is in the class B range, thus it has a network mask of 255.255.0.0. However we want it to be a subset of that network (mas), thus a sub-net(mask) therein. > >11000000.00000000.00000010.00000000 192.0.2.0 >NNNNNNNN.NNNNNNNN.xxxxxxxx.xxxxxxxx network mask >xxxxxxxx.xxxxxxxx.SSSSSSSS.xxxxxxxx sub-network mask >xxxxxxxx.xxxxxxxx.xxxxxxxx.HHHHHHHH host bits >NNNNNNNN.NNNNNNNN.SSSSSSSS.HHHHHHHH combination of network mask, sub-network mask, and host bits > >Conversely for 10.0.0.0/24 > >00001010.00000000.00000000.00000000 10.0.0.0 >NNNNNNNN.xxxxxxxx.xxxxxxxx.xxxxxxxx network mask >xxxxxxxx.SSSSSSSS.SSSSSSSS.xxxxxxxx sub-network mask >xxxxxxxx.xxxxxxxx.xxxxxxxx.HHHHHHHH host bits >NNNNNNNN.SSSSSSSS.SSSSSSSS.HHHHHHHH combination of network mask, sub-network mask, and host bits > >In OS/390 (et al.) the sub-net(work) mask is a separate configuration from the net(work) mask which is based on the class of the IP address. > >At least this is, and has been, my working understanding of OS/390's TCP/IP configuration on my P/390-E. > >The point that I'm trying to emphasize is that the sub-net(work) mask is a separate configuration value than the net(work) mask. The net(work) mask and the sub-net(work) mask are combined to get what we now consider to be one aggregate thing and call sub-net / net-mask in common parlance. > >I suspect this harks back to an older version of the TCP/IP stack that only had support for the net(work) mask. I speculate that newer versions of the software wanted to be compatible with configuration files for older versions and as such added the sub-net(work) portion as an additional parameter. > >But I've seen reference to the difference in net(work) mask and sub-net(work) mask in other things. > >We can also see some hints of this separation of net(work) mask ans bub-net(work) mask in Cisco IOS wherein `show ip route` groups multiple sub-et(work) prefixes under their net(work) prefix. E.g. > >17.0.0.0/8 is variably subnetted ... > 17.43.58.0/24 [200/0] via ... > ... >170.187.0.0/16 is variably subnetted ... > 170.187.6.0/23 [200/0] via ... > ... >216.152.65.0/24 [200/0] via ... >136.237.0.0/24 is subnetted ... > 136.237.45.0 [200/0] via ... > ... > >So the sub-net(work) being separate from the net(work) is not strictly isolated to OS/390. > >IMHO even the names suggest that they are, or at least were / used to be, separate things: "net(work) mask" and "sub-net(work) mask". > >The sub-net(work) mask wasn't anything like Cisco wildcards like I think you might be alluding to or what we might consider to be host-bits in various IPv6 network prefixes today. > >All of this is based on my observations and trying to understand them over the years and as such can be anywhere between not quite correct and completely wrong. > I see what you're saying. In my thinking it's a little related but not the same thing. And I may be mis-remembering something I misunderstood many years ago. From brian.e.carpenter at gmail.com Fri Jan 10 11:35:53 2025 From: brian.e.carpenter at gmail.com (Brian E Carpenter) Date: Sat, 11 Jan 2025 08:35:53 +1300 Subject: [ih] Fwd: Article about Peter Kirstein In-Reply-To: References: <1213214822.7832062.1736477904035@mail.yahoo.com> Message-ID: What an excellent article; I would have expected nothing less from Peter. Regards Brian Carpenter On 10-Jan-25 16:06, Greg Skinner via Internet-history wrote: > Forwarded for Barbara > >> ----- Forwarded Message ----- >> From: Barbara Denny >> To: Internet-history >> Sent: Thursday, January 9, 2025 at 06:20:09 PM PST >> Subject: Article about Peter Kirstein >> >> Link provided by Steve Berson. >> >> https://theconversation.com/how-britain-got-its-first-internet-connection-by-the-late-pioneer-who-created-the-first-password-on-the-internet-45404 >> >> barbara >> > From jeanjour at comcast.net Fri Jan 10 12:29:52 2025 From: jeanjour at comcast.net (John Day) Date: Fri, 10 Jan 2025 15:29:52 -0500 Subject: [ih] Article about Peter Kirstein In-Reply-To: References: <1213214822.7832062.1736477904035@mail.yahoo.com> Message-ID: <402FBB4C-DE0A-47A8-AAA7-99D4F790A85B@comcast.net> Yes, an excellent article. It is amazing how much resistance there was in the UK to doing this! Unbelievable. Just a minor follow-up, the connection to RAL became quite useful (and illegal). ;-) We were the Univ of Illinois node at the time. The physics dept at Illinois was working closely with Argonne National Laboratory on the south side of Chicago and with the development of FermiLab. They were moving large files from CERN to RAL to Illinois and then driving them up to Argonne It has been too many years but we were either the largest ARPANET user of the RAL machine, or flat out the largest user. (I think they were running some programs at RAL as well.) The illegality? There was not supposed to be any ?international traffic? through the TIP.) ;-) (I remember sitting at my desk in Champaign and connecting to 360/195 no password required. So far as I know, there were no passwords on the TIPs in the US. Although use of the timesharing systems on the ARPANET required passwords (no those weren?t Telnet passwords, they were the system's passwords). So the only real point of the passwords was to keep UK users off and to satisfy the bureaucrats. IOW, to hurt themselves.) ;-) It was a very useful connection to have and I am sure others made a lot of use of it. But it is truly unbelievable how many obstacles the UK government and GPO created to either keep it from happening or make it hard to use. Take care, John > On Jan 10, 2025, at 14:35, Brian E Carpenter via Internet-history wrote: > > What an excellent article; I would have expected nothing less from Peter. > > Regards > Brian Carpenter > > On 10-Jan-25 16:06, Greg Skinner via Internet-history wrote: >> Forwarded for Barbara >>> ----- Forwarded Message ----- >>> From: Barbara Denny >>> To: Internet-history >>> Sent: Thursday, January 9, 2025 at 06:20:09 PM PST >>> Subject: Article about Peter Kirstein >>> >>> Link provided by Steve Berson. >>> >>> https://theconversation.com/how-britain-got-its-first-internet-connection-by-the-late-pioneer-who-created-the-first-password-on-the-internet-45404 >>> >>> barbara >>> > -- > Internet-history mailing list > Internet-history at elists.isoc.org > https://elists.isoc.org/mailman/listinfo/internet-history From brian.e.carpenter at gmail.com Fri Jan 10 13:35:40 2025 From: brian.e.carpenter at gmail.com (Brian E Carpenter) Date: Sat, 11 Jan 2025 10:35:40 +1300 Subject: [ih] Article about Peter Kirstein In-Reply-To: <402FBB4C-DE0A-47A8-AAA7-99D4F790A85B@comcast.net> References: <1213214822.7832062.1736477904035@mail.yahoo.com> <402FBB4C-DE0A-47A8-AAA7-99D4F790A85B@comcast.net> Message-ID: <8db67687-5e81-43e2-a002-9ea1529e3cc6@gmail.com> > They were moving large files from CERN to RAL to Illinois That was before I joined the networking group at CERN, but the CERN-RAL link was 9.6 kbaud and installed during 1972 [0]. Apparently it cost about ?14,000 a year. When it first came to my attention (in 1984) it was running the JANET Coloured Book protocols. Peter Kirstein was always a friend of CERN - he'd actually worked at CERN in 1959 to 1963, the real pioneering days [1]. There's a paper [2] about access control of the London ARPANET node, and one of its authors, David Bates, later worked at CERN too (but not on networking). [0] Rutherford: Computing by telephone, CERN Courier 12(12), pp 421-422, December 1972, https://cds.cern.ch/record/1729427 [1] P. T. Kirstein, The design of injection systems with application to the CERN storage ring mode, CERN-62-04, 1962, http://dx.doi.org/10.5170/CERN-1962-004 [2]A. V. Stokes, D. L. Bates, P. T. Kirstein, Monitoring and access control of the London node of ARPANET, AFIPS '76, 1976, https://dl.acm.org/doi/10.1145/1499799.1499882 Regards Brian Carpenter On 11-Jan-25 09:29, John Day wrote: > Yes, an excellent article. It is amazing how much resistance there was in the UK to doing this! Unbelievable. > > Just a minor follow-up, the connection to RAL became quite useful (and illegal). ;-) We were the Univ of Illinois node at the time. The physics dept at Illinois was working closely with Argonne National Laboratory on the south side of Chicago and with the development of FermiLab. They were moving large files from CERN to RAL to Illinois and then driving them up to Argonne It has been too many years but we were either the largest ARPANET user of the RAL machine, or flat out the largest user. (I think they were running some programs at RAL as well.) The illegality? There was not supposed to be any ?international traffic? through the TIP.) ;-) > > (I remember sitting at my desk in Champaign and connecting to 360/195 no password required. So far as I know, there were no passwords on the TIPs in the US. Although use of the timesharing systems on the ARPANET required passwords (no those weren?t Telnet passwords, they were the system's passwords). So the only real point of the passwords was to keep UK users off and to satisfy the bureaucrats. IOW, to hurt themselves.) ;-) > > It was a very useful connection to have and I am sure others made a lot of use of it. But it is truly unbelievable how many obstacles the UK government and GPO created to either keep it from happening or make it hard to use. > > Take care, > John > >> On Jan 10, 2025, at 14:35, Brian E Carpenter via Internet-history wrote: >> >> What an excellent article; I would have expected nothing less from Peter. >> >> Regards >> Brian Carpenter >> >> On 10-Jan-25 16:06, Greg Skinner via Internet-history wrote: >>> Forwarded for Barbara >>>> ----- Forwarded Message ----- >>>> From: Barbara Denny >>>> To: Internet-history >>>> Sent: Thursday, January 9, 2025 at 06:20:09 PM PST >>>> Subject: Article about Peter Kirstein >>>> >>>> Link provided by Steve Berson. >>>> >>>> https://theconversation.com/how-britain-got-its-first-internet-connection-by-the-late-pioneer-who-created-the-first-password-on-the-internet-45404 >>>> >>>> barbara >>>> >> -- >> Internet-history mailing list >> Internet-history at elists.isoc.org >> https://elists.isoc.org/mailman/listinfo/internet-history > From jnc at mercury.lcs.mit.edu Fri Jan 24 05:52:19 2025 From: jnc at mercury.lcs.mit.edu (Noel Chiappa) Date: Fri, 24 Jan 2025 08:52:19 -0500 (EST) Subject: [ih] Correct name for early TCP/IP working group? Message-ID: <20250124135219.CA54118C0D4@mercury.lcs.mit.edu> So, what was the correct full name for the early engineering group that met every couple of months to work on TCP/IP, before the IETF was set up? (I have looked for it in Jon Postel's excellent minutes of its meetings, preserved in the Internet Experiment Note series, but I havn't found it there.) I have this memory that we called it the 'Internet Working Group', but of course (especially in shortened form) that could be confused with the INWG (the 'International Network Working Group'), a different early group entirly. Noel From galmes at tamu.edu Fri Jan 24 06:08:14 2025 From: galmes at tamu.edu (Guy Almes) Date: Fri, 24 Jan 2025 09:08:14 -0500 Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: <20250124135219.CA54118C0D4@mercury.lcs.mit.edu> References: <20250124135219.CA54118C0D4@mercury.lcs.mit.edu> Message-ID: Hi Noel, You were there long before I was, but I recall that, in the upper left corner of RFCs and Internet Drafts, we were always to put down "Network Working Group". Even though we were organized into an IETF with Working Groups, etc., that old name had to be there for some reason. -- Guy On 1/24/25 8:52 AM, Noel Chiappa via Internet-history wrote: > So, what was the correct full name for the early engineering group that met > every couple of months to work on TCP/IP, before the IETF was set up? (I have > looked for it in Jon Postel's excellent minutes of its meetings, preserved in > the Internet Experiment Note series, but I havn't found it there.) > > I have this memory that we called it the 'Internet Working Group', but of > course (especially in shortened form) that could be confused with the INWG > (the 'International Network Working Group'), a different early group entirly. > > Noel > -- > Internet-history mailing list > Internet-history at elists.isoc.org > https://urldefense.com/v3/__https://elists.isoc.org/mailman/listinfo/ > internet-history__;!!KwNVnqRv! > CmzgJn96jrPDxtRTnatHHTCBN2TAXU8BefHlcjZpgPJuRVLhYsdmbUEVlQ7NPmOnBeigt- > Och0cm2u9EORW6deYdkij7FA$ > From steve at shinkuro.com Fri Jan 24 06:19:27 2025 From: steve at shinkuro.com (Steve Crocker) Date: Fri, 24 Jan 2025 09:19:27 -0500 Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: References: Message-ID: It had to be there because I said so :). See RFC 3. Steve Sent from my iPhone > On Jan 24, 2025, at 9:08?AM, Guy Almes via Internet-history wrote: > > ?Hi Noel, > You were there long before I was, but I recall that, in the upper left corner of RFCs and Internet Drafts, we were always to put down "Network Working Group". > Even though we were organized into an IETF with Working Groups, etc., that old name had to be there for some reason. > -- Guy > >> On 1/24/25 8:52 AM, Noel Chiappa via Internet-history wrote: >> So, what was the correct full name for the early engineering group that met >> every couple of months to work on TCP/IP, before the IETF was set up? (I have >> looked for it in Jon Postel's excellent minutes of its meetings, preserved in >> the Internet Experiment Note series, but I havn't found it there.) >> I have this memory that we called it the 'Internet Working Group', but of >> course (especially in shortened form) that could be confused with the INWG >> (the 'International Network Working Group'), a different early group entirly. >> Noel >> -- >> Internet-history mailing list >> Internet-history at elists.isoc.org >> https://urldefense.com/v3/__https://elists.isoc.org/mailman/listinfo/ internet-history__;!!KwNVnqRv! CmzgJn96jrPDxtRTnatHHTCBN2TAXU8BefHlcjZpgPJuRVLhYsdmbUEVlQ7NPmOnBeigt- Och0cm2u9EORW6deYdkij7FA$ > > -- > Internet-history mailing list > Internet-history at elists.isoc.org > https://elists.isoc.org/mailman/listinfo/internet-history From vint at google.com Fri Jan 24 06:28:58 2025 From: vint at google.com (Vint Cerf) Date: Fri, 24 Jan 2025 09:28:58 -0500 Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: References: Message-ID: so there were a few groups. NWG was Steve's creation and was the nexus in which NCP, FTP, SMTP, TELNET and other Arpanet-based protocols were developed. In 1972, an International Network Working Group (INWG) was created and had its own set of notes called INWG Notes - these have all been digitized thanks to Alex McKenzie (ex-BBN). Many of the same people who worked on the Arpanet protocols also worked on the Internet Protocols and the leads formed the Internet Configuration Control Board (ICCB) before it was renamed the Internet Advisory Board by Barry Leiner and then Internet Activities Board and then, with the merge with ISOC, the Internet Architecture Board. The IETF and IRTF grew out of working groups (ten of them more or less) of the IAB, the largest of which was IETF. v On Fri, Jan 24, 2025 at 9:19?AM Steve Crocker via Internet-history < internet-history at elists.isoc.org> wrote: > It had to be there because I said so :). See RFC 3. > > Steve > > Sent from my iPhone > > > On Jan 24, 2025, at 9:08?AM, Guy Almes via Internet-history < > internet-history at elists.isoc.org> wrote: > > > > ?Hi Noel, > > You were there long before I was, but I recall that, in the upper left > corner of RFCs and Internet Drafts, we were always to put down "Network > Working Group". > > Even though we were organized into an IETF with Working Groups, etc., > that old name had to be there for some reason. > > -- Guy > > > >> On 1/24/25 8:52 AM, Noel Chiappa via Internet-history wrote: > >> So, what was the correct full name for the early engineering group that > met > >> every couple of months to work on TCP/IP, before the IETF was set up? > (I have > >> looked for it in Jon Postel's excellent minutes of its meetings, > preserved in > >> the Internet Experiment Note series, but I havn't found it there.) > >> I have this memory that we called it the 'Internet Working Group', but > of > >> course (especially in shortened form) that could be confused with the > INWG > >> (the 'International Network Working Group'), a different early group > entirly. > >> Noel > >> -- > >> Internet-history mailing list > >> Internet-history at elists.isoc.org > >> https://urldefense.com/v3/__https://elists.isoc.org/mailman/listinfo/ > internet-history__;!!KwNVnqRv! > CmzgJn96jrPDxtRTnatHHTCBN2TAXU8BefHlcjZpgPJuRVLhYsdmbUEVlQ7NPmOnBeigt- > Och0cm2u9EORW6deYdkij7FA$ < > https://urldefense.com/v3/__https://elists.isoc.org/mailman/listinfo/internet-history__;!!KwNVnqRv!CmzgJn96jrPDxtRTnatHHTCBN2TAXU8BefHlcjZpgPJuRVLhYsdmbUEVlQ7NPmOnBeigt-Och0cm2u9EORW6deYdkij7FA$ > > > > > > -- > > Internet-history mailing list > > Internet-history at elists.isoc.org > > https://elists.isoc.org/mailman/listinfo/internet-history > -- > Internet-history mailing list > Internet-history at elists.isoc.org > https://elists.isoc.org/mailman/listinfo/internet-history > -- Please send any postal/overnight deliveries to: Vint Cerf Google, LLC 1900 Reston Metro Plaza, 16th Floor Reston, VA 20190 +1 (571) 213 1346 until further notice From dhc at dcrocker.net Fri Jan 24 07:07:30 2025 From: dhc at dcrocker.net (Dave Crocker) Date: Fri, 24 Jan 2025 15:07:30 +0000 (UTC) Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: References: Message-ID: On 1/24/2025 6:28 AM, Vint Cerf via Internet-history wrote: > notes called INWG Notes As an outsider to the effort, who was vaguely 'around', when I read Noel's note and before reading Vint's, INWG was the name that I thought of.? Count this as an informal survey result of community perception... d/ -- Dave Crocker Brandenburg InternetWorking bbiw.net bluesky: @dcrocker.bsky.social mast: @dcrocker at mastodon.social From pugs78 at gmail.com Fri Jan 24 07:28:25 2025 From: pugs78 at gmail.com (Tom Lyon) Date: Fri, 24 Jan 2025 07:28:25 -0800 Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: References: Message-ID: Here's some physical evidence in support of INWG: https://mastodon.social/@aka_pugs/111093333460402486 On Fri, Jan 24, 2025 at 7:07?AM Dave Crocker via Internet-history < internet-history at elists.isoc.org> wrote: > On 1/24/2025 6:28 AM, Vint Cerf via Internet-history wrote: > > notes called INWG Notes > > > As an outsider to the effort, who was vaguely 'around', when I read > Noel's note and before reading Vint's, INWG was the name that I thought > of. Count this as an informal survey result of community perception... > > d/ > > -- > Dave Crocker > > Brandenburg InternetWorking > bbiw.net > bluesky: @dcrocker.bsky.social > mast: @dcrocker at mastodon.social > -- > Internet-history mailing list > Internet-history at elists.isoc.org > https://elists.isoc.org/mailman/listinfo/internet-history > From j at shoch.com Fri Jan 24 12:57:04 2025 From: j at shoch.com (John Shoch) Date: Fri, 24 Jan 2025 12:57:04 -0800 Subject: [ih] Internet-history Digest, Vol 62, Issue 9 In-Reply-To: References: Message-ID: Vint, Glad you could clarify some of the history on ARPA NWG vs INWG -- Alex McKenzie did have a great collection of INWG documents (and he notes your support in getting them digitized!): https://alexmckenzie.weebly.com/inwg-documents.html That initial, informal INWG meeting took place at the ICCC in Oct. 1972, as the Arpanet was having its "coming out" party in Washington DC. You have heard me say that I always thought this was a seminal meeting -- drawing together people working on different networks, and kicking off an incredibly creative period of research on internetworking. [I was at PARC at the time but, sadly, was not in DC.....] Has anyone written up the whole story of that meeting? How it came to be? Who attended? The sub-groups that were formed (you led one, I think)? Did the participants, at the time, anticipate what would evolve? Who was there, and have they written about it? What documents exist? Or do we need to do an oral history? Cheers, John Shoch Message: 4 Date: Fri, 24 Jan 2025 09:28:58 -0500 From: Vint Cerf To: Steve Crocker Cc: Guy Almes , Noel Chiappa , internet-history at elists.isoc.org Subject: Re: [ih] Correct name for early TCP/IP working group? Message-ID: Content-Type: text/plain; charset="UTF-8" so there were a few groups. NWG was Steve's creation and was the nexus in which NCP, FTP, SMTP, TELNET and other Arpanet-based protocols were developed. In 1972, an International Network Working Group (INWG) was created and had its own set of notes called INWG Notes - these have all been digitized thanks to Alex McKenzie (ex-BBN). Many of the same people who worked on the Arpanet protocols also worked on the Internet Protocols and the leads formed the Internet Configuration Control Board (ICCB) before it was renamed the Internet Advisory Board by Barry Leiner and then Internet Activities Board and then, with the merge with ISOC, the Internet Architecture Board. The IETF and IRTF grew out of working groups (ten of them more or less) of the IAB, the largest of which was IETF. v On Fri, Jan 24, 2025 at 7:07?AM wrote: > Send Internet-history mailing list submissions to > internet-history at elists.isoc.org > > To subscribe or unsubscribe via the World Wide Web, visit > https://elists.isoc.org/mailman/listinfo/internet-history > or, via email, send a message with subject or body 'help' to > internet-history-request at elists.isoc.org > > You can reach the person managing the list at > internet-history-owner at elists.isoc.org > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of Internet-history digest..." > > > Today's Topics: > > 1. Correct name for early TCP/IP working group? (Noel Chiappa) > 2. Re: Correct name for early TCP/IP working group? (Guy Almes) > 3. Re: Correct name for early TCP/IP working group? (Steve Crocker) > 4. Re: Correct name for early TCP/IP working group? (Vint Cerf) > 5. Re: Correct name for early TCP/IP working group? (Dave Crocker) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Fri, 24 Jan 2025 08:52:19 -0500 (EST) > From: jnc at mercury.lcs.mit.edu (Noel Chiappa) > To: internet-history at elists.isoc.org > Cc: jnc at mercury.lcs.mit.edu > Subject: [ih] Correct name for early TCP/IP working group? > Message-ID: <20250124135219.CA54118C0D4 at mercury.lcs.mit.edu> > > > So, what was the correct full name for the early engineering group that met > every couple of months to work on TCP/IP, before the IETF was set up? (I > have > looked for it in Jon Postel's excellent minutes of its meetings, preserved > in > the Internet Experiment Note series, but I havn't found it there.) > > I have this memory that we called it the 'Internet Working Group', but of > course (especially in shortened form) that could be confused with the INWG > (the 'International Network Working Group'), a different early group > entirly. > > Noel > > > ------------------------------ > > Message: 2 > Date: Fri, 24 Jan 2025 09:08:14 -0500 > From: Guy Almes > To: Noel Chiappa , > internet-history at elists.isoc.org > Subject: Re: [ih] Correct name for early TCP/IP working group? > Message-ID: > Content-Type: text/plain; charset=UTF-8; format=flowed > > Hi Noel, > You were there long before I was, but I recall that, in the upper > left corner of RFCs and Internet Drafts, we were always to put down > "Network Working Group". > Even though we were organized into an IETF with Working Groups, etc., > that old name had to be there for some reason. > -- Guy > > On 1/24/25 8:52 AM, Noel Chiappa via Internet-history wrote: > > So, what was the correct full name for the early engineering group that > met > > every couple of months to work on TCP/IP, before the IETF was set up? (I > have > > looked for it in Jon Postel's excellent minutes of its meetings, > preserved in > > the Internet Experiment Note series, but I havn't found it there.) > > > > I have this memory that we called it the 'Internet Working Group', but of > > course (especially in shortened form) that could be confused with the > INWG > > (the 'International Network Working Group'), a different early group > entirly. > > > > Noel > > -- > > Internet-history mailing list > > Internet-history at elists.isoc.org > > https://urldefense.com/v3/__https://elists.isoc.org/mailman/listinfo/ > > internet-history__;!!KwNVnqRv! > > CmzgJn96jrPDxtRTnatHHTCBN2TAXU8BefHlcjZpgPJuRVLhYsdmbUEVlQ7NPmOnBeigt- > > Och0cm2u9EORW6deYdkij7FA$ < > https://urldefense.com/v3/__https://elists.isoc.org/mailman/listinfo/internet-history__;!!KwNVnqRv!CmzgJn96jrPDxtRTnatHHTCBN2TAXU8BefHlcjZpgPJuRVLhYsdmbUEVlQ7NPmOnBeigt-Och0cm2u9EORW6deYdkij7FA$ > > > > > > > > ------------------------------ > > Message: 3 > Date: Fri, 24 Jan 2025 09:19:27 -0500 > From: Steve Crocker > To: Guy Almes > Cc: Noel Chiappa , > internet-history at elists.isoc.org > Subject: Re: [ih] Correct name for early TCP/IP working group? > Message-ID: > Content-Type: text/plain; charset=utf-8 > > It had to be there because I said so :). See RFC 3. > > Steve > > Sent from my iPhone > > > On Jan 24, 2025, at 9:08?AM, Guy Almes via Internet-history < > internet-history at elists.isoc.org> wrote: > > > > ?Hi Noel, > > You were there long before I was, but I recall that, in the upper left > corner of RFCs and Internet Drafts, we were always to put down "Network > Working Group". > > Even though we were organized into an IETF with Working Groups, etc., > that old name had to be there for some reason. > > -- Guy > > > >> On 1/24/25 8:52 AM, Noel Chiappa via Internet-history wrote: > >> So, what was the correct full name for the early engineering group that > met > >> every couple of months to work on TCP/IP, before the IETF was set up? > (I have > >> looked for it in Jon Postel's excellent minutes of its meetings, > preserved in > >> the Internet Experiment Note series, but I havn't found it there.) > >> I have this memory that we called it the 'Internet Working Group', but > of > >> course (especially in shortened form) that could be confused with the > INWG > >> (the 'International Network Working Group'), a different early group > entirly. > >> Noel > >> -- > >> Internet-history mailing list > >> Internet-history at elists.isoc.org > >> https://urldefense.com/v3/__https://elists.isoc.org/mailman/listinfo/ > internet-history__;!!KwNVnqRv! > CmzgJn96jrPDxtRTnatHHTCBN2TAXU8BefHlcjZpgPJuRVLhYsdmbUEVlQ7NPmOnBeigt- > Och0cm2u9EORW6deYdkij7FA$ < > https://urldefense.com/v3/__https://elists.isoc.org/mailman/listinfo/internet-history__;!!KwNVnqRv!CmzgJn96jrPDxtRTnatHHTCBN2TAXU8BefHlcjZpgPJuRVLhYsdmbUEVlQ7NPmOnBeigt-Och0cm2u9EORW6deYdkij7FA$ > > > > > > -- > > Internet-history mailing list > > Internet-history at elists.isoc.org > > https://elists.isoc.org/mailman/listinfo/internet-history > > > ------------------------------ > > Message: 4 > Date: Fri, 24 Jan 2025 09:28:58 -0500 > From: Vint Cerf > To: Steve Crocker > Cc: Guy Almes , Noel Chiappa > , internet-history at elists.isoc.org > Subject: Re: [ih] Correct name for early TCP/IP working group? > Message-ID: > < > CAHxHggeMijQarW-fnLzAodG2eibJ5v6oCOT-9yY+kkMMSBMkMQ at mail.gmail.com> > Content-Type: text/plain; charset="UTF-8" > > so there were a few groups. NWG was Steve's creation and was the nexus in > which NCP, FTP, SMTP, TELNET and other Arpanet-based protocols were > developed. In 1972, an International Network Working Group (INWG) was > created and had its own set of notes called INWG Notes - these have all > been digitized thanks to Alex McKenzie (ex-BBN). Many of the same people > who worked on the Arpanet protocols also worked on the Internet Protocols > and the leads formed the Internet Configuration Control Board (ICCB) before > it was renamed the Internet Advisory Board by Barry Leiner and then > Internet Activities Board and then, with the merge with ISOC, the Internet > Architecture Board. The IETF and IRTF grew out of working groups (ten of > them more or less) of the IAB, the largest of which was IETF. > > v > > > On Fri, Jan 24, 2025 at 9:19?AM Steve Crocker via Internet-history < > internet-history at elists.isoc.org> wrote: > > > It had to be there because I said so :). See RFC 3. > > > > Steve > > > > Sent from my iPhone > > > > > On Jan 24, 2025, at 9:08?AM, Guy Almes via Internet-history < > > internet-history at elists.isoc.org> wrote: > > > > > > ?Hi Noel, > > > You were there long before I was, but I recall that, in the upper left > > corner of RFCs and Internet Drafts, we were always to put down "Network > > Working Group". > > > Even though we were organized into an IETF with Working Groups, etc., > > that old name had to be there for some reason. > > > -- Guy > > > > > >> On 1/24/25 8:52 AM, Noel Chiappa via Internet-history wrote: > > >> So, what was the correct full name for the early engineering group > that > > met > > >> every couple of months to work on TCP/IP, before the IETF was set up? > > (I have > > >> looked for it in Jon Postel's excellent minutes of its meetings, > > preserved in > > >> the Internet Experiment Note series, but I havn't found it there.) > > >> I have this memory that we called it the 'Internet Working Group', but > > of > > >> course (especially in shortened form) that could be confused with the > > INWG > > >> (the 'International Network Working Group'), a different early group > > entirly. > > >> Noel > > >> -- > > >> Internet-history mailing list > > >> Internet-history at elists.isoc.org > > >> https://urldefense.com/v3/__https://elists.isoc.org/mailman/listinfo/ > > internet-history__;!!KwNVnqRv! > > CmzgJn96jrPDxtRTnatHHTCBN2TAXU8BefHlcjZpgPJuRVLhYsdmbUEVlQ7NPmOnBeigt- > > Och0cm2u9EORW6deYdkij7FA$ < > > > https://urldefense.com/v3/__https://elists.isoc.org/mailman/listinfo/internet-history__;!!KwNVnqRv!CmzgJn96jrPDxtRTnatHHTCBN2TAXU8BefHlcjZpgPJuRVLhYsdmbUEVlQ7NPmOnBeigt-Och0cm2u9EORW6deYdkij7FA$ > > > > > > > > > -- > > > Internet-history mailing list > > > Internet-history at elists.isoc.org > > > https://elists.isoc.org/mailman/listinfo/internet-history > > -- > > Internet-history mailing list > > Internet-history at elists.isoc.org > > https://elists.isoc.org/mailman/listinfo/internet-history > > > > > -- > Please send any postal/overnight deliveries to: > Vint Cerf > Google, LLC > 1900 Reston Metro Plaza, 16th Floor > Reston, VA 20190 > +1 (571) 213 1346 > > > until further notice > > > ------------------------------ > > Message: 5 > Date: Fri, 24 Jan 2025 15:07:30 +0000 (UTC) > From: Dave Crocker > To: Vint Cerf , Steve Crocker > Cc: Guy Almes , Noel Chiappa > , internet-history at elists.isoc.org > Subject: Re: [ih] Correct name for early TCP/IP working group? > Message-ID: > Content-Type: text/plain; charset=UTF-8; format=flowed > > On 1/24/2025 6:28 AM, Vint Cerf via Internet-history wrote: > > notes called INWG Notes > > > As an outsider to the effort, who was vaguely 'around', when I read > Noel's note and before reading Vint's, INWG was the name that I thought > of.? Count this as an informal survey result of community perception... > > d/ > > -- > Dave Crocker > > Brandenburg InternetWorking > bbiw.net > bluesky: @dcrocker.bsky.social > mast: @dcrocker at mastodon.social > > > ------------------------------ > > Subject: Digest Footer > > Internet-history mailing list > Internet-history at elists.isoc.org > https://elists.isoc.org/mailman/listinfo/internet-history > > > ------------------------------ > > End of Internet-history Digest, Vol 62, Issue 9 > *********************************************** > From jack at 3kitty.org Fri Jan 24 14:15:37 2025 From: jack at 3kitty.org (Jack Haverty) Date: Fri, 24 Jan 2025 14:15:37 -0800 Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: References: Message-ID: <431e9222-d1f4-4245-88ca-d286ae7f44a4@3kitty.org> My recollections, all IIRC of course, after almost 50 years (not necessarily in chronological order): - The "Internet Project" was, circa 1977, actually an informal collection of separate ARPA projects.? Packet Radio (PRNET) was building wireless networks with mobility to be used in jeeps, helicopters, etc.?? SATNET was building transatlantic networking by satellite.? WBNET (WideBand NETwork) was building a "high bandwidth" (3 Mb/sec) satellite network spanning the continental US.? The ARPANET had been operating for almost a decade and was now run by DCA instead of ARPA. The "Gateway Project" was building a device to attach to a PRNET and pass TCP traffic across the boundaries as an initial experiment in the new Cerf/Kahn concept of TCP. - Each of these groups had its own set of contractors working on that project, using what was available at the time to colaborate and communicate, e.g., meetings and email across the ARPANET. - At the time, the "ARPANET crowd" was skeptical that the "datagram" nature of TCP could be made to work.?? Traditional networks, including the ARPANET, had elaborate internal mechanisms to provide a "virtual circuit" service to its users.? Although the ARPANET had a simple "datagram mode" (aka messages of "subtype 3"?), there was strong reluctance to permit its use other than for very limited experiments, for fear such use would crash the ARPANET. - There was some work however within the ARPANET IMP software to acknowledge the need for multiple networks.?? For example, some of the formats of data as it passed through the ARPANET included fields labelled "Network Number".? AFAIK this was never actually fully implemented so the ARPANET itself never achieved connectivity between multiple networks until TCP was deployed. - My recollection is that I had heard of INWG, but never knew much about their work (still don't).?? IIRC, none of the people involved in implementation (building, coding, operating) of the various "Internet projects" were also involved in INWG.? I had the impression that the INWG was part of the group that thought the datagram architecture was unworkable.?? Mentally, I associated it with X.25 and X.75 style of interconnecting networks.?? But perhaps that was a mistake. - As the various ARPA projects matured and were interconnected with Gateways, people involved in each project saw a need to interact with people involved in other projects.? Meetings of the "Internet Project" began, but the scope was very broad, so everyone wanted to attend all such meetings.? At one point ARPA restricted attendance to 2 people per contractor as a way to manage the meeting size and cost. - When TCPV2 was evolving into TCP/IPV4, two separate groups were formed: the "TCP Working Group" and the "Internet Working Group". These met and interacted separately. - After a few months, we observed that the main results of the TCP WG were changes to the IP Header, and the main results of the IP WG were changes to the TCP Header.?? The decision was made to fold the groups back together as the "Internet Working Group".?? That decision was of course Vint's, but by then I think the ICCB had been formed and had a role. - At about the time that Vint was moving from ARPA to MCI, the technical work was reorganized into two components.? I recall it happened at the final ICCB meeting, sometime in early 1983.? The IETF was formed to Engineer the operational Internet as it grew. The IRTF was formed to pursue the Research into all of the unsolved research issues - things like congestion, multi-path routing, etc. - At some point in this progression, Jon started collected IENs rather than RFCs.?? IENs were part of the Internet Experiment, while RFCs were more associated with ARPANET issues. - The size of "Internet Meetings" continued to grow, despite efforts to constrain attendance.? It was difficult to find a location with large enough rooms to accommodate the next meeting.? The ultimate Internet Wheeler Dealer (Dan Lynch) noticed this mismatch of supply and demand.? He organized some early meeting venues, e.g., one in a hotel in Monterey California, which became known as "Geeks On The Bay In Monterey".? That led quickly to the Interop shows.?? Problem solved. Jack Haverty On 1/24/25 07:28, Tom Lyon via Internet-history wrote: > Here's some physical evidence in support of INWG: > https://mastodon.social/@aka_pugs/111093333460402486 > > On Fri, Jan 24, 2025 at 7:07?AM Dave Crocker via Internet-history < > internet-history at elists.isoc.org> wrote: > >> On 1/24/2025 6:28 AM, Vint Cerf via Internet-history wrote: >>> notes called INWG Notes >> >> As an outsider to the effort, who was vaguely 'around', when I read >> Noel's note and before reading Vint's, INWG was the name that I thought >> of. Count this as an informal survey result of community perception... >> >> d/ >> >> -- >> Dave Crocker >> >> Brandenburg InternetWorking >> bbiw.net >> bluesky: @dcrocker.bsky.social >> mast: @dcrocker at mastodon.social >> -- >> Internet-history mailing list >> Internet-history at elists.isoc.org >> https://elists.isoc.org/mailman/listinfo/internet-history >> -------------- next part -------------- A non-text attachment was scrubbed... Name: OpenPGP_signature.asc Type: application/pgp-signature Size: 665 bytes Desc: OpenPGP digital signature URL: From vgcerf at gmail.com Fri Jan 24 14:57:29 2025 From: vgcerf at gmail.com (vinton cerf) Date: Fri, 24 Jan 2025 17:57:29 -0500 Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: <431e9222-d1f4-4245-88ca-d286ae7f44a4@3kitty.org> References: <431e9222-d1f4-4245-88ca-d286ae7f44a4@3kitty.org> Message-ID: On Fri, Jan 24, 2025 at 5:15?PM Jack Haverty via Internet-history < internet-history at elists.isoc.org> wrote: > My recollections, all IIRC of course, after almost 50 years (not > necessarily in chronological order): > > - The "Internet Project" was, circa 1977, actually an informal > collection of separate ARPA projects. Packet Radio (PRNET) was building > wireless networks with mobility to be used in jeeps, helicopters, etc. > SATNET was building transatlantic networking by satellite. WBNET > (WideBand NETwork) was building a "high bandwidth" (3 Mb/sec) satellite > network spanning the continental US. The ARPANET had been operating for > almost a decade and was now run by DCA instead of ARPA. The "Gateway > Project" was building a device to attach to a PRNET and pass TCP traffic > across the boundaries as an initial experiment in the new Cerf/Kahn > concept of TCP. > Jack, the Internet project started in 1973, about four years after the first IMPs were delivered. The Packet Radio and Packet Satellite networks were working around 1976 > > - Each of these groups had its own set of contractors working on that > project, using what was available at the time to colaborate and > communicate, e.g., meetings and email across the ARPANET. > > - At the time, the "ARPANET crowd" was skeptical that the "datagram" > nature of TCP could be made to work. Traditional networks, including > the ARPANET, had elaborate internal mechanisms to provide a "virtual > circuit" service to its users. Although the ARPANET had a simple > "datagram mode" (aka messages of "subtype 3"?), there was strong > reluctance to permit its use other than for very limited experiments, > for fear such use would crash the ARPANET. > > - There was some work however within the ARPANET IMP software to > acknowledge the need for multiple networks. For example, some of the > formats of data as it passed through the ARPANET included fields > labelled "Network Number". AFAIK this was never actually fully > implemented so the ARPANET itself never achieved connectivity between > multiple networks until TCP was deployed. > > - My recollection is that I had heard of INWG, but never knew much about > their work (still don't). IIRC, none of the people involved in > implementation (building, coding, operating) of the various "Internet > projects" were also involved in INWG. I don't think that is correct. INWG had a blend of people from the US and Europe and Japan (I had some of these people at Stanford). > I had the impression that the > INWG was part of the group that thought the datagram architecture was > unworkable. Mentally, I associated it with X.25 and X.75 style of > interconnecting networks. But perhaps that was a mistake. > yes - there were factions favoring virtual circuits but others who were persuaded that datagram operation was preferable. > > - As the various ARPA projects matured and were interconnected with > Gateways, people involved in each project saw a need to interact with > people involved in other projects. Meetings of the "Internet Project" > began, but the scope was very broad, so everyone wanted to attend all > such meetings. At one point ARPA restricted attendance to 2 people per > contractor as a way to manage the meeting size and cost. > > - When TCPV2 was evolving into TCP/IPV4, two separate groups were > formed: the "TCP Working Group" and the "Internet Working Group". These > met and interacted separately. > This must have been after I got to ARPA because I don't recall such a grouping. > > - After a few months, we observed that the main results of the TCP WG > were changes to the IP Header, and the main results of the IP WG were > changes to the TCP Header. The decision was made to fold the groups > back together as the "Internet Working Group". That decision was of > course Vint's, but by then I think the ICCB had been formed and had a role. > ICCB was formed in 1979 > > - At about the time that Vint was moving from ARPA to MCI, the technical > work was reorganized into two components. I recall it happened at the > final ICCB meeting, sometime in early 1983. The IETF was formed to > Engineer the operational Internet as it grew. The IRTF was formed to > pursue the Research into all of the unsolved research issues - things > like congestion, multi-path routing, etc. > I think you skipped a step or three here. Barry Leiner took over the Internet, Packet Radio and Packet Satellite programs around 1983 (late?). He morphed the ICCB into the Internet Activities Board > > - At some point in this progression, Jon started collected IENs rather > than RFCs. IENs were part of the Internet Experiment, while RFCs were > more associated with ARPANET issues. > IENs and INWG Notes were more or less concurrent. The latter included a lot of European participants and the INWG eventually became IFIP 6.1 > > - The size of "Internet Meetings" continued to grow, despite efforts to > constrain attendance. It was difficult to find a location with large > enough rooms to accommodate the next meeting. The ultimate Internet > Wheeler Dealer (Dan Lynch) noticed this mismatch of supply and demand. > He organized some early meeting venues, e.g., one in a hotel in Monterey > California, which became known as "Geeks On The Bay In Monterey". That > led quickly to the Interop shows. Problem solved. That would have been > about 1986 which is also the time that the IETF was formed, along with the > IRTF by merging a number of IAB working groups into these two categories. > > Jack Haverty > > > > On 1/24/25 07:28, Tom Lyon via Internet-history wrote: > > Here's some physical evidence in support of INWG: > > https://mastodon.social/@aka_pugs/111093333460402486 > > > > On Fri, Jan 24, 2025 at 7:07?AM Dave Crocker via Internet-history < > > internet-history at elists.isoc.org> wrote: > > > >> On 1/24/2025 6:28 AM, Vint Cerf via Internet-history wrote: > >>> notes called INWG Notes > >> > >> As an outsider to the effort, who was vaguely 'around', when I read > >> Noel's note and before reading Vint's, INWG was the name that I thought > >> of. Count this as an informal survey result of community perception... > >> > >> d/ > >> > >> -- > >> Dave Crocker > >> > >> Brandenburg InternetWorking > >> bbiw.net > >> bluesky: @dcrocker.bsky.social > >> mast: @dcrocker at mastodon.social > >> -- > >> Internet-history mailing list > >> Internet-history at elists.isoc.org > >> https://elists.isoc.org/mailman/listinfo/internet-history > >> > > -- > Internet-history mailing list > Internet-history at elists.isoc.org > https://elists.isoc.org/mailman/listinfo/internet-history > From jnc at mercury.lcs.mit.edu Fri Jan 24 16:06:56 2025 From: jnc at mercury.lcs.mit.edu (Noel Chiappa) Date: Fri, 24 Jan 2025 19:06:56 -0500 (EST) Subject: [ih] Correct name for early TCP/IP working group? Message-ID: <20250125000656.0514D18C083@mercury.lcs.mit.edu> > From: Vint Cerf > so there were a few groups. ... the leads formed the Internet > Configuration Control Board (ICCB) Right, but the ICCB was formed in about 1979. I have created a list of the early meetings (starting in 1977), here: https://gunkies.org/wiki/TCP_and_Internet_Meetings and there were about a dozen over two years before that - most of the ones in that list were of the body I'm asking about. I'm not sure it was a formal group; it was basically just your DARPA contractors on the project. So maybe it didn't have a formal name. (If so, maybe we should make one up! :-) Noel From jack at 3kitty.org Fri Jan 24 16:54:33 2025 From: jack at 3kitty.org (Jack Haverty) Date: Fri, 24 Jan 2025 16:54:33 -0800 Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: References: <431e9222-d1f4-4245-88ca-d286ae7f44a4@3kitty.org> Message-ID: <3c6d581a-927c-40b6-9f70-54cb3e727b0e@3kitty.org> On 1/24/25 14:57, vinton cerf wrote: > > - When TCPV2 was evolving into TCP/IPV4, two separate groups were > formed: the "TCP Working Group" and the "Internet Working Group". > These > met and interacted separately. > > This must have been after I got to ARPA because I don't recall such a > grouping. There were separate meetings for TCP and Internet during around 1978-79.?? IIRC that's when there were two separate (but overlapping) sets of people.? One was supposed to refine TCP, and the other to finalize IP.?? Those two groups may not have had separate names, but the notes from the meetings had different titles.?? Noel's website captures that some meetings were "TCP Meetings" and others were "Internet Meetings".?? After 1979 I think they were all Internet Meetings.?? Of course there was much more constant interaction by email. I migrated from Lick's group at MIT to Heart's group at BBN somewhere around Fall 1977 and my first assignment was to get TCP implemented on PDP-11 Unix.? That was interesting because I had never before worked on or even used Unix, hadn't implemented any networking software below the level of mail servers, had never heard of TCP, and had never seen a PDP-11.?? So I was perfect for the task....? Jack From Noel's website: EN's (in IEN number order): * INTERNET Meeting Notes 15 August 1977 (IEN 3) * Meeting Notes - 1 February 1978 (IEN-22) * Meeting Notes - 2,3&4 August 1978 (IEN-53) * Boston Area Meeting of the Internet Working Group to Discuss Interactions With Gateways (IEN-60) * Internet Meeting Notes - 30 & 31 October 1978 (IEN-63) * Minutes of TCP Meeting, March 12, 1977, Washington, D.C. (IEN-64) * TCP Meeting Notes - 14 & 15 July 1977 (IEN-65) * TCP Meeting Notes - 13 & 14 October 1977 (IEN-66 * TCP Meeting Notes 30 & 31 January 1978 (IEN-67) * TCP Meeting Notes - 15 & 16 June 1978 (IEN-68) * TCP Meeting Notes - 18 & 19 September 1978 (IEN 69) * Internet Meeting Notes - 4 December 1978 (IEN-70) * TCP Meeting Notes - 29 January 1979 (IEN-77) * Minutes of the Fault Isolation Meeting (IEN-104) * Internet Meeting Notes - 8, 9, 10 & 11 May 1979 (IEN-106) * Internet Meeting Notes - 10, 11, 12 & 13 September 1979 (IEN-121) * Internet Meeting Notes - 4, 5, & 6 February 1980 (IEN-134) * Internet Meeting Notes - 14 & 15 May 1980 (IEN-145) * Internet Meeting Notes -- 7-8-9 October 1980 (IEN-160) * Internet Meeting Notes -- 28-29-30 January 1981 (IEN-175) -------------- next part -------------- A non-text attachment was scrubbed... Name: OpenPGP_signature.asc Type: application/pgp-signature Size: 665 bytes Desc: OpenPGP digital signature URL: From vgcerf at gmail.com Fri Jan 24 18:05:34 2025 From: vgcerf at gmail.com (vinton cerf) Date: Fri, 24 Jan 2025 21:05:34 -0500 Subject: [ih] Internet-history Digest, Vol 62, Issue 9 In-Reply-To: References: Message-ID: that's a good question - I am not sure it was ever documented with minutes. It was pretty informal. Pouzin, Crocker, Cerf, Metcalfe (?) and many others were there because of ICCC and the demonstration of the Arpanet. v On Fri, Jan 24, 2025 at 3:57?PM John Shoch via Internet-history < internet-history at elists.isoc.org> wrote: > Vint, > > Glad you could clarify some of the history on ARPA NWG vs INWG -- Alex > McKenzie did have a great collection of INWG documents (and he notes your > support in getting them digitized!): > https://alexmckenzie.weebly.com/inwg-documents.html > > That initial, informal INWG meeting took place at the ICCC in Oct. 1972, as > the Arpanet was having its "coming out" party in Washington DC. > You have heard me say that I always thought this was a seminal meeting -- > drawing together people working on different networks, and kicking off an > incredibly creative period of research on internetworking. [I was at PARC > at the time but, sadly, was not in DC.....] > > Has anyone written up the whole story of that meeting? How it came to be? > Who attended? The sub-groups that were formed (you led one, I think)? Did > the participants, at the time, anticipate what would evolve? > > Who was there, and have they written about it? What documents exist? > Or do we need to do an oral history? > > Cheers, > > John Shoch > > > Message: 4 > Date: Fri, 24 Jan 2025 09:28:58 -0500 > From: Vint Cerf > To: Steve Crocker > Cc: Guy Almes , Noel Chiappa > , internet-history at elists.isoc.org > Subject: Re: [ih] Correct name for early TCP/IP working group? > Message-ID: > < > CAHxHggeMijQarW-fnLzAodG2eibJ5v6oCOT-9yY+kkMMSBMkMQ at mail.gmail.com> > Content-Type: text/plain; charset="UTF-8" > > so there were a few groups. NWG was Steve's creation and was the nexus in > which NCP, FTP, SMTP, TELNET and other Arpanet-based protocols were > developed. In 1972, an International Network Working Group (INWG) was > created and had its own set of notes called INWG Notes - these have all > been digitized thanks to Alex McKenzie (ex-BBN). Many of the same people > who worked on the Arpanet protocols also worked on the Internet Protocols > and the leads formed the Internet Configuration Control Board (ICCB) before > it was renamed the Internet Advisory Board by Barry Leiner and then > Internet Activities Board and then, with the merge with ISOC, the Internet > Architecture Board. The IETF and IRTF grew out of working groups (ten of > them more or less) of the IAB, the largest of which was IETF. > > v > > On Fri, Jan 24, 2025 at 7:07?AM > wrote: > > > Send Internet-history mailing list submissions to > > internet-history at elists.isoc.org > > > > To subscribe or unsubscribe via the World Wide Web, visit > > https://elists.isoc.org/mailman/listinfo/internet-history > > or, via email, send a message with subject or body 'help' to > > internet-history-request at elists.isoc.org > > > > You can reach the person managing the list at > > internet-history-owner at elists.isoc.org > > > > When replying, please edit your Subject line so it is more specific > > than "Re: Contents of Internet-history digest..." > > > > > > Today's Topics: > > > > 1. Correct name for early TCP/IP working group? (Noel Chiappa) > > 2. Re: Correct name for early TCP/IP working group? (Guy Almes) > > 3. Re: Correct name for early TCP/IP working group? (Steve Crocker) > > 4. Re: Correct name for early TCP/IP working group? (Vint Cerf) > > 5. Re: Correct name for early TCP/IP working group? (Dave Crocker) > > > > > > ---------------------------------------------------------------------- > > > > Message: 1 > > Date: Fri, 24 Jan 2025 08:52:19 -0500 (EST) > > From: jnc at mercury.lcs.mit.edu (Noel Chiappa) > > To: internet-history at elists.isoc.org > > Cc: jnc at mercury.lcs.mit.edu > > Subject: [ih] Correct name for early TCP/IP working group? > > Message-ID: <20250124135219.CA54118C0D4 at mercury.lcs.mit.edu> > > > > > > So, what was the correct full name for the early engineering group that > met > > every couple of months to work on TCP/IP, before the IETF was set up? (I > > have > > looked for it in Jon Postel's excellent minutes of its meetings, > preserved > > in > > the Internet Experiment Note series, but I havn't found it there.) > > > > I have this memory that we called it the 'Internet Working Group', but of > > course (especially in shortened form) that could be confused with the > INWG > > (the 'International Network Working Group'), a different early group > > entirly. > > > > Noel > > > > > > ------------------------------ > > > > Message: 2 > > Date: Fri, 24 Jan 2025 09:08:14 -0500 > > From: Guy Almes > > To: Noel Chiappa , > > internet-history at elists.isoc.org > > Subject: Re: [ih] Correct name for early TCP/IP working group? > > Message-ID: > > Content-Type: text/plain; charset=UTF-8; format=flowed > > > > Hi Noel, > > You were there long before I was, but I recall that, in the upper > > left corner of RFCs and Internet Drafts, we were always to put down > > "Network Working Group". > > Even though we were organized into an IETF with Working Groups, etc., > > that old name had to be there for some reason. > > -- Guy > > > > On 1/24/25 8:52 AM, Noel Chiappa via Internet-history wrote: > > > So, what was the correct full name for the early engineering group that > > met > > > every couple of months to work on TCP/IP, before the IETF was set up? > (I > > have > > > looked for it in Jon Postel's excellent minutes of its meetings, > > preserved in > > > the Internet Experiment Note series, but I havn't found it there.) > > > > > > I have this memory that we called it the 'Internet Working Group', but > of > > > course (especially in shortened form) that could be confused with the > > INWG > > > (the 'International Network Working Group'), a different early group > > entirly. > > > > > > Noel > > > -- > > > Internet-history mailing list > > > Internet-history at elists.isoc.org > > > https://urldefense.com/v3/__https://elists.isoc.org/mailman/listinfo/ > > > internet-history__;!!KwNVnqRv! > > > CmzgJn96jrPDxtRTnatHHTCBN2TAXU8BefHlcjZpgPJuRVLhYsdmbUEVlQ7NPmOnBeigt- > > > Och0cm2u9EORW6deYdkij7FA$ < > > > https://urldefense.com/v3/__https://elists.isoc.org/mailman/listinfo/internet-history__;!!KwNVnqRv!CmzgJn96jrPDxtRTnatHHTCBN2TAXU8BefHlcjZpgPJuRVLhYsdmbUEVlQ7NPmOnBeigt-Och0cm2u9EORW6deYdkij7FA$ > > > > > > > > > > > > > > ------------------------------ > > > > Message: 3 > > Date: Fri, 24 Jan 2025 09:19:27 -0500 > > From: Steve Crocker > > To: Guy Almes > > Cc: Noel Chiappa , > > internet-history at elists.isoc.org > > Subject: Re: [ih] Correct name for early TCP/IP working group? > > Message-ID: > > Content-Type: text/plain; charset=utf-8 > > > > It had to be there because I said so :). See RFC 3. > > > > Steve > > > > Sent from my iPhone > > > > > On Jan 24, 2025, at 9:08?AM, Guy Almes via Internet-history < > > internet-history at elists.isoc.org> wrote: > > > > > > ?Hi Noel, > > > You were there long before I was, but I recall that, in the upper left > > corner of RFCs and Internet Drafts, we were always to put down "Network > > Working Group". > > > Even though we were organized into an IETF with Working Groups, etc., > > that old name had to be there for some reason. > > > -- Guy > > > > > >> On 1/24/25 8:52 AM, Noel Chiappa via Internet-history wrote: > > >> So, what was the correct full name for the early engineering group > that > > met > > >> every couple of months to work on TCP/IP, before the IETF was set up? > > (I have > > >> looked for it in Jon Postel's excellent minutes of its meetings, > > preserved in > > >> the Internet Experiment Note series, but I havn't found it there.) > > >> I have this memory that we called it the 'Internet Working Group', but > > of > > >> course (especially in shortened form) that could be confused with the > > INWG > > >> (the 'International Network Working Group'), a different early group > > entirly. > > >> Noel > > >> -- > > >> Internet-history mailing list > > >> Internet-history at elists.isoc.org > > >> https://urldefense.com/v3/__https://elists.isoc.org/mailman/listinfo/ > > internet-history__;!!KwNVnqRv! > > CmzgJn96jrPDxtRTnatHHTCBN2TAXU8BefHlcjZpgPJuRVLhYsdmbUEVlQ7NPmOnBeigt- > > Och0cm2u9EORW6deYdkij7FA$ < > > > https://urldefense.com/v3/__https://elists.isoc.org/mailman/listinfo/internet-history__;!!KwNVnqRv!CmzgJn96jrPDxtRTnatHHTCBN2TAXU8BefHlcjZpgPJuRVLhYsdmbUEVlQ7NPmOnBeigt-Och0cm2u9EORW6deYdkij7FA$ > > > > > > > > > -- > > > Internet-history mailing list > > > Internet-history at elists.isoc.org > > > https://elists.isoc.org/mailman/listinfo/internet-history > > > > > > ------------------------------ > > > > Message: 4 > > Date: Fri, 24 Jan 2025 09:28:58 -0500 > > From: Vint Cerf > > To: Steve Crocker > > Cc: Guy Almes , Noel Chiappa > > , internet-history at elists.isoc.org > > Subject: Re: [ih] Correct name for early TCP/IP working group? > > Message-ID: > > < > > CAHxHggeMijQarW-fnLzAodG2eibJ5v6oCOT-9yY+kkMMSBMkMQ at mail.gmail.com> > > Content-Type: text/plain; charset="UTF-8" > > > > so there were a few groups. NWG was Steve's creation and was the nexus in > > which NCP, FTP, SMTP, TELNET and other Arpanet-based protocols were > > developed. In 1972, an International Network Working Group (INWG) was > > created and had its own set of notes called INWG Notes - these have all > > been digitized thanks to Alex McKenzie (ex-BBN). Many of the same people > > who worked on the Arpanet protocols also worked on the Internet Protocols > > and the leads formed the Internet Configuration Control Board (ICCB) > before > > it was renamed the Internet Advisory Board by Barry Leiner and then > > Internet Activities Board and then, with the merge with ISOC, the > Internet > > Architecture Board. The IETF and IRTF grew out of working groups (ten of > > them more or less) of the IAB, the largest of which was IETF. > > > > v > > > > > > On Fri, Jan 24, 2025 at 9:19?AM Steve Crocker via Internet-history < > > internet-history at elists.isoc.org> wrote: > > > > > It had to be there because I said so :). See RFC 3. > > > > > > Steve > > > > > > Sent from my iPhone > > > > > > > On Jan 24, 2025, at 9:08?AM, Guy Almes via Internet-history < > > > internet-history at elists.isoc.org> wrote: > > > > > > > > ?Hi Noel, > > > > You were there long before I was, but I recall that, in the upper > left > > > corner of RFCs and Internet Drafts, we were always to put down "Network > > > Working Group". > > > > Even though we were organized into an IETF with Working Groups, > etc., > > > that old name had to be there for some reason. > > > > -- Guy > > > > > > > >> On 1/24/25 8:52 AM, Noel Chiappa via Internet-history wrote: > > > >> So, what was the correct full name for the early engineering group > > that > > > met > > > >> every couple of months to work on TCP/IP, before the IETF was set > up? > > > (I have > > > >> looked for it in Jon Postel's excellent minutes of its meetings, > > > preserved in > > > >> the Internet Experiment Note series, but I havn't found it there.) > > > >> I have this memory that we called it the 'Internet Working Group', > but > > > of > > > >> course (especially in shortened form) that could be confused with > the > > > INWG > > > >> (the 'International Network Working Group'), a different early group > > > entirly. > > > >> Noel > > > >> -- > > > >> Internet-history mailing list > > > >> Internet-history at elists.isoc.org > > > >> > https://urldefense.com/v3/__https://elists.isoc.org/mailman/listinfo/ > > > internet-history__;!!KwNVnqRv! > > > CmzgJn96jrPDxtRTnatHHTCBN2TAXU8BefHlcjZpgPJuRVLhYsdmbUEVlQ7NPmOnBeigt- > > > Och0cm2u9EORW6deYdkij7FA$ < > > > > > > https://urldefense.com/v3/__https://elists.isoc.org/mailman/listinfo/internet-history__;!!KwNVnqRv!CmzgJn96jrPDxtRTnatHHTCBN2TAXU8BefHlcjZpgPJuRVLhYsdmbUEVlQ7NPmOnBeigt-Och0cm2u9EORW6deYdkij7FA$ > > > > > > > > > > > > -- > > > > Internet-history mailing list > > > > Internet-history at elists.isoc.org > > > > https://elists.isoc.org/mailman/listinfo/internet-history > > > -- > > > Internet-history mailing list > > > Internet-history at elists.isoc.org > > > https://elists.isoc.org/mailman/listinfo/internet-history > > > > > > > > > -- > > Please send any postal/overnight deliveries to: > > Vint Cerf > > Google, LLC > > 1900 Reston Metro Plaza, 16th Floor > > Reston, VA 20190 > > +1 (571) 213 1346 > > > > > > until further notice > > > > > > ------------------------------ > > > > Message: 5 > > Date: Fri, 24 Jan 2025 15:07:30 +0000 (UTC) > > From: Dave Crocker > > To: Vint Cerf , Steve Crocker > > Cc: Guy Almes , Noel Chiappa > > , internet-history at elists.isoc.org > > Subject: Re: [ih] Correct name for early TCP/IP working group? > > Message-ID: > > Content-Type: text/plain; charset=UTF-8; format=flowed > > > > On 1/24/2025 6:28 AM, Vint Cerf via Internet-history wrote: > > > notes called INWG Notes > > > > > > As an outsider to the effort, who was vaguely 'around', when I read > > Noel's note and before reading Vint's, INWG was the name that I thought > > of.? Count this as an informal survey result of community perception... > > > > d/ > > > > -- > > Dave Crocker > > > > Brandenburg InternetWorking > > bbiw.net > > bluesky: @dcrocker.bsky.social > > mast: @dcrocker at mastodon.social > > > > > > ------------------------------ > > > > Subject: Digest Footer > > > > Internet-history mailing list > > Internet-history at elists.isoc.org > > https://elists.isoc.org/mailman/listinfo/internet-history > > > > > > ------------------------------ > > > > End of Internet-history Digest, Vol 62, Issue 9 > > *********************************************** > > > -- > Internet-history mailing list > Internet-history at elists.isoc.org > https://elists.isoc.org/mailman/listinfo/internet-history > From dhc at dcrocker.net Fri Jan 24 18:10:03 2025 From: dhc at dcrocker.net (Dave Crocker) Date: Sat, 25 Jan 2025 02:10:03 +0000 (UTC) Subject: [ih] Internet-history Digest, Vol 62, Issue 9 In-Reply-To: References: Message-ID: <4a8b1e37-99f1-4690-a622-6168c7b22796@dcrocker.net> On 1/24/2025 6:05 PM, vinton cerf via Internet-history wrote: > that's a good question - I am not sure it was ever documented with minutes. > It was pretty informal. Pouzin, Crocker, Cerf, Metcalfe (?) and many others > were there because of ICCC and the demonstration of the Arpanet. Whoever was there an is available should record their recollections from then.? Don't worry about the burden of writing.? Get the snapshots. d/ -- Dave Crocker Brandenburg InternetWorking bbiw.net bluesky: @dcrocker.bsky.social mast: @dcrocker at mastodon.social From aam3sendonly at gmail.com Fri Jan 24 18:49:30 2025 From: aam3sendonly at gmail.com (Alexander McKenzie) Date: Fri, 24 Jan 2025 21:49:30 -0500 Subject: [ih] Fwd: Fw: Internet-history Digest, Vol 62, Issue 9 In-Reply-To: <701149288.3600971.1737773233989@mail.yahoo.com> References: <964595491.35548.1737764588011@mail.yahoo.com> <701149288.3600971.1737773233989@mail.yahoo.com> Message-ID: For my view of what happened at the 1972 ICCC and the subsequent work of INWG see: http://alexmckenzie.weebly.com/inwg-and-the-conception-of-the-internet-an-eyewitness-account.html rnet.html Cheers, Alex On Friday, January 24, 2025 at 03:57:30 PM EST, John Shoch via Internet-history wrote: Vint, Glad you could clarify some of the history on ARPA NWG vs INWG -- Alex McKenzie did have a great collection of INWG documents (and he notes your support in getting them digitized!): https://alexmckenzie.weebly.com/inwg-documents.html That initial, informal INWG meeting took place at the ICCC in Oct. 1972, as the Arpanet was having its "coming out" party in Washington DC. You have heard me say that I always thought this was a seminal meeting -- drawing together people working on different networks, and kicking off an incredibly creative period of research on internetworking. [I was at PARC at the time but, sadly, was not in DC.....] Has anyone written up the whole story of that meeting? How it came to be? Who attended? The sub-groups that were formed (you led one, I think)? Did the participants, at the time, anticipate what would evolve? Who was there, and have they written about it? What documents exist? Or do we need to do an oral history? Cheers, John Shoch Message: 4 Date: Fri, 24 Jan 2025 09:28:58 -0500 From: Vint Cerf To: Steve Crocker Cc: Guy Almes , Noel Chiappa , internet-history at elists.isoc.org Subject: Re: [ih] Correct name for early TCP/IP working group? Message-ID: Content-Type: text/plain; charset="UTF-8" so there were a few groups. NWG was Steve's creation and was the nexus in which NCP, FTP, SMTP, TELNET and other Arpanet-based protocols were developed. In 1972, an International Network Working Group (INWG) was created and had its own set of notes called INWG Notes - these have all been digitized thanks to Alex McKenzie (ex-BBN). Many of the same people who worked on the Arpanet protocols also worked on the Internet Protocols and the leads formed the Internet Configuration Control Board (ICCB) before it was renamed the Internet Advisory Board by Barry Leiner and then Internet Activities Board and then, with the merge with ISOC, the Internet Architecture Board. The IETF and IRTF grew out of working groups (ten of them more or less) of the IAB, the largest of which was IETF. v On Fri, Jan 24, 2025 at 7:07?AM wrote: > Send Internet-history mailing list submissions to > internet-history at elists.isoc.org > > To subscribe or unsubscribe via the World Wide Web, visit > https://elists.isoc.org/mailman/listinfo/internet-history > or, via email, send a message with subject or body 'help' to > internet-history-request at elists.isoc.org > > You can reach the person managing the list at > internet-history-owner at elists.isoc.org > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of Internet-history digest..." > > > Today's Topics: > > 1. Correct name for early TCP/IP working group? (Noel Chiappa) > 2. Re: Correct name for early TCP/IP working group? (Guy Almes) > 3. Re: Correct name for early TCP/IP working group? (Steve Crocker) > 4. Re: Correct name for early TCP/IP working group? (Vint Cerf) > 5. Re: Correct name for early TCP/IP working group? (Dave Crocker) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Fri, 24 Jan 2025 08:52:19 -0500 (EST) > From: jnc at mercury.lcs.mit.edu (Noel Chiappa) > To: internet-history at elists.isoc.org > Cc: jnc at mercury.lcs.mit.edu > Subject: [ih] Correct name for early TCP/IP working group? > Message-ID: <20250124135219.CA54118C0D4 at mercury.lcs.mit.edu> > > > So, what was the correct full name for the early engineering group that met > every couple of months to work on TCP/IP, before the IETF was set up? (I > have > looked for it in Jon Postel's excellent minutes of its meetings, preserved > in > the Internet Experiment Note series, but I havn't found it there.) > > I have this memory that we called it the 'Internet Working Group', but of > course (especially in shortened form) that could be confused with the INWG > (the 'International Network Working Group'), a different early group > entirly. > > Noel > > > ------------------------------ > > Message: 2 > Date: Fri, 24 Jan 2025 09:08:14 -0500 > From: Guy Almes > To: Noel Chiappa , > internet-history at elists.isoc.org > Subject: Re: [ih] Correct name for early TCP/IP working group? > Message-ID: > Content-Type: text/plain; charset=UTF-8; format=flowed > > Hi Noel, > You were there long before I was, but I recall that, in the upper > left corner of RFCs and Internet Drafts, we were always to put down > "Network Working Group". > Even though we were organized into an IETF with Working Groups, etc., > that old name had to be there for some reason. > -- Guy > > On 1/24/25 8:52 AM, Noel Chiappa via Internet-history wrote: > > So, what was the correct full name for the early engineering group that > met > > every couple of months to work on TCP/IP, before the IETF was set up? (I > have > > looked for it in Jon Postel's excellent minutes of its meetings, > preserved in > > the Internet Experiment Note series, but I havn't found it there.) > > > > I have this memory that we called it the 'Internet Working Group', but of > > course (especially in shortened form) that could be confused with the > INWG > > (the 'International Network Working Group'), a different early group > entirly. > > > > Noel > > -- > > Internet-history mailing list > > Internet-history at elists.isoc.org > > https://urldefense.com/v3/__https://elists.isoc.org/mailman/listinfo/ > > internet-history__;!!KwNVnqRv! > > CmzgJn96jrPDxtRTnatHHTCBN2TAXU8BefHlcjZpgPJuRVLhYsdmbUEVlQ7NPmOnBeigt- > > Och0cm2u9EORW6deYdkij7FA$ < > https://urldefense.com/v3/__https://elists.isoc.org/mailman/listinfo/internet-history__;!!KwNVnqRv!CmzgJn96jrPDxtRTnatHHTCBN2TAXU8BefHlcjZpgPJuRVLhYsdmbUEVlQ7NPmOnBeigt-Och0cm2u9EORW6deYdkij7FA$ > > > > > > > > ------------------------------ > > Message: 3 > Date: Fri, 24 Jan 2025 09:19:27 -0500 > From: Steve Crocker > To: Guy Almes > Cc: Noel Chiappa , > internet-history at elists.isoc.org > Subject: Re: [ih] Correct name for early TCP/IP working group? > Message-ID: > Content-Type: text/plain; charset=utf-8 > > It had to be there because I said so :). See RFC 3. > > Steve > > Sent from my iPhone > > > On Jan 24, 2025, at 9:08?AM, Guy Almes via Internet-history < > internet-history at elists.isoc.org> wrote: > > > > ?Hi Noel, > > You were there long before I was, but I recall that, in the upper left > corner of RFCs and Internet Drafts, we were always to put down "Network > Working Group". > > Even though we were organized into an IETF with Working Groups, etc., > that old name had to be there for some reason. > > -- Guy > > > >> On 1/24/25 8:52 AM, Noel Chiappa via Internet-history wrote: > >> So, what was the correct full name for the early engineering group that > met > >> every couple of months to work on TCP/IP, before the IETF was set up? > (I have > >> looked for it in Jon Postel's excellent minutes of its meetings, > preserved in > >> the Internet Experiment Note series, but I havn't found it there.) > >> I have this memory that we called it the 'Internet Working Group', but > of > >> course (especially in shortened form) that could be confused with the > INWG > >> (the 'International Network Working Group'), a different early group > entirly. > >> Noel > >> -- > >> Internet-history mailing list > >> Internet-history at elists.isoc.org > >> https://urldefense.com/v3/__https://elists.isoc.org/mailman/listinfo/ > internet-history__;!!KwNVnqRv! > CmzgJn96jrPDxtRTnatHHTCBN2TAXU8BefHlcjZpgPJuRVLhYsdmbUEVlQ7NPmOnBeigt- > Och0cm2u9EORW6deYdkij7FA$ < > https://urldefense.com/v3/__https://elists.isoc.org/mailman/listinfo/internet-history__;!!KwNVnqRv!CmzgJn96jrPDxtRTnatHHTCBN2TAXU8BefHlcjZpgPJuRVLhYsdmbUEVlQ7NPmOnBeigt-Och0cm2u9EORW6deYdkij7FA$ > > > > > > -- > > Internet-history mailing list > > Internet-history at elists.isoc.org > > https://elists.isoc.org/mailman/listinfo/internet-history > > > ------------------------------ > > Message: 4 > Date: Fri, 24 Jan 2025 09:28:58 -0500 > From: Vint Cerf > To: Steve Crocker > Cc: Guy Almes , Noel Chiappa > , internet-history at elists.isoc.org > Subject: Re: [ih] Correct name for early TCP/IP working group? > Message-ID: > < > CAHxHggeMijQarW-fnLzAodG2eibJ5v6oCOT-9yY+kkMMSBMkMQ at mail.gmail.com> > Content-Type: text/plain; charset="UTF-8" > > so there were a few groups. NWG was Steve's creation and was the nexus in > which NCP, FTP, SMTP, TELNET and other Arpanet-based protocols were > developed. In 1972, an International Network Working Group (INWG) was > created and had its own set of notes called INWG Notes - these have all > been digitized thanks to Alex McKenzie (ex-BBN). Many of the same people > who worked on the Arpanet protocols also worked on the Internet Protocols > and the leads formed the Internet Configuration Control Board (ICCB) before > it was renamed the Internet Advisory Board by Barry Leiner and then > Internet Activities Board and then, with the merge with ISOC, the Internet > Architecture Board. The IETF and IRTF grew out of working groups (ten of > them more or less) of the IAB, the largest of which was IETF. > > v > > > On Fri, Jan 24, 2025 at 9:19?AM Steve Crocker via Internet-history < > internet-history at elists.isoc.org> wrote: > > > It had to be there because I said so :). See RFC 3. > > > > Steve > > > > Sent from my iPhone > > > > > On Jan 24, 2025, at 9:08?AM, Guy Almes via Internet-history < > > internet-history at elists.isoc.org> wrote: > > > > > > ?Hi Noel, > > > You were there long before I was, but I recall that, in the upper left > > corner of RFCs and Internet Drafts, we were always to put down "Network > > Working Group". > > > Even though we were organized into an IETF with Working Groups, etc., > > that old name had to be there for some reason. > > > -- Guy > > > > > >> On 1/24/25 8:52 AM, Noel Chiappa via Internet-history wrote: > > >> So, what was the correct full name for the early engineering group > that > > met > > >> every couple of months to work on TCP/IP, before the IETF was set up? > > (I have > > >> looked for it in Jon Postel's excellent minutes of its meetings, > > preserved in > > >> the Internet Experiment Note series, but I havn't found it there.) > > >> I have this memory that we called it the 'Internet Working Group', but > > of > > >> course (especially in shortened form) that could be confused with the > > INWG > > >> (the 'International Network Working Group'), a different early group > > entirly. > > >> Noel > > >> -- > > >> Internet-history mailing list > > >> Internet-history at elists.isoc.org > > >> https://urldefense.com/v3/__https://elists.isoc.org/mailman/listinfo/ > > internet-history__;!!KwNVnqRv! > > CmzgJn96jrPDxtRTnatHHTCBN2TAXU8BefHlcjZpgPJuRVLhYsdmbUEVlQ7NPmOnBeigt- > > Och0cm2u9EORW6deYdkij7FA$ < > > > https://urldefense.com/v3/__https://elists.isoc.org/mailman/listinfo/internet-history__;!!KwNVnqRv!CmzgJn96jrPDxtRTnatHHTCBN2TAXU8BefHlcjZpgPJuRVLhYsdmbUEVlQ7NPmOnBeigt-Och0cm2u9EORW6deYdkij7FA$ > > > > > > > > > -- > > > Internet-history mailing list > > > Internet-history at elists.isoc.org > > > https://elists.isoc.org/mailman/listinfo/internet-history > > -- > > Internet-history mailing list > > Internet-history at elists.isoc.org > > https://elists.isoc.org/mailman/listinfo/internet-history > > > > > -- > Please send any postal/overnight deliveries to: > Vint Cerf > Google, LLC > 1900 Reston Metro Plaza, 16th Floor > Reston, VA 20190 > +1 (571) 213 1346 > > > until further notice > > > ------------------------------ > > Message: 5 > Date: Fri, 24 Jan 2025 15:07:30 +0000 (UTC) > From: Dave Crocker > To: Vint Cerf , Steve Crocker > Cc: Guy Almes , Noel Chiappa > , internet-history at elists.isoc.org > Subject: Re: [ih] Correct name for early TCP/IP working group? > Message-ID: > Content-Type: text/plain; charset=UTF-8; format=flowed > > On 1/24/2025 6:28 AM, Vint Cerf via Internet-history wrote: > > notes called INWG Notes > > > As an outsider to the effort, who was vaguely 'around', when I read > Noel's note and before reading Vint's, INWG was the name that I thought > of.? Count this as an informal survey result of community perception... > > d/ > > -- > Dave Crocker > > Brandenburg InternetWorking > bbiw.net > bluesky: @dcrocker.bsky.social > mast: @dcrocker at mastodon.social > > > ------------------------------ > > Subject: Digest Footer > > Internet-history mailing list > Internet-history at elists.isoc.org > https://elists.isoc.org/mailman/listinfo/internet-history > > > ------------------------------ > > End of Internet-history Digest, Vol 62, Issue 9 > *********************************************** > -- Internet-history mailing list Internet-history at elists.isoc.org https://elists.isoc.org/mailman/listinfo/internet-history From jnc at mercury.lcs.mit.edu Fri Jan 24 19:46:49 2025 From: jnc at mercury.lcs.mit.edu (Noel Chiappa) Date: Fri, 24 Jan 2025 22:46:49 -0500 (EST) Subject: [ih] Correct name for early TCP/IP working group? Message-ID: <20250125034649.473FB18C086@mercury.lcs.mit.edu> > From: Tom Lyon > Here's some physical evidence in support of INWG: Err, no. As my original message said: could be confused with the INWG (the 'International Network Working Group'), a different early group entirly. Your scanned message refers to a meeting in 1973; the meetings I am talking about: https://gunkies.org/wiki/TCP_and_Internet_Meetings started in 1977. Also, that meeting was in NY; none of the meetings I am talking about were in NY; they were all at various DARPA contractors who were working on pieces of the DARPA internetworking project, none of them in NY. So maybe that's the name to give them 'DARPA internetwork project meetings'? (We can't use 'DARPA internet project meetings', because the morons at the AP have completely screwed up the careful distinction between 'internet' and 'Internet'.) Nobody seems to be able to point to a name for them? We are so lucky we have all of Jon Postel's minutes for these meetings. When we're all gone, his record of them will remain; and they were an _incredibly_ important part of the early history of what later became the Internet. Noel From jack at 3kitty.org Fri Jan 24 20:04:03 2025 From: jack at 3kitty.org (Jack Haverty) Date: Fri, 24 Jan 2025 20:04:03 -0800 Subject: [ih] Internet-history Digest, Vol 62, Issue 9 In-Reply-To: <4a8b1e37-99f1-4690-a622-6168c7b22796@dcrocker.net> References: <4a8b1e37-99f1-4690-a622-6168c7b22796@dcrocker.net> Message-ID: <8a6ada4f-b089-4e72-8d42-ac3f2ff62449@3kitty.org> On 1/24/25 18:10, Dave Crocker via Internet-history wrote: > On 1/24/2025 6:05 PM, vinton cerf via Internet-history wrote: >> that's a good question - I am not sure it was ever documented with >> minutes. >> It was pretty informal. Pouzin, Crocker, Cerf, Metcalfe (?) and many >> others >> were there because of ICCC and the demonstration of the Arpanet. > > Whoever was there an is available should record their recollections > from then.? Don't worry about the burden of writing.? Get the snapshots. > > > d/ > OK.... At ICCC'72, I was one of the crew setting up the exhibition in the ballroom of the Washington Hilton and getting it to work.?? The ARPANET was only a few years old, and Lick had tasked his "chief of staff", Al Vezza, to help make ICCC'72 a success as the "coming out party" for the ARPANET.?? Lick was my thesis advisor and I had stayed on after graduation in his group as a Member of the Research Staff.? Metcalfe had also been in Lick's group, with the job of building our PDP-10's IMP interface while simultaneously doing his thesis at Harvard and then joining Xerox PARC. A bunch of us from MIT went down to DC to help set things up for the ICCC exhibit.? That involved stuff we knew how to do -- setting up a raised floor, running cables, unpacking and assembling equipment, etc.?? It also included things we didn't know how to do -- such as getting all sorts of different terminals connected to, and successfully talking to, the TIP in the middle of the room.? In retrospect, I suspect many of those terminals, loaned to the Exhibit by all sorts of vendors, had never been connected to a TIP before. We tweaked hardware, software, rewired connectors, and even rewired a backplane or two to get everything (more or less?) working. There were some interesting experiences...?? The Ballroom was in the basement of the hotel, and we all had rooms somewhere up on the guests' floors.? We were working, so of course we were dressed in boots, jeans, T-shirts, etc.? I remember at one point several of us getting in the elevator in the basement to go up to our rooms.? I was probably still wearing my tool belt and carrying my Craftsman toolbox from MIT brought in the plane as carryon.? The door opened on some intermediate floor and a whole bunch of people in ball gowns and tuxedos got in, obviously displeased to be riding with members of the working class.? A hotel manager shortly accosted us to advise that we had to use the freight elevators, since the primary elevators were only for use by registered guests.?? I wish I had a camera to take his picture when we showed him our room keys, proving we actually were registered guests! It is important to remember that the world of computing was quite different 50+ years ago.? Computers were expensive and required large spaces.? Humans used terminals on a network to access a remote computer in much the same way they used a terminal in their office to access a computer elsewhere in the building.? Most "network usage" was terminals interacting with some distant computer, with occasional data transfers between two computers, to move files as needed.? No web, no PCs, no smartphones, no speeds faster than about 100 kilobits per second, no Terabyte disk drives or SD cards. Dozens of computers, not millions.? To access a remote computer you used a telephone to "dial up" to that computer, and you were charged by the minute. I didn't have time to see any of the presentations.?? We were busy at first getting everything to work, then we served as docents for the Conference attendees as they came to view the Exhibit.? Bob Metcalfe had collected a bunch of "scenarios" for the various ARPANET sites, and the idea was that the attendees would get a hands-on experience by actually using the ARPANET themselves.?? A scan of my copy of that "scenarios" handout is here: https://drive.google.com/file/d/1F6OJOu8K7Mcfwdh_oTpZAO1J0U71VyBP/view?usp=sharing The main controversy at the time seemed to be the difference between existing networking techniques and the new-fangled technique called "packet switching".? I recall the ATT attendees were quite pleased when the TIP crashed as they were watching a demo. Circuit-switching wins! Most of the attendees, at least the ones that I helped, didn't care much about the technical aspects.?? They were more focused on what they could do by using the ARPANET, and how they could access some useful program or data at some other distant site.? And of course the fact that it worked.? They had used it themselves and not just watched a demo by Marketing. I don't recall hearing any discussions about "internetworking", distinct from discussions of "networking".?? Established networking practice used circuit switching, and interconnecting networks was accomplished by "plugging" virtual circuits from one network to another.?? That was basically how the telephony world had worked for quite a while. Inside the ARPANET IMPs, algorithms implemented a virtual circuit service between any two computers attached to the net.?? Packet switching was used between the IMPs, but the users' computers saw a virtual circuit service.? What you sent came out the other end, intact and in order.?? There were lots of mechanisms and algorithms inside the IMP code to accomplish that.? So it would have been straightforward to interconnect the ARPANET with a clone of itself, patching together virtual circuits at the interface. Again in retrospect, it's not surprising that the CCITT evolved to adopt X.25 and X.75 as internetworking techniques for data communications.? That was the obvious extension of the mechanisms that had been used to interconnect PTTs' networks to carry voice calls across country borders. The TCP approach was an unproven competitor.?? Hence the "Internet Experiment". Alex's recollections and the old INWG notes probably contain a lot more detail about those meeting discussions. Sadly I suspect there are no Youtube videos available..... Jack Haverty -------------- next part -------------- A non-text attachment was scrubbed... Name: OpenPGP_signature.asc Type: application/pgp-signature Size: 665 bytes Desc: OpenPGP digital signature URL: From agmalis at gmail.com Sat Jan 25 06:28:32 2025 From: agmalis at gmail.com (Andrew G. Malis) Date: Sat, 25 Jan 2025 09:28:32 -0500 Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: <431e9222-d1f4-4245-88ca-d286ae7f44a4@3kitty.org> References: <431e9222-d1f4-4245-88ca-d286ae7f44a4@3kitty.org> Message-ID: Jack, - There was some work however within the ARPANET IMP software to > acknowledge the need for multiple networks. For example, some of the > formats of data as it passed through the ARPANET included fields > labelled "Network Number". AFAIK this was never actually fully > implemented so the ARPANET itself never achieved connectivity between > multiple networks until TCP was deployed. As I recall, the ARPANET's "network number" field was used during the ARPANET/MILNET split to allow logical separation of the IMPs and hosts sharing the same backbone infrastructure until the physical split could be completed. Cheers, Andy From vgcerf at gmail.com Sat Jan 25 06:43:30 2025 From: vgcerf at gmail.com (vinton cerf) Date: Sat, 25 Jan 2025 09:43:30 -0500 Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: References: <431e9222-d1f4-4245-88ca-d286ae7f44a4@3kitty.org> Message-ID: Thanks for that, Andy - I had not realized that this was used as a scaffolding - must have affected routing in some way and created a kind of gateway IMP notion for IMPs lying along a border between two connected subsets? v On Sat, Jan 25, 2025 at 9:28?AM Andrew G. Malis via Internet-history < internet-history at elists.isoc.org> wrote: > Jack, > > - There was some work however within the ARPANET IMP software to > > acknowledge the need for multiple networks. For example, some of the > > formats of data as it passed through the ARPANET included fields > > labelled "Network Number". AFAIK this was never actually fully > > implemented so the ARPANET itself never achieved connectivity between > > multiple networks until TCP was deployed. > > > As I recall, the ARPANET's "network number" field was used during the > ARPANET/MILNET split to allow logical separation of the IMPs and hosts > sharing the same backbone infrastructure until the physical split could be > completed. > > Cheers, > Andy > -- > Internet-history mailing list > Internet-history at elists.isoc.org > https://elists.isoc.org/mailman/listinfo/internet-history > From jeanjour at comcast.net Sat Jan 25 06:48:46 2025 From: jeanjour at comcast.net (John Day) Date: Sat, 25 Jan 2025 09:48:46 -0500 Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: References: <431e9222-d1f4-4245-88ca-d286ae7f44a4@3kitty.org> Message-ID: <5A3B769F-F760-476C-B963-9D63B9BCE680@comcast.net> This brings up the question, when did IMPs cease to be front-end+router and at least some of them become just routers? I know that we started to think of them that way long before they actually were that way, if they were. Take care, John > On Jan 25, 2025, at 09:43, vinton cerf via Internet-history wrote: > > Thanks for that, Andy - I had not realized that this was used as a > scaffolding - must have affected routing in some way and created a kind of > gateway IMP notion for IMPs lying along a border between two connected > subsets? > > v > > > On Sat, Jan 25, 2025 at 9:28?AM Andrew G. Malis via Internet-history < > internet-history at elists.isoc.org> wrote: > >> Jack, >> >> - There was some work however within the ARPANET IMP software to >>> acknowledge the need for multiple networks. For example, some of the >>> formats of data as it passed through the ARPANET included fields >>> labelled "Network Number". AFAIK this was never actually fully >>> implemented so the ARPANET itself never achieved connectivity between >>> multiple networks until TCP was deployed. >> >> >> As I recall, the ARPANET's "network number" field was used during the >> ARPANET/MILNET split to allow logical separation of the IMPs and hosts >> sharing the same backbone infrastructure until the physical split could be >> completed. >> >> Cheers, >> Andy >> -- >> Internet-history mailing list >> Internet-history at elists.isoc.org >> https://elists.isoc.org/mailman/listinfo/internet-history >> > -- > Internet-history mailing list > Internet-history at elists.isoc.org > https://elists.isoc.org/mailman/listinfo/internet-history From vgcerf at gmail.com Sat Jan 25 06:58:45 2025 From: vgcerf at gmail.com (vinton cerf) Date: Sat, 25 Jan 2025 09:58:45 -0500 Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: <5A3B769F-F760-476C-B963-9D63B9BCE680@comcast.net> References: <431e9222-d1f4-4245-88ca-d286ae7f44a4@3kitty.org> <5A3B769F-F760-476C-B963-9D63B9BCE680@comcast.net> Message-ID: John, that might depend on what you mean by "front-end" - Maybe you are distinguishing between the host/imp interface which accepted "messages" and turned them into packets and the part that just routed packets. Similarly on receipt, reassembling packets into messages before delivering them to hosts. If that is the correct parse, I am not sure there was ever a change. Host/IMP and IMP/IMP interactions were always pretty distinguishable, I think. Once we put in "real" gateway between Arpanet and Milnet, the Host/IMP interfaces were used to serve the gateway hosts, presumably. v v On Sat, Jan 25, 2025 at 9:48?AM John Day wrote: > This brings up the question, when did IMPs cease to be front-end+router > and at least some of them become just routers? > > I know that we started to think of them that way long before they actually > were that way, if they were. > > Take care, > John > > > On Jan 25, 2025, at 09:43, vinton cerf via Internet-history < > internet-history at elists.isoc.org> wrote: > > > > Thanks for that, Andy - I had not realized that this was used as a > > scaffolding - must have affected routing in some way and created a kind > of > > gateway IMP notion for IMPs lying along a border between two connected > > subsets? > > > > v > > > > > > On Sat, Jan 25, 2025 at 9:28?AM Andrew G. Malis via Internet-history < > > internet-history at elists.isoc.org> wrote: > > > >> Jack, > >> > >> - There was some work however within the ARPANET IMP software to > >>> acknowledge the need for multiple networks. For example, some of the > >>> formats of data as it passed through the ARPANET included fields > >>> labelled "Network Number". AFAIK this was never actually fully > >>> implemented so the ARPANET itself never achieved connectivity between > >>> multiple networks until TCP was deployed. > >> > >> > >> As I recall, the ARPANET's "network number" field was used during the > >> ARPANET/MILNET split to allow logical separation of the IMPs and hosts > >> sharing the same backbone infrastructure until the physical split could > be > >> completed. > >> > >> Cheers, > >> Andy > >> -- > >> Internet-history mailing list > >> Internet-history at elists.isoc.org > >> https://elists.isoc.org/mailman/listinfo/internet-history > >> > > -- > > Internet-history mailing list > > Internet-history at elists.isoc.org > > https://elists.isoc.org/mailman/listinfo/internet-history > > From jeanjour at comcast.net Sat Jan 25 07:32:31 2025 From: jeanjour at comcast.net (John Day) Date: Sat, 25 Jan 2025 10:32:31 -0500 Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: References: <431e9222-d1f4-4245-88ca-d286ae7f44a4@3kitty.org> <5A3B769F-F760-476C-B963-9D63B9BCE680@comcast.net> Message-ID: <034327A0-3647-459C-8C33-3707090AE213@comcast.net> That would be my guess, that there really was no change. Prior to sending out the ARPANET RFQ, Roberts had a meeting of the potential sites at Ann Arbor. There he got a lot of pushback that the sites' systems were already so overloaded that they couldn?t possibly support a network. Roberts left the meeting very unhappy. ;-) On the way to the airport, Wes Clark suggested putting a minicomputer in front of each host as a front-end, which is what Roberts wrote into the RFQ. Hence, IMPs. The initial distance restrictions of 1822 and the IMP-Host protocol reflect that front-end functionality as well as that IMPs delivered whole messages to the Hosts, reassembling packets into messages. (To satisfy the objections from the Ann Arbor meeting.) The host requested that its IMP create a ?connection? to the destination IMP and host. Then NCP operated process-to-process over that. NCP carried socket ids, not host ids. (There is a good chance the terminology then is the same as the terminology now.) ;-) Take care, John > On Jan 25, 2025, at 09:58, vinton cerf wrote: > > John, that might depend on what you mean by "front-end" - > Maybe you are distinguishing between the host/imp interface which accepted "messages" and turned them into packets and the part that just routed packets. Similarly on receipt, reassembling packets into messages before delivering them to hosts. > > If that is the correct parse, I am not sure there was ever a change. Host/IMP and IMP/IMP interactions were always pretty distinguishable, I think. Once we put in "real" gateway between Arpanet and Milnet, the Host/IMP interfaces were used to serve the gateway hosts, presumably. > > v > > > v > > > On Sat, Jan 25, 2025 at 9:48?AM John Day > wrote: >> This brings up the question, when did IMPs cease to be front-end+router and at least some of them become just routers? >> >> I know that we started to think of them that way long before they actually were that way, if they were. >> >> Take care, >> John >> >> > On Jan 25, 2025, at 09:43, vinton cerf via Internet-history > wrote: >> > >> > Thanks for that, Andy - I had not realized that this was used as a >> > scaffolding - must have affected routing in some way and created a kind of >> > gateway IMP notion for IMPs lying along a border between two connected >> > subsets? >> > >> > v >> > >> > >> > On Sat, Jan 25, 2025 at 9:28?AM Andrew G. Malis via Internet-history < >> > internet-history at elists.isoc.org > wrote: >> > >> >> Jack, >> >> >> >> - There was some work however within the ARPANET IMP software to >> >>> acknowledge the need for multiple networks. For example, some of the >> >>> formats of data as it passed through the ARPANET included fields >> >>> labelled "Network Number". AFAIK this was never actually fully >> >>> implemented so the ARPANET itself never achieved connectivity between >> >>> multiple networks until TCP was deployed. >> >> >> >> >> >> As I recall, the ARPANET's "network number" field was used during the >> >> ARPANET/MILNET split to allow logical separation of the IMPs and hosts >> >> sharing the same backbone infrastructure until the physical split could be >> >> completed. >> >> >> >> Cheers, >> >> Andy >> >> -- >> >> Internet-history mailing list >> >> Internet-history at elists.isoc.org >> >> https://elists.isoc.org/mailman/listinfo/internet-history >> >> >> > -- >> > Internet-history mailing list >> > Internet-history at elists.isoc.org >> > https://elists.isoc.org/mailman/listinfo/internet-history >> From lars at nocrew.org Sat Jan 25 07:42:26 2025 From: lars at nocrew.org (Lars Brinkhoff) Date: Sat, 25 Jan 2025 15:42:26 +0000 Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: <034327A0-3647-459C-8C33-3707090AE213@comcast.net> (John Day via Internet-history's message of "Sat, 25 Jan 2025 10:32:31 -0500") References: <431e9222-d1f4-4245-88ca-d286ae7f44a4@3kitty.org> <5A3B769F-F760-476C-B963-9D63B9BCE680@comcast.net> <034327A0-3647-459C-8C33-3707090AE213@comcast.net> Message-ID: <7wikq23nod.fsf@junk.nocrew.org> John Day wrote: > Then NCP operated process-to-process over that. NCP carried socket > ids, not host ids. And links. RFC (STR or RTS) and CLS messages identify a connection by the remote and local socket numbers. Data messages and flow control identify the connection by the link number set up during RFC. (I recently wrote a new NCP to test an emulated ARPANET.) From jeanjour at comcast.net Sat Jan 25 08:04:00 2025 From: jeanjour at comcast.net (John Day) Date: Sat, 25 Jan 2025 11:04:00 -0500 Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: <7wikq23nod.fsf@junk.nocrew.org> References: <431e9222-d1f4-4245-88ca-d286ae7f44a4@3kitty.org> <5A3B769F-F760-476C-B963-9D63B9BCE680@comcast.net> <034327A0-3647-459C-8C33-3707090AE213@comcast.net> <7wikq23nod.fsf@junk.nocrew.org> Message-ID: <4D9AE787-5508-4952-9C2F-B717A345F462@comcast.net> Correct. Forgot about that and don?t forget the control link!! Believe it or not, a US Defense Contractor (CSC) creating a copy of the ARPANET thought it would be too much trouble to open a connection to send mail and was going to send it on the Control Link. (!!) A friend at MITRE made them at least change the title of the document to ?Non-functional Specification . . .? ;-) lol (Not surprising, the copy was 7 times slower.) lol Ahh, those were the days! John > On Jan 25, 2025, at 10:42, Lars Brinkhoff wrote: > > John Day wrote: >> Then NCP operated process-to-process over that. NCP carried socket >> ids, not host ids. > > And links. > > RFC (STR or RTS) and CLS messages identify a connection by the remote > and local socket numbers. Data messages and flow control identify the > connection by the link number set up during RFC. (I recently wrote a > new NCP to test an emulated ARPANET.) From agmalis at gmail.com Sat Jan 25 08:42:19 2025 From: agmalis at gmail.com (Andrew G. Malis) Date: Sat, 25 Jan 2025 11:42:19 -0500 Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: References: <431e9222-d1f4-4245-88ca-d286ae7f44a4@3kitty.org> Message-ID: Vint and Jack, There weren't any "gateway" functions per se in the IMPs. Rather, the IMPs kept the same unique IMP numbering (no IMP duplicate numbers) until the physical split was complete. Thus, routing continued as usual for the single merged network. The logical split was enforced by only allowing hosts on the same "network" to communicate. Thus, any gateway functionality would have to occur in a host, as usual. Once the physical split was complete, each network then had its own independent routing, and we could have duplicate IMP numbers in the two networks. We then added network numbering to the routing updates and link management to forestall any issues if AT&T accidentally cross-connected the network links. As I recall, this did happen on occasion. Alan Hill would probably have the best recollection of any operational issues that came up as a result of accidental cross-connection. Cheers, Andy On Sat, Jan 25, 2025 at 9:43?AM vinton cerf wrote: > Thanks for that, Andy - I had not realized that this was used as a > scaffolding - must have affected routing in some way and created a kind of > gateway IMP notion for IMPs lying along a border between two connected > subsets? > > v > > > On Sat, Jan 25, 2025 at 9:28?AM Andrew G. Malis via Internet-history < > internet-history at elists.isoc.org> wrote: > >> Jack, >> >> - There was some work however within the ARPANET IMP software to >> > acknowledge the need for multiple networks. For example, some of the >> > formats of data as it passed through the ARPANET included fields >> > labelled "Network Number". AFAIK this was never actually fully >> > implemented so the ARPANET itself never achieved connectivity between >> > multiple networks until TCP was deployed. >> >> >> As I recall, the ARPANET's "network number" field was used during the >> ARPANET/MILNET split to allow logical separation of the IMPs and hosts >> sharing the same backbone infrastructure until the physical split could be >> completed. >> >> Cheers, >> Andy >> -- >> Internet-history mailing list >> Internet-history at elists.isoc.org >> https://elists.isoc.org/mailman/listinfo/internet-history >> > From vint at google.com Sat Jan 25 09:05:06 2025 From: vint at google.com (Vint Cerf) Date: Sat, 25 Jan 2025 12:05:06 -0500 Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: References: <431e9222-d1f4-4245-88ca-d286ae7f44a4@3kitty.org> Message-ID: Thanks for that bit of lore, Andy!! v On Sat, Jan 25, 2025 at 11:42?AM Andrew G. Malis via Internet-history < internet-history at elists.isoc.org> wrote: > Vint and Jack, > > There weren't any "gateway" functions per se in the IMPs. Rather, the IMPs > kept the same unique IMP numbering (no IMP duplicate numbers) until the > physical split was complete. Thus, routing continued as usual for the > single merged network. The logical split was enforced by only allowing > hosts on the same "network" to communicate. Thus, any gateway functionality > would have to occur in a host, as usual. > > Once the physical split was complete, each network then had its own > independent routing, and we could have duplicate IMP numbers in the two > networks. We then added network numbering to the routing updates and link > management to forestall any issues if AT&T accidentally cross-connected the > network links. As I recall, this did happen on occasion. Alan Hill would > probably have the best recollection of any operational issues that came up > as a result of accidental cross-connection. > > Cheers, > Andy > > > On Sat, Jan 25, 2025 at 9:43?AM vinton cerf wrote: > > > Thanks for that, Andy - I had not realized that this was used as a > > scaffolding - must have affected routing in some way and created a kind > of > > gateway IMP notion for IMPs lying along a border between two connected > > subsets? > > > > v > > > > > > On Sat, Jan 25, 2025 at 9:28?AM Andrew G. Malis via Internet-history < > > internet-history at elists.isoc.org> wrote: > > > >> Jack, > >> > >> - There was some work however within the ARPANET IMP software to > >> > acknowledge the need for multiple networks. For example, some of the > >> > formats of data as it passed through the ARPANET included fields > >> > labelled "Network Number". AFAIK this was never actually fully > >> > implemented so the ARPANET itself never achieved connectivity between > >> > multiple networks until TCP was deployed. > >> > >> > >> As I recall, the ARPANET's "network number" field was used during the > >> ARPANET/MILNET split to allow logical separation of the IMPs and hosts > >> sharing the same backbone infrastructure until the physical split could > be > >> completed. > >> > >> Cheers, > >> Andy > >> -- > >> Internet-history mailing list > >> Internet-history at elists.isoc.org > >> https://elists.isoc.org/mailman/listinfo/internet-history > >> > > > -- > Internet-history mailing list > Internet-history at elists.isoc.org > https://elists.isoc.org/mailman/listinfo/internet-history > -- Please send any postal/overnight deliveries to: Vint Cerf Google, LLC 1900 Reston Metro Plaza, 16th Floor Reston, VA 20190 +1 (571) 213 1346 until further notice From aam3sendonly at gmail.com Sat Jan 25 11:16:00 2025 From: aam3sendonly at gmail.com (Alexander McKenzie) Date: Sat, 25 Jan 2025 14:16:00 -0500 Subject: [ih] Some random thoughts about ICCC 72 Message-ID: As I recall, Bob Kahn was in charge of the ARPAnet demo. As I recall, Al Vezza was Bob's assistant. ARPA had BBN send a TIP to the hotel. AT&T donated (I believe) a 50kbps circuit from some other nearby Washington site, including the loan of two Bell 303 50kbps modems. I believe ARPA paid for the lease of the ballroom, but it may have been donated by the ICCC management. Bob and Al arranged for the loan of a raised floor and about 60 character-at-a-time terminals, including one or two Imlacs, a line printer, Model 33 TTY's, and glass TTYs. Some contractor (MIT, BBN, .. ?) brought a Logo turtle. Most of these terminals had never before been connected to a TIP and there were exciting times getting them all working properly. Most of the work was done by engineers and grad students being supported by ARPA contracts who were sent by their managers to support the demo. IMPs/TIPs expected to be connected to the rest of the net by two or more circuits. They were quick to declare a circuit "down" so traffic would be routed through an alternate path, rather than queuing for a flaky circuit. In view of the fact that the demo TIP was connected by only a single circuit, we patched the demo TIP and the IMP it was connected to to be a lot slower to declare the line down, in hopes that a flaky line would get at least some traffic to and from the demo. This turned out to be a disaster - when the line got flaky, congestion quickly spread through the entire ARPAnet and the demos stopped working anyway. After a little experience we removed the patches. Bob and AL had arranged to have a film made that would explain the ARPAnet and why ARPA had built it - the title included the phrase "Heralds of Resource Sharing." For some reason the film wasn't finished in time, so Bob and Al put together a slide show (one full slide carousel) that ran continuously in a corridor outside the ballroom. At one point, the Logo turtle appeared to have gone crazy. It turned out someone had switched TIP port assignments so the line printer output was going to the turtle, and turtle commands were going to the line printer. I believe most conference attendees arrived thinking packet switching was a foolish departure from standard communication concepts, and thanks to the demo went home believing that packet switching was a viable technology. In this sense the domo was a success. A few months before the demo, many (most?) of the ARPAnet host sites were not able to make use of the ARPAnet to provide access to other hosts or to support remote users. Pressure on every site by Larry Roberts to participate in the demo got almost every site up to speed, and the ARPAnet began to be used much more. In this sense, too, the demo was a success. Cheers, Alex From lk at cs.ucla.edu Sat Jan 25 12:55:30 2025 From: lk at cs.ucla.edu (Leonard Kleinrock) Date: Sat, 25 Jan 2025 12:55:30 -0800 (PST) Subject: [ih] Some random thoughts about ICCC 72 Message-ID: <1027308858.296176289.1737838530269.JavaMail.zimbra@mail.cs.ucla.edu> ? Alex got the story correct and added some great color to its description. Well done Alex. I remember the turtle disaster since we practiced the UCLA demonstration the night before and saw our great experiment of moving moving files back-and-forth across the country to deliver a printed output to Jon Postel sitting at a terminal there in the basement, but instead, as Alex described, the output went to turtle and turtle jumped around the room with the output - hilarious!. We fixed it in time for a proper and successful demo the next morning. Interestingly, not only did the public walk away, convinced that packet switching was a viable technology, but also the many ARPA contractors and PI?s were themselves convinced of its viability and usefulness, which was not the case before ICCC 72. All the best, Len Sent from my iPhone > On Jan 25, 2025, at 11:16?AM, Alexander McKenzie via Internet-history wrote: > ?As I recall, Bob Kahn was in charge of the ARPAnet demo. As I recall, Al > Vezza was Bob's assistant. > > ARPA had BBN send a TIP to the hotel. AT&T donated (I believe) a 50kbps > circuit from some other nearby Washington site, including the loan of two > Bell 303 50kbps modems. > > I believe ARPA paid for the lease of the ballroom, but it may have been > donated by the ICCC management. > > Bob and Al arranged for the loan of a raised floor and about 60 > character-at-a-time terminals, including one or two Imlacs, a line printer, > Model 33 TTY's, and glass TTYs. Some contractor (MIT, BBN, .. ?) brought a > Logo turtle. Most of these terminals had never before been connected to a > TIP and there were exciting times getting them all working properly. Most > of the work was done by engineers and grad students being supported by ARPA > contracts who were sent by their managers to support the demo. > > IMPs/TIPs expected to be connected to the rest of the net by two or more > circuits. They were quick to declare a circuit "down" so traffic would be > routed through an alternate path, rather than queuing for a flaky circuit. > In view of the fact that the demo TIP was connected by only a single > circuit, we patched the demo TIP and the IMP it was connected to to be a > lot slower to declare the line down, in hopes that a flaky line would get > at least some traffic to and from the demo. This turned out to be a > disaster - when the line got flaky, congestion quickly spread through the > entire ARPAnet and the demos stopped working anyway. After a little > experience we removed the patches. > > Bob and AL had arranged to have a film made that would explain the ARPAnet > and why ARPA had built it - the title included the phrase "Heralds of > Resource Sharing." For some reason the film wasn't finished in time, so > Bob and Al put together a slide show (one full slide carousel) that ran > continuously in a corridor outside the ballroom. > > At one point, the Logo turtle appeared to have gone crazy. It turned out > someone had switched TIP port assignments so the line printer output was > going to the turtle, and turtle commands were going to the line printer. > > I believe most conference attendees arrived thinking packet switching was a > foolish departure from standard communication concepts, and thanks to the > demo went home believing that packet switching was a viable technology. In > this sense the domo was a success. > > A few months before the demo, many (most?) of the ARPAnet host sites were > not able to make use of the ARPAnet to provide access to other hosts or to > support remote users. Pressure on every site by Larry Roberts to > participate in the demo got almost every site up to speed, and the ARPAnet > began to be used much more. In this sense, too, the demo was a success. > > Cheers, > Alex > -- > Internet-history mailing list > Internet-history at elists.isoc.org > https://elists.isoc.org/mailman/listinfo/internet-history From j at shoch.com Sat Jan 25 15:38:37 2025 From: j at shoch.com (John Shoch) Date: Sat, 25 Jan 2025 15:38:37 -0800 Subject: [ih] Internet-history Digest, Vol 62, Issue 12 In-Reply-To: References: Message-ID: Noel, Thanks for putting up a page on the Internet Experiment Notes, and early meetings. I was at some of the meetings, beginning ca. 1977 (but was certainly not at all of them). The minutes reflect the work of many smart people grappling with new challenges. They also show "how the sausage was made....." ca. 1977-1978. Other people probably know a lot more than I do, but some recollections and context: --They were well organized meetings, chaired by Vint. Most attendees were funded by Arpa, working directly on the projects. --This started in the TCP-era, pre-TCP/IP. --There were, at that time, two sets of meetings: Internet meetings and TCP meetings. --In general, the TCP meetings were focused on specifying and implementing TCP, and Internet meetings were focused on networks and gateways (PRNet, SatNet, etc. -- see some of the minutes). --Notes from both sets of meetings often emerged in the unified series of Internet Experiment Notes. --In some cases the two meetings were on adjacent days, with some number of people attending both. --Thus, there was sometimes important "spill over" from one meeting to the next. --Famously, in late Jan./early Feb. 1978 there were back-to-back meetings in LA, at ISI. --The first was a TCP meeting, documented by Jon Postel (but, for some reason, assigned an IEN number much later, #67). There were many status reports, and comments on many subjects. Danny Cohen talked about packet service for voice; that and fragmentation issues helped precipitate an important conversation about the structure of TCP. --From Postel's agenda and notes about the first day: "Discuss the multidestination and broadcast topic, set up a working group on it, and introduce the issue of incorporating a datagram mode and an emmission control mode of operation into TCP." "Introduction and Objectives -- Cerf. The main objective is to get TCP-3 straightened out, and to discuss extensions." "Cohen -- Uses this time to complain about TCP-3 becoming all things to all people. Also illustration of the approach to voice service via an "unreliability" package on top of TCP." [Danny at his best, using humor and sarcasm to make his point.] "Fragmentation Issues and Choices -- Cerf. Vint proposes that fragmentation be removed from TCP and be designated an internet protocol task. John Shoch presented a brief summary of his memo on this topic. A working group is to resolve this issue on Tuesday." "NSW Protocols and their Requirements -- Thomas. ...There was some discussion of how TCP could help and how MSG could use a datagram mode." --On the second day of the TCP meeting there were working groups in the morning, and reports after lunch: "(b) Fragmentation -- Cerf This group had a lengthly [sic] discussion. One immediate result was the decision to remove fragmentation from TCP and place it in the internet protocol." --The Internet meeting convened the following day, Feb. 1, 1978, again at ISI. It was also documented by Jon Postel, in IEN #22. --The minutes of the prior TCP meeting show 22 attendees. The minutes of the Internet meeting, on the 3rd day, show 26 attendees -- 13 of whom had been at the prior TCP meeting. "Introduction and Objectives -- Cerf Vint primarily presented a summary of the results of the TCP meeting which took place the preceding two days. The main result is a decision not to extend the TCP to include alternate modes of service, and instead to provide for a set of parallel protocols that all utilize a common "datagram" type internet protocol. The functions of the internet layer would be addressing and fragmentation. ..." "Other TCP meeting results were a decision to revise the TCP3 specification to correct minor points, and sometime in the future to undertake a major revision of the documentation." --That, of course, eventually led to the design and specification of the now-split architecture, TCP4 and IP4. As I said, "that's how the sausage got made....." John Shoch Message: 2 Date: Fri, 24 Jan 2025 19:06:56 -0500 (EST) From: jnc at mercury.lcs.mit.edu (Noel Chiappa) To: internet-history at elists.isoc.org Cc: jnc at mercury.lcs.mit.edu Subject: Re: [ih] Correct name for early TCP/IP working group? Message-ID: <20250125000656.0514D18C083 at mercury.lcs.mit.edu> > From: Vint Cerf > so there were a few groups. ... the leads formed the Internet > Configuration Control Board (ICCB) Right, but the ICCB was formed in about 1979. I have created a list of the early meetings (starting in 1977), here: https://gunkies.org/wiki/TCP_and_Internet_Meetings and there were about a dozen over two years before that - most of the ones in that list were of the body I'm asking about. I'm not sure it was a formal group; it was basically just your DARPA contractors on the project. So maybe it didn't have a formal name. (If so, maybe we should make one up! :-) Noel From dhc at dcrocker.net Sat Jan 25 15:52:45 2025 From: dhc at dcrocker.net (Dave Crocker) Date: Sat, 25 Jan 2025 23:52:45 +0000 (UTC) Subject: [ih] Some random thoughts about ICCC 72 In-Reply-To: <1027308858.296176289.1737838530269.JavaMail.zimbra@mail.cs.ucla.edu> References: <1027308858.296176289.1737838530269.JavaMail.zimbra@mail.cs.ucla.edu> Message-ID: <459c23f5-b65f-4c84-9959-77f4e1f31082@dcrocker.net> On 1/25/2025 12:55 PM, Leonard Kleinrock via Internet-history wrote: > Interestingly, not only did the public walk away, convinced that packet > switching was a viable technology, but also the many ARPA contractors > and PI?s were themselves convinced of its viability and usefulness, > which was not the case before ICCC 72. I was one of the floor walkers, giving demos.? At one point, someone handed me Lipinski Sr, of the Institute of the Future.? I connected to a BBN Tenex machine, did some stuff, closed the connection, opened an ISI Tenex, and started to do some stuff. Lipinski looked entirely bored by all of this. I was just demoing a time-sharing system... And then things froze.? I looked to the center of the room, saw a flurry of activity around the TIP, and told Lipinski it appeared the TIP had crashed.? He thanked me and said he'd go off to a session and come back after the machine had been rebooted. I said it would only take a couple of minutes and he said, "but they haven't rolled out the paper tape reader yet." I explained about reloading from a neighboring IMP at 56Kbps. His entire demeanor changed. In that moment, he registered what was being demonstrated. Now he understood networking. d/ ps.? I also learned a bit about doing demos... d/ -- Dave Crocker Brandenburg InternetWorking bbiw.net bluesky: @dcrocker.bsky.social mast: @dcrocker at mastodon.social From vint at google.com Sat Jan 25 15:59:52 2025 From: vint at google.com (Vint Cerf) Date: Sat, 25 Jan 2025 18:59:52 -0500 Subject: [ih] Some random thoughts about ICCC 72 In-Reply-To: <459c23f5-b65f-4c84-9959-77f4e1f31082@dcrocker.net> References: <1027308858.296176289.1737838530269.JavaMail.zimbra@mail.cs.ucla.edu> <459c23f5-b65f-4c84-9959-77f4e1f31082@dcrocker.net> Message-ID: I seem to recall one case where a biology researcher was using the terminal to connect to something and the host was not online. The TIP typed out "Host Dead" and this caused a lot of anxiety.... v On Sat, Jan 25, 2025 at 6:52?PM Dave Crocker via Internet-history < internet-history at elists.isoc.org> wrote: > On 1/25/2025 12:55 PM, Leonard Kleinrock via Internet-history wrote: > > Interestingly, not only did the public walk away, convinced that packet > > switching was a viable technology, but also the many ARPA contractors > > and PI?s were themselves convinced of its viability and usefulness, > > which was not the case before ICCC 72. > > I was one of the floor walkers, giving demos. At one point, someone > handed me Lipinski Sr, of the Institute of the Future. I connected to a > BBN Tenex machine, did some stuff, closed the connection, opened an ISI > Tenex, and started to do some stuff. Lipinski looked entirely bored by > all of this. I was just demoing a time-sharing system... > > And then things froze. I looked to the center of the room, saw a flurry > of activity around the TIP, and told Lipinski it appeared the TIP had > crashed. He thanked me and said he'd go off to a session and come back > after the machine had been rebooted. > > I said it would only take a couple of minutes and he said, "but they > haven't rolled out the paper tape reader yet." > > I explained about reloading from a neighboring IMP at 56Kbps. > > His entire demeanor changed. In that moment, he registered what was > being demonstrated. > > Now he understood networking. > > d/ > > ps. I also learned a bit about doing demos... > > d/ > > -- > Dave Crocker > > Brandenburg InternetWorking > bbiw.net > bluesky: @dcrocker.bsky.social > mast: @dcrocker at mastodon.social > -- > Internet-history mailing list > Internet-history at elists.isoc.org > https://elists.isoc.org/mailman/listinfo/internet-history > -- Please send any postal/overnight deliveries to: Vint Cerf Google, LLC 1900 Reston Metro Plaza, 16th Floor Reston, VA 20190 +1 (571) 213 1346 until further notice From lars at nocrew.org Sat Jan 25 23:28:33 2025 From: lars at nocrew.org (Lars Brinkhoff) Date: Sun, 26 Jan 2025 07:28:33 +0000 Subject: [ih] Some random thoughts about ICCC 72 In-Reply-To: (Alexander McKenzie via Internet-history's message of "Sat, 25 Jan 2025 14:16:00 -0500") References: Message-ID: <7w7c6i2fvi.fsf@junk.nocrew.org> Alexander McKenzie wrote: > Bob and AL had arranged to have a film made that would explain the > ARPAnet and why ARPA had built it - the title included the phrase > "Heralds of Resource Sharing." There are several copies online. Here's one: https://www.youtube.com/watch?v=Cdf5fMAL994 From galmes at tamu.edu Sun Jan 26 07:34:40 2025 From: galmes at tamu.edu (Guy Almes) Date: Sun, 26 Jan 2025 10:34:40 -0500 Subject: [ih] Some random thoughts about ICCC 72 In-Reply-To: <7w7c6i2fvi.fsf@junk.nocrew.org> References: <7w7c6i2fvi.fsf@junk.nocrew.org> Message-ID: <5657a8e7-fe0e-4953-bf7b-93ebc1609594@tamu.edu> Lars, I'd heard about ICCC 72 for years, but this film is so informative. Thanks, -- Guy On 1/26/25 2:28 AM, Lars Brinkhoff via Internet-history wrote: > Alexander McKenzie wrote: >> Bob and AL had arranged to have a film made that would explain the >> ARPAnet and why ARPA had built it - the title included the phrase >> "Heralds of Resource Sharing." > > There are several copies online. Here's one: > > https://urldefense.com/v3/__https://www.youtube.com/watch? > v=Cdf5fMAL994__;!!KwNVnqRv!HZNzFktVPkM- > k6hZQqAvW5b82_L1M7iaw6UvnsDirrfNV96_RR9lJwBXdg8byGT78LCQXESkvTlTPacEa4cpEepJxgP6Lg$ > -- > Internet-history mailing list > Internet-history at elists.isoc.org > https://urldefense.com/v3/__https://elists.isoc.org/mailman/listinfo/ > internet-history__;!!KwNVnqRv!HZNzFktVPkM- > k6hZQqAvW5b82_L1M7iaw6UvnsDirrfNV96_RR9lJwBXdg8byGT78LCQXESkvTlTPacEa4cpEeqYTkPxEg$ > From jnc at mercury.lcs.mit.edu Sun Jan 26 13:25:18 2025 From: jnc at mercury.lcs.mit.edu (Noel Chiappa) Date: Sun, 26 Jan 2025 16:25:18 -0500 (EST) Subject: [ih] Correct name for early TCP/IP working group? Message-ID: <20250126212518.AC40818C079@mercury.lcs.mit.edu> > From: Jack Haverty jack at 3kitty.org > At the time, the "ARPANET crowd" was skeptical that the "datagram" > nature of TCP could be made to work. Traditional networks, including > the ARPANET, had elaborate internal mechanisms to provide a "virtual > circuit" service to its users. I was thinkking about this, and wondering if internetworking was a more fundamental advance than the ARPANET (relegating the latter to a 'ground-breaking experiment'), and I had another thought. Internetworking (following in the track of CYCLADES) made much of the fate-sharing aspect - that the data needed to ensure reliable transmission was co-located was the application. One good reason for that (that we knew at the time) was that it made the network itself simpler. But there's another side to that, one that was even more important, and which I'm not sure was obvious to us at the time (1977-79), which is that because it means the intermediate packet switches in the overall internet carry no state about the connections travelling through them, there's no scaling limit. This, to me, has been the single biggest reason why the Internet has been able to grow to the stupendous size it has. I don't think we could have been thinking 'this aspect of lack of state in the internet packet switches neans it will scale indefinitely', because I don't think we had any idea, at that point, about how to do path selection in a global-scale internet - so global-scale internets could not have been in our thinking. Did that infinite scalability turn out to be just a happy accident, a side-effect of good fundamental design (but one whose true complete value wasn't obvious to us at the time), one that moved state out of the internet packet switches? Noel From vint at google.com Sun Jan 26 14:16:28 2025 From: vint at google.com (Vint Cerf) Date: Sun, 26 Jan 2025 17:16:28 -0500 Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: <20250126212518.AC40818C079@mercury.lcs.mit.edu> References: <20250126212518.AC40818C079@mercury.lcs.mit.edu> Message-ID: statelessness was an important design choice and was made consciously so that paths were not critical to successful transport. For example we did not want to have to reassemble along a particular path. Even though we deprecated fragmentation, at the time we thought it was important, we did not want gateway (router) state to be necessary to accomplish reassembly regardless of path. I don't know that we recognized the scalability aspect but we definitely cared a lot about statelessness of the gateways. v On Sun, Jan 26, 2025 at 4:25?PM Noel Chiappa via Internet-history < internet-history at elists.isoc.org> wrote: > > From: Jack Haverty jack at 3kitty.org > > > At the time, the "ARPANET crowd" was skeptical that the "datagram" > > nature of TCP could be made to work. Traditional networks, including > > the ARPANET, had elaborate internal mechanisms to provide a "virtual > > circuit" service to its users. > > I was thinkking about this, and wondering if internetworking was a more > fundamental advance than the ARPANET (relegating the latter to a > 'ground-breaking experiment'), and I had another thought. > > > Internetworking (following in the track of CYCLADES) made much of the > fate-sharing aspect - that the data needed to ensure reliable transmission > was co-located was the application. One good reason for that (that we knew > at > the time) was that it made the network itself simpler. > > But there's another side to that, one that was even more important, and > which > I'm not sure was obvious to us at the time (1977-79), which is that because > it means the intermediate packet switches in the overall internet carry no > state about the connections travelling through them, there's no scaling > limit. This, to me, has been the single biggest reason why the Internet has > been able to grow to the stupendous size it has. > > I don't think we could have been thinking 'this aspect of lack of state in > the internet packet switches neans it will scale indefinitely', because I > don't think we had any idea, at that point, about how to do path selection > in > a global-scale internet - so global-scale internets could not have been in > our thinking. > > Did that infinite scalability turn out to be just a happy accident, a > side-effect of good fundamental design (but one whose true complete value > wasn't obvious to us at the time), one that moved state out of the internet > packet switches? > > Noel > -- > Internet-history mailing list > Internet-history at elists.isoc.org > https://elists.isoc.org/mailman/listinfo/internet-history > -- Please send any postal/overnight deliveries to: Vint Cerf Google, LLC 1900 Reston Metro Plaza, 16th Floor Reston, VA 20190 +1 (571) 213 1346 until further notice From brian.e.carpenter at gmail.com Mon Jan 27 17:47:41 2025 From: brian.e.carpenter at gmail.com (Brian E Carpenter) Date: Tue, 28 Jan 2025 14:47:41 +1300 Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: References: <20250126212518.AC40818C079@mercury.lcs.mit.edu> Message-ID: <0405134f-a905-43d1-991c-49aca2c203c8@gmail.com> Vint, and Noel, I just glanced through Baran's 1964 paper, and it clearly recognized statelessnesss (and a standard packet header) as important for network survivability and adaptive routing. But although he mentions networks of intercontinental size, I didn't spot any discussion of scalability as such. Interestingly, exactly the same applies to Dave Clark's 1988 "Design Philosophy" paper. In RFC 1958, we did note as principle 3.3 that "All designs must scale readily to very many nodes per site and to many millions of sites". I guess that by then (1996) this was too obvious to ignore, and it was written when IPv4 address exhaustion was considered inevitable. Maybe somebody who knows the early literature better than me can find something. But it's almost as if the intrinsic scalability of stateless packet switching was an unnoticed and accidental property. Regards Brian On 27-Jan-25 11:16, Vint Cerf via Internet-history wrote: > statelessness was an important design choice and was made consciously so > that paths were not critical to successful transport. For example we did > not want to have to reassemble along a particular path. Even though we > deprecated fragmentation, at the time we thought it was important, we did > not want gateway (router) state to be necessary to accomplish reassembly > regardless of path. I don't know that we recognized the scalability aspect > but we definitely cared a lot about statelessness of the gateways. > > v > > On Sun, Jan 26, 2025 at 4:25?PM Noel Chiappa via Internet-history < > internet-history at elists.isoc.org> wrote: > >> > From: Jack Haverty jack at 3kitty.org >> >> > At the time, the "ARPANET crowd" was skeptical that the "datagram" >> > nature of TCP could be made to work. Traditional networks, including >> > the ARPANET, had elaborate internal mechanisms to provide a "virtual >> > circuit" service to its users. >> >> I was thinkking about this, and wondering if internetworking was a more >> fundamental advance than the ARPANET (relegating the latter to a >> 'ground-breaking experiment'), and I had another thought. >> >> >> Internetworking (following in the track of CYCLADES) made much of the >> fate-sharing aspect - that the data needed to ensure reliable transmission >> was co-located was the application. One good reason for that (that we knew >> at >> the time) was that it made the network itself simpler. >> >> But there's another side to that, one that was even more important, and >> which >> I'm not sure was obvious to us at the time (1977-79), which is that because >> it means the intermediate packet switches in the overall internet carry no >> state about the connections travelling through them, there's no scaling >> limit. This, to me, has been the single biggest reason why the Internet has >> been able to grow to the stupendous size it has. >> >> I don't think we could have been thinking 'this aspect of lack of state in >> the internet packet switches neans it will scale indefinitely', because I >> don't think we had any idea, at that point, about how to do path selection >> in >> a global-scale internet - so global-scale internets could not have been in >> our thinking. >> >> Did that infinite scalability turn out to be just a happy accident, a >> side-effect of good fundamental design (but one whose true complete value >> wasn't obvious to us at the time), one that moved state out of the internet >> packet switches? >> >> Noel >> -- >> Internet-history mailing list >> Internet-history at elists.isoc.org >> https://elists.isoc.org/mailman/listinfo/internet-history >> > > From sob at sobco.com Tue Jan 28 03:55:35 2025 From: sob at sobco.com (Scott Bradner) Date: Tue, 28 Jan 2025 06:55:35 -0500 Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: <0405134f-a905-43d1-991c-49aca2c203c8@gmail.com> References: <20250126212518.AC40818C079@mercury.lcs.mit.edu> <0405134f-a905-43d1-991c-49aca2c203c8@gmail.com> Message-ID: <954AAF1D-C208-4ABB-BCF5-946F2447E0AE@sobco.com> fwiw - the 1976 Kleinrock & Kamoun paper "Hierarchical Routing for Large Networks, Performance evaluation and optimization" is all about scaling of packet switched networks Scott > On Jan 27, 2025, at 8:47?PM, Brian E Carpenter via Internet-history wrote: > > Vint, and Noel, > > I just glanced through Baran's 1964 paper, and it clearly recognized > statelessnesss (and a standard packet header) as important for network > survivability and adaptive routing. But although he mentions networks > of intercontinental size, I didn't spot any discussion of scalability > as such. > > Interestingly, exactly the same applies to Dave Clark's 1988 "Design > Philosophy" paper. > > In RFC 1958, we did note as principle 3.3 that "All designs must scale > readily to very many nodes per site and to many millions of sites". > I guess that by then (1996) this was too obvious to ignore, and it was > written when IPv4 address exhaustion was considered inevitable. > > Maybe somebody who knows the early literature better than me can find > something. But it's almost as if the intrinsic scalability of stateless > packet switching was an unnoticed and accidental property. > > Regards > Brian > > On 27-Jan-25 11:16, Vint Cerf via Internet-history wrote: >> statelessness was an important design choice and was made consciously so >> that paths were not critical to successful transport. For example we did >> not want to have to reassemble along a particular path. Even though we >> deprecated fragmentation, at the time we thought it was important, we did >> not want gateway (router) state to be necessary to accomplish reassembly >> regardless of path. I don't know that we recognized the scalability aspect >> but we definitely cared a lot about statelessness of the gateways. >> v >> On Sun, Jan 26, 2025 at 4:25?PM Noel Chiappa via Internet-history < >> internet-history at elists.isoc.org> wrote: >>> > From: Jack Haverty jack at 3kitty.org >>> >>> > At the time, the "ARPANET crowd" was skeptical that the "datagram" >>> > nature of TCP could be made to work. Traditional networks, including >>> > the ARPANET, had elaborate internal mechanisms to provide a "virtual >>> > circuit" service to its users. >>> >>> I was thinkking about this, and wondering if internetworking was a more >>> fundamental advance than the ARPANET (relegating the latter to a >>> 'ground-breaking experiment'), and I had another thought. >>> >>> >>> Internetworking (following in the track of CYCLADES) made much of the >>> fate-sharing aspect - that the data needed to ensure reliable transmission >>> was co-located was the application. One good reason for that (that we knew >>> at >>> the time) was that it made the network itself simpler. >>> >>> But there's another side to that, one that was even more important, and >>> which >>> I'm not sure was obvious to us at the time (1977-79), which is that because >>> it means the intermediate packet switches in the overall internet carry no >>> state about the connections travelling through them, there's no scaling >>> limit. This, to me, has been the single biggest reason why the Internet has >>> been able to grow to the stupendous size it has. >>> >>> I don't think we could have been thinking 'this aspect of lack of state in >>> the internet packet switches neans it will scale indefinitely', because I >>> don't think we had any idea, at that point, about how to do path selection >>> in >>> a global-scale internet - so global-scale internets could not have been in >>> our thinking. >>> >>> Did that infinite scalability turn out to be just a happy accident, a >>> side-effect of good fundamental design (but one whose true complete value >>> wasn't obvious to us at the time), one that moved state out of the internet >>> packet switches? >>> >>> Noel >>> -- >>> Internet-history mailing list >>> Internet-history at elists.isoc.org >>> https://elists.isoc.org/mailman/listinfo/internet-history >>> > -- > Internet-history mailing list > Internet-history at elists.isoc.org > https://elists.isoc.org/mailman/listinfo/internet-history From sob at sobco.com Tue Jan 28 04:06:22 2025 From: sob at sobco.com (Scott Bradner) Date: Tue, 28 Jan 2025 07:06:22 -0500 Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: <954AAF1D-C208-4ABB-BCF5-946F2447E0AE@sobco.com> References: <20250126212518.AC40818C079@mercury.lcs.mit.edu> <0405134f-a905-43d1-991c-49aca2c203c8@gmail.com> <954AAF1D-C208-4ABB-BCF5-946F2447E0AE@sobco.com> Message-ID: ps - you can find the paper on Kleinrock's website https://www.lk.cs.ucla.edu/data/files/Kamoun/Data%20Communications%20through%20Large%20Packet-Switching%20Networks.pdf > On Jan 28, 2025, at 6:55?AM, Scott Bradner via Internet-history wrote: > > fwiw - the 1976 Kleinrock & Kamoun paper "Hierarchical Routing for Large Networks, Performance evaluation and optimization" is all about scaling of packet switched networks > > Scott > > > >> On Jan 27, 2025, at 8:47?PM, Brian E Carpenter via Internet-history wrote: >> >> Vint, and Noel, >> >> I just glanced through Baran's 1964 paper, and it clearly recognized >> statelessnesss (and a standard packet header) as important for network >> survivability and adaptive routing. But although he mentions networks >> of intercontinental size, I didn't spot any discussion of scalability >> as such. >> >> Interestingly, exactly the same applies to Dave Clark's 1988 "Design >> Philosophy" paper. >> >> In RFC 1958, we did note as principle 3.3 that "All designs must scale >> readily to very many nodes per site and to many millions of sites". >> I guess that by then (1996) this was too obvious to ignore, and it was >> written when IPv4 address exhaustion was considered inevitable. >> >> Maybe somebody who knows the early literature better than me can find >> something. But it's almost as if the intrinsic scalability of stateless >> packet switching was an unnoticed and accidental property. >> >> Regards >> Brian >> >> On 27-Jan-25 11:16, Vint Cerf via Internet-history wrote: >>> statelessness was an important design choice and was made consciously so >>> that paths were not critical to successful transport. For example we did >>> not want to have to reassemble along a particular path. Even though we >>> deprecated fragmentation, at the time we thought it was important, we did >>> not want gateway (router) state to be necessary to accomplish reassembly >>> regardless of path. I don't know that we recognized the scalability aspect >>> but we definitely cared a lot about statelessness of the gateways. >>> v >>> On Sun, Jan 26, 2025 at 4:25?PM Noel Chiappa via Internet-history < >>> internet-history at elists.isoc.org> wrote: >>>>> From: Jack Haverty jack at 3kitty.org >>>> >>>>> At the time, the "ARPANET crowd" was skeptical that the "datagram" >>>>> nature of TCP could be made to work. Traditional networks, including >>>>> the ARPANET, had elaborate internal mechanisms to provide a "virtual >>>>> circuit" service to its users. >>>> >>>> I was thinkking about this, and wondering if internetworking was a more >>>> fundamental advance than the ARPANET (relegating the latter to a >>>> 'ground-breaking experiment'), and I had another thought. >>>> >>>> >>>> Internetworking (following in the track of CYCLADES) made much of the >>>> fate-sharing aspect - that the data needed to ensure reliable transmission >>>> was co-located was the application. One good reason for that (that we knew >>>> at >>>> the time) was that it made the network itself simpler. >>>> >>>> But there's another side to that, one that was even more important, and >>>> which >>>> I'm not sure was obvious to us at the time (1977-79), which is that because >>>> it means the intermediate packet switches in the overall internet carry no >>>> state about the connections travelling through them, there's no scaling >>>> limit. This, to me, has been the single biggest reason why the Internet has >>>> been able to grow to the stupendous size it has. >>>> >>>> I don't think we could have been thinking 'this aspect of lack of state in >>>> the internet packet switches neans it will scale indefinitely', because I >>>> don't think we had any idea, at that point, about how to do path selection >>>> in >>>> a global-scale internet - so global-scale internets could not have been in >>>> our thinking. >>>> >>>> Did that infinite scalability turn out to be just a happy accident, a >>>> side-effect of good fundamental design (but one whose true complete value >>>> wasn't obvious to us at the time), one that moved state out of the internet >>>> packet switches? >>>> >>>> Noel >>>> -- >>>> Internet-history mailing list >>>> Internet-history at elists.isoc.org >>>> https://elists.isoc.org/mailman/listinfo/internet-history >>>> >> -- >> Internet-history mailing list >> Internet-history at elists.isoc.org >> https://elists.isoc.org/mailman/listinfo/internet-history > > -- > Internet-history mailing list > Internet-history at elists.isoc.org > https://elists.isoc.org/mailman/listinfo/internet-history From jeanjour at comcast.net Tue Jan 28 04:50:23 2025 From: jeanjour at comcast.net (John Day) Date: Tue, 28 Jan 2025 07:50:23 -0500 Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: <0405134f-a905-43d1-991c-49aca2c203c8@gmail.com> References: <20250126212518.AC40818C079@mercury.lcs.mit.edu> <0405134f-a905-43d1-991c-49aca2c203c8@gmail.com> Message-ID: <52E30F78-6795-4C7D-A7DA-6EA3FC71F964@comcast.net> Brian, I agree with you. In the Baran reports, he describes something that sounds like a datagram. However, he never explores it much other than to define hot-potato routing. His focus is very centered on survivability and resilience, which makes sense it was research for the DoD. There is also the consideration that so far as I have been able to determine, all of the projects Baran was involved in afterwards were virtual-circuit, as were Roberts. OTOH, NPL didn?t do military research, so I guessed that their impetus for exploring packet switching had to be different and indeed it was. We have found a memo Davies wrote (and a similar one by Derek Barber) that Davies had attended the IFIP Congress in the US in 1965 and heard lots of papers on timesharing and the time slicing scheduling that timesharing used. The advantage being that while batch systems did FCFS and short jobs got stuck behind long ones, timesharut ing interleaved jobs and while short jobs were still delayed but their *completion* times were shorter. Davies told Derek that that was what they should do with communications and they did. Their impetus for packet switching was very different. (But there were two problems. ;-) 1) Donald got promoted and less time for research, ;-) and 2) the GPO got involved and the government directed NPL to concentrate on ?practical? projects, so they had to move to virtual-circuits. Scantlebury told Roberts about packet switching at the Gatlinburg conference and convinced him to use it. When Roberts returned to DC, he found he had Baran?s reports in a stack of documents but hadn?t read it yet. Based on the NPL experience Roger also convinced Roberts not to use 2.4Kbps lines but 50Kbps, which was a large part of the ARPANET success. (Slower speed would have worked but been so slow people would have said it wasn?t practical, etc.) There is much more to be said about all of this. But that seems to be the core of it. I find it very interesting how minor events have major effects. Take care, John Day > On Jan 27, 2025, at 20:47, Brian E Carpenter via Internet-history wrote: > > Vint, and Noel, > > I just glanced through Baran's 1964 paper, and it clearly recognized > statelessnesss (and a standard packet header) as important for network > survivability and adaptive routing. But although he mentions networks > of intercontinental size, I didn't spot any discussion of scalability > as such. > > Interestingly, exactly the same applies to Dave Clark's 1988 "Design > Philosophy" paper. > > In RFC 1958, we did note as principle 3.3 that "All designs must scale > readily to very many nodes per site and to many millions of sites". > I guess that by then (1996) this was too obvious to ignore, and it was > written when IPv4 address exhaustion was considered inevitable. > > Maybe somebody who knows the early literature better than me can find > something. But it's almost as if the intrinsic scalability of stateless > packet switching was an unnoticed and accidental property. > > Regards > Brian > > On 27-Jan-25 11:16, Vint Cerf via Internet-history wrote: >> statelessness was an important design choice and was made consciously so >> that paths were not critical to successful transport. For example we did >> not want to have to reassemble along a particular path. Even though we >> deprecated fragmentation, at the time we thought it was important, we did >> not want gateway (router) state to be necessary to accomplish reassembly >> regardless of path. I don't know that we recognized the scalability aspect >> but we definitely cared a lot about statelessness of the gateways. >> v >> On Sun, Jan 26, 2025 at 4:25?PM Noel Chiappa via Internet-history < >> internet-history at elists.isoc.org> wrote: >>> > From: Jack Haverty jack at 3kitty.org >>> >>> > At the time, the "ARPANET crowd" was skeptical that the "datagram" >>> > nature of TCP could be made to work. Traditional networks, including >>> > the ARPANET, had elaborate internal mechanisms to provide a "virtual >>> > circuit" service to its users. >>> >>> I was thinkking about this, and wondering if internetworking was a more >>> fundamental advance than the ARPANET (relegating the latter to a >>> 'ground-breaking experiment'), and I had another thought. >>> >>> >>> Internetworking (following in the track of CYCLADES) made much of the >>> fate-sharing aspect - that the data needed to ensure reliable transmission >>> was co-located was the application. One good reason for that (that we knew >>> at >>> the time) was that it made the network itself simpler. >>> >>> But there's another side to that, one that was even more important, and >>> which >>> I'm not sure was obvious to us at the time (1977-79), which is that because >>> it means the intermediate packet switches in the overall internet carry no >>> state about the connections travelling through them, there's no scaling >>> limit. This, to me, has been the single biggest reason why the Internet has >>> been able to grow to the stupendous size it has. >>> >>> I don't think we could have been thinking 'this aspect of lack of state in >>> the internet packet switches neans it will scale indefinitely', because I >>> don't think we had any idea, at that point, about how to do path selection >>> in >>> a global-scale internet - so global-scale internets could not have been in >>> our thinking. >>> >>> Did that infinite scalability turn out to be just a happy accident, a >>> side-effect of good fundamental design (but one whose true complete value >>> wasn't obvious to us at the time), one that moved state out of the internet >>> packet switches? >>> >>> Noel >>> -- >>> Internet-history mailing list >>> Internet-history at elists.isoc.org >>> https://elists.isoc.org/mailman/listinfo/internet-history >>> > -- > Internet-history mailing list > Internet-history at elists.isoc.org > https://elists.isoc.org/mailman/listinfo/internet-history From vint at google.com Tue Jan 28 05:49:49 2025 From: vint at google.com (Vint Cerf) Date: Tue, 28 Jan 2025 08:49:49 -0500 Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: <52E30F78-6795-4C7D-A7DA-6EA3FC71F964@comcast.net> References: <20250126212518.AC40818C079@mercury.lcs.mit.edu> <0405134f-a905-43d1-991c-49aca2c203c8@gmail.com> <52E30F78-6795-4C7D-A7DA-6EA3FC71F964@comcast.net> Message-ID: roberts would also have been aware of Len Kleinrock's queueing theory analysis of message switching which, mathematically, was not very different from packet switching. They were both at MIT if memory serves and got their Ph.D's the same year, 1963. v On Tue, Jan 28, 2025 at 7:50?AM John Day wrote: > Brian, > I agree with you. In the Baran reports, he describes something that sounds > like a datagram. However, he never explores it much other than to define > hot-potato routing. His focus is very centered on survivability and > resilience, which makes sense it was research for the DoD. There is also > the consideration that so far as I have been able to determine, all of the > projects Baran was involved in afterwards were virtual-circuit, as were > Roberts. > > OTOH, NPL didn?t do military research, so I guessed that their impetus for > exploring packet switching had to be different and indeed it was. We have > found a memo Davies wrote (and a similar one by Derek Barber) that Davies > had attended the IFIP Congress in the US in 1965 and heard lots of papers > on timesharing and the time slicing scheduling that timesharing used. The > advantage being that while batch systems did FCFS and short jobs got stuck > behind long ones, timesharut ing interleaved jobs and while short jobs were > still delayed but their *completion* times were shorter. Davies told Derek > that that was what they should do with communications and they did. Their > impetus for packet switching was very different. (But there were two > problems. ;-) 1) Donald got promoted and less time for research, ;-) and 2) > the GPO got involved and the government directed NPL to concentrate on > ?practical? projects, so they had to move to virtual-circuits. > > Scantlebury told Roberts about packet switching at the Gatlinburg > conference and convinced him to use it. When Roberts returned to DC, he > found he had Baran?s reports in a stack of documents but hadn?t read it > yet. Based on the NPL experience Roger also convinced Roberts not to use > 2.4Kbps lines but 50Kbps, which was a large part of the ARPANET success. > (Slower speed would have worked but been so slow people would have said it > wasn?t practical, etc.) > > There is much more to be said about all of this. But that seems to be the > core of it. I find it very interesting how minor events have major effects. > > Take care, > John Day > > > On Jan 27, 2025, at 20:47, Brian E Carpenter via Internet-history < > internet-history at elists.isoc.org> wrote: > > > > Vint, and Noel, > > > > I just glanced through Baran's 1964 paper, and it clearly recognized > > statelessnesss (and a standard packet header) as important for network > > survivability and adaptive routing. But although he mentions networks > > of intercontinental size, I didn't spot any discussion of scalability > > as such. > > > > Interestingly, exactly the same applies to Dave Clark's 1988 "Design > > Philosophy" paper. > > > > In RFC 1958, we did note as principle 3.3 that "All designs must scale > > readily to very many nodes per site and to many millions of sites". > > I guess that by then (1996) this was too obvious to ignore, and it was > > written when IPv4 address exhaustion was considered inevitable. > > > > Maybe somebody who knows the early literature better than me can find > > something. But it's almost as if the intrinsic scalability of stateless > > packet switching was an unnoticed and accidental property. > > > > Regards > > Brian > > > > On 27-Jan-25 11:16, Vint Cerf via Internet-history wrote: > >> statelessness was an important design choice and was made consciously so > >> that paths were not critical to successful transport. For example we did > >> not want to have to reassemble along a particular path. Even though we > >> deprecated fragmentation, at the time we thought it was important, we > did > >> not want gateway (router) state to be necessary to accomplish reassembly > >> regardless of path. I don't know that we recognized the scalability > aspect > >> but we definitely cared a lot about statelessness of the gateways. > >> v > >> On Sun, Jan 26, 2025 at 4:25?PM Noel Chiappa via Internet-history < > >> internet-history at elists.isoc.org> wrote: > >>> > From: Jack Haverty jack at 3kitty.org > >>> > >>> > At the time, the "ARPANET crowd" was skeptical that the > "datagram" > >>> > nature of TCP could be made to work. Traditional networks, > including > >>> > the ARPANET, had elaborate internal mechanisms to provide a > "virtual > >>> > circuit" service to its users. > >>> > >>> I was thinkking about this, and wondering if internetworking was a more > >>> fundamental advance than the ARPANET (relegating the latter to a > >>> 'ground-breaking experiment'), and I had another thought. > >>> > >>> > >>> Internetworking (following in the track of CYCLADES) made much of the > >>> fate-sharing aspect - that the data needed to ensure reliable > transmission > >>> was co-located was the application. One good reason for that (that we > knew > >>> at > >>> the time) was that it made the network itself simpler. > >>> > >>> But there's another side to that, one that was even more important, and > >>> which > >>> I'm not sure was obvious to us at the time (1977-79), which is that > because > >>> it means the intermediate packet switches in the overall internet > carry no > >>> state about the connections travelling through them, there's no scaling > >>> limit. This, to me, has been the single biggest reason why the > Internet has > >>> been able to grow to the stupendous size it has. > >>> > >>> I don't think we could have been thinking 'this aspect of lack of > state in > >>> the internet packet switches neans it will scale indefinitely', > because I > >>> don't think we had any idea, at that point, about how to do path > selection > >>> in > >>> a global-scale internet - so global-scale internets could not have > been in > >>> our thinking. > >>> > >>> Did that infinite scalability turn out to be just a happy accident, a > >>> side-effect of good fundamental design (but one whose true complete > value > >>> wasn't obvious to us at the time), one that moved state out of the > internet > >>> packet switches? > >>> > >>> Noel > >>> -- > >>> Internet-history mailing list > >>> Internet-history at elists.isoc.org > >>> https://elists.isoc.org/mailman/listinfo/internet-history > >>> > > -- > > Internet-history mailing list > > Internet-history at elists.isoc.org > > https://elists.isoc.org/mailman/listinfo/internet-history > > -- Please send any postal/overnight deliveries to: Vint Cerf Google, LLC 1900 Reston Metro Plaza, 16th Floor Reston, VA 20190 +1 (571) 213 1346 until further notice From steve at shinkuro.com Tue Jan 28 05:56:02 2025 From: steve at shinkuro.com (Steve Crocker) Date: Tue, 28 Jan 2025 08:56:02 -0500 Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: References: <20250126212518.AC40818C079@mercury.lcs.mit.edu> <0405134f-a905-43d1-991c-49aca2c203c8@gmail.com> <52E30F78-6795-4C7D-A7DA-6EA3FC71F964@comcast.net> Message-ID: And they were office mates. On Tue, Jan 28, 2025 at 8:50?AM Vint Cerf via Internet-history < internet-history at elists.isoc.org> wrote: > roberts would also have been aware of Len Kleinrock's queueing theory > analysis of message switching which, mathematically, was not very different > from packet switching. They were both at MIT if memory serves and got their > Ph.D's the same year, 1963. > > v > > > On Tue, Jan 28, 2025 at 7:50?AM John Day wrote: > > > Brian, > > I agree with you. In the Baran reports, he describes something that > sounds > > like a datagram. However, he never explores it much other than to define > > hot-potato routing. His focus is very centered on survivability and > > resilience, which makes sense it was research for the DoD. There is also > > the consideration that so far as I have been able to determine, all of > the > > projects Baran was involved in afterwards were virtual-circuit, as were > > Roberts. > > > > OTOH, NPL didn?t do military research, so I guessed that their impetus > for > > exploring packet switching had to be different and indeed it was. We have > > found a memo Davies wrote (and a similar one by Derek Barber) that Davies > > had attended the IFIP Congress in the US in 1965 and heard lots of papers > > on timesharing and the time slicing scheduling that timesharing used. The > > advantage being that while batch systems did FCFS and short jobs got > stuck > > behind long ones, timesharut ing interleaved jobs and while short jobs > were > > still delayed but their *completion* times were shorter. Davies told > Derek > > that that was what they should do with communications and they did. Their > > impetus for packet switching was very different. (But there were two > > problems. ;-) 1) Donald got promoted and less time for research, ;-) and > 2) > > the GPO got involved and the government directed NPL to concentrate on > > ?practical? projects, so they had to move to virtual-circuits. > > > > Scantlebury told Roberts about packet switching at the Gatlinburg > > conference and convinced him to use it. When Roberts returned to DC, he > > found he had Baran?s reports in a stack of documents but hadn?t read it > > yet. Based on the NPL experience Roger also convinced Roberts not to use > > 2.4Kbps lines but 50Kbps, which was a large part of the ARPANET success. > > (Slower speed would have worked but been so slow people would have said > it > > wasn?t practical, etc.) > > > > There is much more to be said about all of this. But that seems to be the > > core of it. I find it very interesting how minor events have major > effects. > > > > Take care, > > John Day > > > > > On Jan 27, 2025, at 20:47, Brian E Carpenter via Internet-history < > > internet-history at elists.isoc.org> wrote: > > > > > > Vint, and Noel, > > > > > > I just glanced through Baran's 1964 paper, and it clearly recognized > > > statelessnesss (and a standard packet header) as important for network > > > survivability and adaptive routing. But although he mentions networks > > > of intercontinental size, I didn't spot any discussion of scalability > > > as such. > > > > > > Interestingly, exactly the same applies to Dave Clark's 1988 "Design > > > Philosophy" paper. > > > > > > In RFC 1958, we did note as principle 3.3 that "All designs must scale > > > readily to very many nodes per site and to many millions of sites". > > > I guess that by then (1996) this was too obvious to ignore, and it was > > > written when IPv4 address exhaustion was considered inevitable. > > > > > > Maybe somebody who knows the early literature better than me can find > > > something. But it's almost as if the intrinsic scalability of stateless > > > packet switching was an unnoticed and accidental property. > > > > > > Regards > > > Brian > > > > > > On 27-Jan-25 11:16, Vint Cerf via Internet-history wrote: > > >> statelessness was an important design choice and was made consciously > so > > >> that paths were not critical to successful transport. For example we > did > > >> not want to have to reassemble along a particular path. Even though we > > >> deprecated fragmentation, at the time we thought it was important, we > > did > > >> not want gateway (router) state to be necessary to accomplish > reassembly > > >> regardless of path. I don't know that we recognized the scalability > > aspect > > >> but we definitely cared a lot about statelessness of the gateways. > > >> v > > >> On Sun, Jan 26, 2025 at 4:25?PM Noel Chiappa via Internet-history < > > >> internet-history at elists.isoc.org> wrote: > > >>> > From: Jack Haverty jack at 3kitty.org > > >>> > > >>> > At the time, the "ARPANET crowd" was skeptical that the > > "datagram" > > >>> > nature of TCP could be made to work. Traditional networks, > > including > > >>> > the ARPANET, had elaborate internal mechanisms to provide a > > "virtual > > >>> > circuit" service to its users. > > >>> > > >>> I was thinkking about this, and wondering if internetworking was a > more > > >>> fundamental advance than the ARPANET (relegating the latter to a > > >>> 'ground-breaking experiment'), and I had another thought. > > >>> > > >>> > > >>> Internetworking (following in the track of CYCLADES) made much of the > > >>> fate-sharing aspect - that the data needed to ensure reliable > > transmission > > >>> was co-located was the application. One good reason for that (that we > > knew > > >>> at > > >>> the time) was that it made the network itself simpler. > > >>> > > >>> But there's another side to that, one that was even more important, > and > > >>> which > > >>> I'm not sure was obvious to us at the time (1977-79), which is that > > because > > >>> it means the intermediate packet switches in the overall internet > > carry no > > >>> state about the connections travelling through them, there's no > scaling > > >>> limit. This, to me, has been the single biggest reason why the > > Internet has > > >>> been able to grow to the stupendous size it has. > > >>> > > >>> I don't think we could have been thinking 'this aspect of lack of > > state in > > >>> the internet packet switches neans it will scale indefinitely', > > because I > > >>> don't think we had any idea, at that point, about how to do path > > selection > > >>> in > > >>> a global-scale internet - so global-scale internets could not have > > been in > > >>> our thinking. > > >>> > > >>> Did that infinite scalability turn out to be just a happy accident, a > > >>> side-effect of good fundamental design (but one whose true complete > > value > > >>> wasn't obvious to us at the time), one that moved state out of the > > internet > > >>> packet switches? > > >>> > > >>> Noel > > >>> -- > > >>> Internet-history mailing list > > >>> Internet-history at elists.isoc.org > > >>> https://elists.isoc.org/mailman/listinfo/internet-history > > >>> > > > -- > > > Internet-history mailing list > > > Internet-history at elists.isoc.org > > > https://elists.isoc.org/mailman/listinfo/internet-history > > > > > > -- > Please send any postal/overnight deliveries to: > Vint Cerf > Google, LLC > 1900 Reston Metro Plaza, 16th Floor > Reston, VA 20190 > +1 (571) 213 1346 > > > until further notice > -- > Internet-history mailing list > Internet-history at elists.isoc.org > https://elists.isoc.org/mailman/listinfo/internet-history > -- Sent by a Verified sender From jeanjour at comcast.net Tue Jan 28 07:53:20 2025 From: jeanjour at comcast.net (John Day) Date: Tue, 28 Jan 2025 10:53:20 -0500 Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: References: <20250126212518.AC40818C079@mercury.lcs.mit.edu> <0405134f-a905-43d1-991c-49aca2c203c8@gmail.com> <52E30F78-6795-4C7D-A7DA-6EA3FC71F964@comcast.net> <34D32C24-C418-4D8A-B96E-66441F93107F@comcast.net> Message-ID: Right, and at some point (possibly then) length was less important than processing time. Packets were pretty much all the same size, although message length varied. But that was probably much less important. John > On Jan 28, 2025, at 09:04, Vint Cerf wrote: > > basically it was networks of queues with variable length messages - averages and moments as well as optimizations. We used the same mathematics to predict Arpanet behavior. I collected data via the Sigma-7 and other students modeled performance (mostly delay and throughput for message completion). The models were adapted to the Arpanet message/packet practice. > > v > > > On Tue, Jan 28, 2025 at 9:00?AM John Day > wrote: >> Correct. >> >> But I would think there would have been a big difference in the results, but I guess it would depend on what was being analyzed. If it was just the queue behavior, then there wouldn?t be a big difference. The difference between Message switching with long and short messages, vs packet switching with more homogenous traffic of short messages. >> >> But there is a big difference is 'completion time' between FCFS (message-switching) and round-robin (packet switching). In this case, completion time is ?end-to-end? delay. ;-) >> >> Just rambling off the top of my head too early in the morning. ;-) >> >> Take care, >> John >> >>> On Jan 28, 2025, at 08:49, Vint Cerf > wrote: >>> >>> roberts would also have been aware of Len Kleinrock's queueing theory analysis of message switching which, mathematically, was not very different from packet switching. They were both at MIT if memory serves and got their Ph.D's the same year, 1963. >>> >>> v >>> >>> >>> On Tue, Jan 28, 2025 at 7:50?AM John Day > wrote: >>>> Brian, >>>> I agree with you. In the Baran reports, he describes something that sounds like a datagram. However, he never explores it much other than to define hot-potato routing. His focus is very centered on survivability and resilience, which makes sense it was research for the DoD. There is also the consideration that so far as I have been able to determine, all of the projects Baran was involved in afterwards were virtual-circuit, as were Roberts. >>>> >>>> OTOH, NPL didn?t do military research, so I guessed that their impetus for exploring packet switching had to be different and indeed it was. We have found a memo Davies wrote (and a similar one by Derek Barber) that Davies had attended the IFIP Congress in the US in 1965 and heard lots of papers on timesharing and the time slicing scheduling that timesharing used. The advantage being that while batch systems did FCFS and short jobs got stuck behind long ones, timesharut ing interleaved jobs and while short jobs were still delayed but their *completion* times were shorter. Davies told Derek that that was what they should do with communications and they did. Their impetus for packet switching was very different. (But there were two problems. ;-) 1) Donald got promoted and less time for research, ;-) and 2) the GPO got involved and the government directed NPL to concentrate on ?practical? projects, so they had to move to virtual-circuits. >>>> >>>> Scantlebury told Roberts about packet switching at the Gatlinburg conference and convinced him to use it. When Roberts returned to DC, he found he had Baran?s reports in a stack of documents but hadn?t read it yet. Based on the NPL experience Roger also convinced Roberts not to use 2.4Kbps lines but 50Kbps, which was a large part of the ARPANET success. (Slower speed would have worked but been so slow people would have said it wasn?t practical, etc.) >>>> >>>> There is much more to be said about all of this. But that seems to be the core of it. I find it very interesting how minor events have major effects. >>>> >>>> Take care, >>>> John Day >>>> >>>> > On Jan 27, 2025, at 20:47, Brian E Carpenter via Internet-history > wrote: >>>> > >>>> > Vint, and Noel, >>>> > >>>> > I just glanced through Baran's 1964 paper, and it clearly recognized >>>> > statelessnesss (and a standard packet header) as important for network >>>> > survivability and adaptive routing. But although he mentions networks >>>> > of intercontinental size, I didn't spot any discussion of scalability >>>> > as such. >>>> > >>>> > Interestingly, exactly the same applies to Dave Clark's 1988 "Design >>>> > Philosophy" paper. >>>> > >>>> > In RFC 1958, we did note as principle 3.3 that "All designs must scale >>>> > readily to very many nodes per site and to many millions of sites". >>>> > I guess that by then (1996) this was too obvious to ignore, and it was >>>> > written when IPv4 address exhaustion was considered inevitable. >>>> > >>>> > Maybe somebody who knows the early literature better than me can find >>>> > something. But it's almost as if the intrinsic scalability of stateless >>>> > packet switching was an unnoticed and accidental property. >>>> > >>>> > Regards >>>> > Brian >>>> > >>>> > On 27-Jan-25 11:16, Vint Cerf via Internet-history wrote: >>>> >> statelessness was an important design choice and was made consciously so >>>> >> that paths were not critical to successful transport. For example we did >>>> >> not want to have to reassemble along a particular path. Even though we >>>> >> deprecated fragmentation, at the time we thought it was important, we did >>>> >> not want gateway (router) state to be necessary to accomplish reassembly >>>> >> regardless of path. I don't know that we recognized the scalability aspect >>>> >> but we definitely cared a lot about statelessness of the gateways. >>>> >> v >>>> >> On Sun, Jan 26, 2025 at 4:25?PM Noel Chiappa via Internet-history < >>>> >> internet-history at elists.isoc.org > wrote: >>>> >>> > From: Jack Haverty jack at 3kitty.org >>>> >>> >>>> >>> > At the time, the "ARPANET crowd" was skeptical that the "datagram" >>>> >>> > nature of TCP could be made to work. Traditional networks, including >>>> >>> > the ARPANET, had elaborate internal mechanisms to provide a "virtual >>>> >>> > circuit" service to its users. >>>> >>> >>>> >>> I was thinkking about this, and wondering if internetworking was a more >>>> >>> fundamental advance than the ARPANET (relegating the latter to a >>>> >>> 'ground-breaking experiment'), and I had another thought. >>>> >>> >>>> >>> >>>> >>> Internetworking (following in the track of CYCLADES) made much of the >>>> >>> fate-sharing aspect - that the data needed to ensure reliable transmission >>>> >>> was co-located was the application. One good reason for that (that we knew >>>> >>> at >>>> >>> the time) was that it made the network itself simpler. >>>> >>> >>>> >>> But there's another side to that, one that was even more important, and >>>> >>> which >>>> >>> I'm not sure was obvious to us at the time (1977-79), which is that because >>>> >>> it means the intermediate packet switches in the overall internet carry no >>>> >>> state about the connections travelling through them, there's no scaling >>>> >>> limit. This, to me, has been the single biggest reason why the Internet has >>>> >>> been able to grow to the stupendous size it has. >>>> >>> >>>> >>> I don't think we could have been thinking 'this aspect of lack of state in >>>> >>> the internet packet switches neans it will scale indefinitely', because I >>>> >>> don't think we had any idea, at that point, about how to do path selection >>>> >>> in >>>> >>> a global-scale internet - so global-scale internets could not have been in >>>> >>> our thinking. >>>> >>> >>>> >>> Did that infinite scalability turn out to be just a happy accident, a >>>> >>> side-effect of good fundamental design (but one whose true complete value >>>> >>> wasn't obvious to us at the time), one that moved state out of the internet >>>> >>> packet switches? >>>> >>> >>>> >>> Noel >>>> >>> -- >>>> >>> Internet-history mailing list >>>> >>> Internet-history at elists.isoc.org >>>> >>> https://elists.isoc.org/mailman/listinfo/internet-history >>>> >>> >>>> > -- >>>> > Internet-history mailing list >>>> > Internet-history at elists.isoc.org >>>> > https://elists.isoc.org/mailman/listinfo/internet-history >>>> >>> >>> >>> >>> -- >>> Please send any postal/overnight deliveries to: >>> Vint Cerf >>> Google, LLC >>> 1900 Reston Metro Plaza, 16th Floor >>> Reston, VA 20190 >>> +1 (571) 213 1346 >>> >>> >>> until further notice >>> >>> >>> >> > > > > -- > Please send any postal/overnight deliveries to: > Vint Cerf > Google, LLC > 1900 Reston Metro Plaza, 16th Floor > Reston, VA 20190 > +1 (571) 213 1346 > > > until further notice > > > From jack at 3kitty.org Tue Jan 28 08:41:41 2025 From: jack at 3kitty.org (Jack Haverty) Date: Tue, 28 Jan 2025 08:41:41 -0800 Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: References: <20250126212518.AC40818C079@mercury.lcs.mit.edu> <0405134f-a905-43d1-991c-49aca2c203c8@gmail.com> <52E30F78-6795-4C7D-A7DA-6EA3FC71F964@comcast.net> <34D32C24-C418-4D8A-B96E-66441F93107F@comcast.net> Message-ID: [I sent this yesterday, but it hasn't come back to me yet from the elists.isoc.org server.?? So I'm not sure what happened.?? Apologies if you already got this.? /Jack] The design goal of statelessness certainly made the switching fabric of routers/gateways simpler and more easily scaled. But that choice didn't remove the complexity;? it simply moved it from the lowest level switches into the computers attached to them. I just viewed the ICCC'72 video (Thanks, Lars!), which I think I never had time to see in 1972.?? One of the points it makes is that the introduction of another computer (the IMP) as a front end for one or several "user" computers had a lot of advantages.? It simplified the tasks for each computer's programmer, minimized the workload on those expensive machines, and made it much easier to evolve the various mechanisms that provided error control, flow control, et al, by putting all that complexity in the IMP under single management. When I joined BBN in 1977, I didn't know much about how the IMPs worked.? But I was surrounded by ARPANET people, who by then had been refining the IMP's internal mechanisms for 8 years and had made many changes as real-world operation surfaced issues.? I learned a lot by osmosis. My assignment was to implement TCP as part of the "Internet Experiment".? I heard plenty of reasons why it was a bad idea to rely on "datagrams", and intense reluctance to allow TCPs to use the IMP's datagram mode, for fear it would result in a crash of the entire network. Experiments are classically done to test some theory/hypothesis.?? I don't recall ever seeing an explanation of the "Theory of the Internet".? But I've always assumed it was something like "It is possible to construct a global data communications network using simple switching computers with no maintenance of state information." The Internet Experiment was testing that hypothesis.?? Today's network seems to say "Yes!".? It is possible to build such a network. But there seem to be some undesired consequences. For example, the IMP code went through multiple releases over the lifetime of the Arpanet.?? All were carefully orchestrated to convert the entire network from the "old" to "new" mechanisms, only rarely with the need for any of the "host" computers to change their code. In contrast, TCP rapidly went through multiple releases, including the major changes from TCP2 to TCP/IP4, over a period of just a few years, while it was still controlled by the research community. However, as the technology migrated into the commercial environment and grew rapidly, such evolution seems to have become much more difficult.? TCP/IP6 was defined more than 25 years ago, but TCP/IP4 is still used, which I have to believe makes the underlying mechanisms more complex inside the switching machinery. Similarly, changes have been made to TCP and related mechanisms, which are all implemented in the "host" computers and are responsible for error control, retransmissions, algorithms and protocols for interactions, and general monitoring and management of such technology.? But few end-users have the interest or capability to watch how their computer systems are working. The IETF has released thousands of Standards, RFCs, and related documents, but it's almost impossible to tell which if them are actually implemented, correctly, in a product I might buy.? ?In my own LAN at home, I have more than 60 devices.?? I have no idea what Standards they implement, how efficiently they use the Internet, or if I should replace them with some other device.? My ISP is not helpful.? They just carry datagrams. In addition, there were a number of unsolved research questions concerning the switching fabric.? For example, can the network provide multiple types of service, such as one for bulk transfer of data and another for interactive needs?? At the time, we thought such a capability was necessary. Now 40 years later, perhaps such research topics are no longer relevant, given today's technology such as fiber.? Or maybe not. Jack Haverty On 1/28/25 07:53, John Day via Internet-history wrote: > Right, and at some point (possibly then) length was less important than processing time. > > Packets were pretty much all the same size, although message length varied. But that was probably much less important. > > John > >> On Jan 28, 2025, at 09:04, Vint Cerf wrote: >> >> basically it was networks of queues with variable length messages - averages and moments as well as optimizations. We used the same mathematics to predict Arpanet behavior. I collected data via the Sigma-7 and other students modeled performance (mostly delay and throughput for message completion). The models were adapted to the Arpanet message/packet practice. >> >> v >> >> >> On Tue, Jan 28, 2025 at 9:00?AM John Day > wrote: >>> Correct. >>> >>> But I would think there would have been a big difference in the results, but I guess it would depend on what was being analyzed. If it was just the queue behavior, then there wouldn?t be a big difference. The difference between Message switching with long and short messages, vs packet switching with more homogenous traffic of short messages. >>> >>> But there is a big difference is 'completion time' between FCFS (message-switching) and round-robin (packet switching). In this case, completion time is ?end-to-end? delay. ;-) >>> >>> Just rambling off the top of my head too early in the morning. ;-) >>> >>> Take care, >>> John >>> >>>> On Jan 28, 2025, at 08:49, Vint Cerf > wrote: >>>> >>>> roberts would also have been aware of Len Kleinrock's queueing theory analysis of message switching which, mathematically, was not very different from packet switching. They were both at MIT if memory serves and got their Ph.D's the same year, 1963. >>>> >>>> v >>>> >>>> >>>> On Tue, Jan 28, 2025 at 7:50?AM John Day > wrote: >>>>> Brian, >>>>> I agree with you. In the Baran reports, he describes something that sounds like a datagram. However, he never explores it much other than to define hot-potato routing. His focus is very centered on survivability and resilience, which makes sense it was research for the DoD. There is also the consideration that so far as I have been able to determine, all of the projects Baran was involved in afterwards were virtual-circuit, as were Roberts. >>>>> >>>>> OTOH, NPL didn?t do military research, so I guessed that their impetus for exploring packet switching had to be different and indeed it was. We have found a memo Davies wrote (and a similar one by Derek Barber) that Davies had attended the IFIP Congress in the US in 1965 and heard lots of papers on timesharing and the time slicing scheduling that timesharing used. The advantage being that while batch systems did FCFS and short jobs got stuck behind long ones, timesharut ing interleaved jobs and while short jobs were still delayed but their *completion* times were shorter. Davies told Derek that that was what they should do with communications and they did. Their impetus for packet switching was very different. (But there were two problems. ;-) 1) Donald got promoted and less time for research, ;-) and 2) the GPO got involved and the government directed NPL to concentrate on ?practical? projects, so they had to move to virtual-circuits. >>>>> >>>>> Scantlebury told Roberts about packet switching at the Gatlinburg conference and convinced him to use it. When Roberts returned to DC, he found he had Baran?s reports in a stack of documents but hadn?t read it yet. Based on the NPL experience Roger also convinced Roberts not to use 2.4Kbps lines but 50Kbps, which was a large part of the ARPANET success. (Slower speed would have worked but been so slow people would have said it wasn?t practical, etc.) >>>>> >>>>> There is much more to be said about all of this. But that seems to be the core of it. I find it very interesting how minor events have major effects. >>>>> >>>>> Take care, >>>>> John Day >>>>> >>>>>> On Jan 27, 2025, at 20:47, Brian E Carpenter via Internet-history > wrote: >>>>>> >>>>>> Vint, and Noel, >>>>>> >>>>>> I just glanced through Baran's 1964 paper, and it clearly recognized >>>>>> statelessnesss (and a standard packet header) as important for network >>>>>> survivability and adaptive routing. But although he mentions networks >>>>>> of intercontinental size, I didn't spot any discussion of scalability >>>>>> as such. >>>>>> >>>>>> Interestingly, exactly the same applies to Dave Clark's 1988 "Design >>>>>> Philosophy" paper. >>>>>> >>>>>> In RFC 1958, we did note as principle 3.3 that "All designs must scale >>>>>> readily to very many nodes per site and to many millions of sites". >>>>>> I guess that by then (1996) this was too obvious to ignore, and it was >>>>>> written when IPv4 address exhaustion was considered inevitable. >>>>>> >>>>>> Maybe somebody who knows the early literature better than me can find >>>>>> something. But it's almost as if the intrinsic scalability of stateless >>>>>> packet switching was an unnoticed and accidental property. >>>>>> >>>>>> Regards >>>>>> Brian >>>>>> >>>>>> On 27-Jan-25 11:16, Vint Cerf via Internet-history wrote: >>>>>>> statelessness was an important design choice and was made consciously so >>>>>>> that paths were not critical to successful transport. For example we did >>>>>>> not want to have to reassemble along a particular path. Even though we >>>>>>> deprecated fragmentation, at the time we thought it was important, we did >>>>>>> not want gateway (router) state to be necessary to accomplish reassembly >>>>>>> regardless of path. I don't know that we recognized the scalability aspect >>>>>>> but we definitely cared a lot about statelessness of the gateways. >>>>>>> v >>>>>>> On Sun, Jan 26, 2025 at 4:25?PM Noel Chiappa via Internet-history < >>>>>>> internet-history at elists.isoc.org > wrote: >>>>>>>> > From: Jack Haverty jack at 3kitty.org >>>>>>>> >>>>>>>> > At the time, the "ARPANET crowd" was skeptical that the "datagram" >>>>>>>> > nature of TCP could be made to work. Traditional networks, including >>>>>>>> > the ARPANET, had elaborate internal mechanisms to provide a "virtual >>>>>>>> > circuit" service to its users. >>>>>>>> >>>>>>>> I was thinkking about this, and wondering if internetworking was a more >>>>>>>> fundamental advance than the ARPANET (relegating the latter to a >>>>>>>> 'ground-breaking experiment'), and I had another thought. >>>>>>>> >>>>>>>> >>>>>>>> Internetworking (following in the track of CYCLADES) made much of the >>>>>>>> fate-sharing aspect - that the data needed to ensure reliable transmission >>>>>>>> was co-located was the application. One good reason for that (that we knew >>>>>>>> at >>>>>>>> the time) was that it made the network itself simpler. >>>>>>>> >>>>>>>> But there's another side to that, one that was even more important, and >>>>>>>> which >>>>>>>> I'm not sure was obvious to us at the time (1977-79), which is that because >>>>>>>> it means the intermediate packet switches in the overall internet carry no >>>>>>>> state about the connections travelling through them, there's no scaling >>>>>>>> limit. This, to me, has been the single biggest reason why the Internet has >>>>>>>> been able to grow to the stupendous size it has. >>>>>>>> >>>>>>>> I don't think we could have been thinking 'this aspect of lack of state in >>>>>>>> the internet packet switches neans it will scale indefinitely', because I >>>>>>>> don't think we had any idea, at that point, about how to do path selection >>>>>>>> in >>>>>>>> a global-scale internet - so global-scale internets could not have been in >>>>>>>> our thinking. >>>>>>>> >>>>>>>> Did that infinite scalability turn out to be just a happy accident, a >>>>>>>> side-effect of good fundamental design (but one whose true complete value >>>>>>>> wasn't obvious to us at the time), one that moved state out of the internet >>>>>>>> packet switches? >>>>>>>> >>>>>>>> Noel >>>>>>>> -- >>>>>>>> Internet-history mailing list >>>>>>>> Internet-history at elists.isoc.org >>>>>>>> https://elists.isoc.org/mailman/listinfo/internet-history >>>>>>>> >>>>>> -- >>>>>> Internet-history mailing list >>>>>> Internet-history at elists.isoc.org >>>>>> https://elists.isoc.org/mailman/listinfo/internet-history >>>> >>>> >>>> -- >>>> Please send any postal/overnight deliveries to: >>>> Vint Cerf >>>> Google, LLC >>>> 1900 Reston Metro Plaza, 16th Floor >>>> Reston, VA 20190 >>>> +1 (571) 213 1346 >>>> >>>> >>>> until further notice >>>> >>>> >>>> >> >> >> -- >> Please send any postal/overnight deliveries to: >> Vint Cerf >> Google, LLC >> 1900 Reston Metro Plaza, 16th Floor >> Reston, VA 20190 >> +1 (571) 213 1346 >> >> >> until further notice >> >> >> -------------- next part -------------- A non-text attachment was scrubbed... Name: OpenPGP_signature.asc Type: application/pgp-signature Size: 665 bytes Desc: OpenPGP digital signature URL: From jnc at mercury.lcs.mit.edu Tue Jan 28 09:13:48 2025 From: jnc at mercury.lcs.mit.edu (Noel Chiappa) Date: Tue, 28 Jan 2025 12:13:48 -0500 (EST) Subject: [ih] Correct name for early TCP/IP working group? Message-ID: <20250128171348.41F9218C073@mercury.lcs.mit.edu> > From: Jack Haverty > My recollections, all IIRC of course Thanks very much for those; I had previously set out to improve the "TCP and Internet Meetings" page: https://gunkies.org/wiki/TCP_and_Internet_Meetings a bit (I'd be very pleased to hear any comments about any remaining errors, and what else it needs), and I 'borrowed' a few chunks of your message, to explain the context, and who came - I hope that's OK. I too did live through most of this (the first meeting I came to was the August 1978 one). I just re-read most of the early minutes, looking for mentions of TCP 2.5 (didn't find much, alas), and most of the discussion seems to be about topics that later turned out to have been irrelevant, like EOL/Urgent and fragmentation. Perhaps I have missed something, but it seems, in retrospect, that the only really significant change from TCP 2 to TCP 4 was the TCP/IP split, and the creation of UDP. Note to future historians: there may be some detail errors in that message (well, it does say "IIRC"), so cross-check. E.g. "The IETF was formed to Engineer the operational Internet as it grew. The IRTF was formed to pursue the Research" - There was an 'InArc' formed at the same time as the IETF: https://www.eecis.udel.edu/~mills/database/papers/inarc.pdf but it never went anywhere. I think the IRTF was created later - I'm not sure exactly when, perhaps around the time of RFC 2014? (I have this vague memory of the IETF and InArc being initially announced at the same West Coast meeting, around the time of IETF 1.) Maybe the IRTF was created to do what Inarc should have done? Another one: > I had the impression that the INWG was part of the group that thought > the datagram architecture was unworkable. Mentally, I associated it > with X.25 and X.75 style of interconnecting networks. But perhaps that > was a mistake. I wasn't there, but I get the impression that some considerable part of the INWG actually was sold on datagrams: Pouzin, whose CYCLADES/CIGALE was the key step from the ARPANET to internets, was a big player in INWG; and Cerf and Kahn's "A Protocol for Packet Network Interconnection" originally came out as an INWG document. See Alexander McKenzie's "INWG and the Conception of the Internet: An Eyewitness Account": https://alexmckenzie.weebly.com/inwg-and-the-conception-of-the-internet-an-eyewitness-account.html for more detail. Noel From vint at google.com Tue Jan 28 09:26:48 2025 From: vint at google.com (Vint Cerf) Date: Tue, 28 Jan 2025 12:26:48 -0500 Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: <20250128171348.41F9218C073@mercury.lcs.mit.edu> References: <20250128171348.41F9218C073@mercury.lcs.mit.edu> Message-ID: Barry Leiner succeeded me at DARPA and created the Internet Advisory Board as a successor to the Internet Configuration Control Board. He later renamed the IAB: Internet Activities Board and created about ten task forces of which INARC was one. The fastest growing task force was IETF run first by Mike Corrigan and then by Phill Gross. Around 1989 it became clear that the IETF was the big task force kahuna and the other task forces were either abandoned or consolidated into the Internet Research Task Force, first chaired by David Clark and then Jon Postel. v On Tue, Jan 28, 2025 at 12:14?PM Noel Chiappa via Internet-history < internet-history at elists.isoc.org> wrote: > > From: Jack Haverty > > > My recollections, all IIRC of course > > Thanks very much for those; I had previously set out to improve the "TCP > and > Internet Meetings" page: > > https://gunkies.org/wiki/TCP_and_Internet_Meetings > > a bit (I'd be very pleased to hear any comments about any remaining errors, > and what else it needs), and I 'borrowed' a few chunks of your message, to > explain the context, and who came - I hope that's OK. > > I too did live through most of this (the first meeting I came to was the > August 1978 one). I just re-read most of the early minutes, looking for > mentions of TCP 2.5 (didn't find much, alas), and most of the discussion > seems to be about topics that later turned out to have been irrelevant, > like > EOL/Urgent and fragmentation. > > Perhaps I have missed something, but it seems, in retrospect, that the only > really significant change from TCP 2 to TCP 4 was the TCP/IP split, and the > creation of UDP. > > > Note to future historians: there may be some detail errors in that message > (well, it does say "IIRC"), so cross-check. E.g. "The IETF was formed to > Engineer the operational Internet as it grew. The IRTF was formed to pursue > the Research" - There was an 'InArc' formed at the same time as the IETF: > > https://www.eecis.udel.edu/~mills/database/papers/inarc.pdf > > but it never went anywhere. I think the IRTF was created later - I'm not > sure > exactly when, perhaps around the time of RFC 2014? (I have this vague > memory > of the IETF and InArc being initially announced at the same West Coast > meeting, around the time of IETF 1.) Maybe the IRTF was created to do what > Inarc should have done? > > Another one: > > > I had the impression that the INWG was part of the group that thought > > the datagram architecture was unworkable. Mentally, I associated it > > with X.25 and X.75 style of interconnecting networks. But perhaps > that > > was a mistake. > > I wasn't there, but I get the impression that some considerable part of the > INWG actually was sold on datagrams: Pouzin, whose CYCLADES/CIGALE was the > key step from the ARPANET to internets, was a big player in INWG; and Cerf > and Kahn's "A Protocol for Packet Network Interconnection" originally came > out as an INWG document. See Alexander McKenzie's "INWG and the Conception > of > the Internet: An Eyewitness Account": > > > https://alexmckenzie.weebly.com/inwg-and-the-conception-of-the-internet-an-eyewitness-account.html > > for more detail. > > Noel > -- > Internet-history mailing list > Internet-history at elists.isoc.org > https://elists.isoc.org/mailman/listinfo/internet-history > -- Please send any postal/overnight deliveries to: Vint Cerf Google, LLC 1900 Reston Metro Plaza, 16th Floor Reston, VA 20190 +1 (571) 213 1346 until further notice From jack at 3kitty.org Tue Jan 28 13:03:00 2025 From: jack at 3kitty.org (Jack Haverty) Date: Tue, 28 Jan 2025 13:03:00 -0800 Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: <20250128171348.41F9218C073@mercury.lcs.mit.edu> References: <20250128171348.41F9218C073@mercury.lcs.mit.edu> Message-ID: Noel said: "Perhaps I have missed something, but it seems, in retrospect, that the only really significant change from TCP 2 to TCP 4 was the TCP/IP split, and the creation of UDP." During that era, Vint moved the "Gateway" work from the Packet Radio group to the group I was managing at BBN. At the time, that included SATNET, which had previously evolved from a research project into an operational network. SATNET, and ARPANET, were both continuously managed by the NOC at BBN. TCP, and a collection of "core gateways" had become increasingly important to users, especially in the UK in Peter Kirstein's group, which had to rely on TCP through SATNET to access computers in the US. The charter for our new contract was to move the Internet "core" to operational status, i.e., "make the Internet a 24x7 reliable service", or something like that. The obvious way to do that was to adopt the NOC's techniques and tools that had been refined over a decade or so of operations. The Internet needed ways to diagnose and isolate problems, monitor activity, deploy new software, and use the network to perform such tasks even when it wasn't working well at all. Just like the ARPANET. To do that required adding a lot of technology to the base TCP system as defined in TCP2. Lots of stuff was "ported" from the ARPANET world, such as XNET. The concept of "fake hosts" from the IMPs was brought into the Internet, so that useful operations tools were available - things like traffic generators. Various IP "options" were defined, such as several varieties of "source routing", which were very useful when the normal routing mechanisms were failing, or when there was a need to test one of a particular gateway's interfaces that had been declared "down", by forcing traffic to use it. "Ping" was possible by implementing a simple "fake host" in a gateway or host. SNMP was defined, including facilities to manage not only IP traffic but also TCP, since that was where much of the error detection and response happens in response to failures that someone ought to notice and fix. In research environments, the equipment involved is usually located near the researchers, so such tools are not much of a concern to them. In operational environments, the equipment is often very far away and in multiple locations. Because of the pressure to get out a DoD Standard, much of that new "operations" technology didn't make it into the TCP or IP spec that Jon assembled. Rather, some of it was collected and named ICMP - Internet Control Message Protocol. Others just got their own documentation, sometimes in an RFC, sometimes just in email messages. That incompleteness of the specs caused some problems especially as DoD systems began to come online after traditional government contractors got involved. They had been tasked to implement the DoD standards. Their contract didn't say that they had to implement ICMP. So they didn't. That made it very difficult to actually operate a network based solely on the DoD Standards. We lobbied very hard to get the additional technology implemented in systems being deployed by the Army, Air Force, etc. I'm not sure what, if anything, changed in the "DoD Standards" regulations. One of our "operational" government clients tasked us to write a handbook on Network Operations. They couldn't find any such "User's Manual" in all of the RFCs et al. So I wrote a large document titled something like "How To Manage A Large Network". Sadly I didn't keep a copy but it may exist still, buried somewhere in DTIC or a government warehouse. So, while the split between TCP and IP was the most prominent change from 2 to 4, there were a lot of other changes to make the Internet "reliable 24x7", which continued well after the DoD Standard was created. There were a bunch of unfinished "research topics" that also weren't there yet - e.g., the desire for routing mechanisms based on actual transit time rather than hops. Jack Haverty On 1/28/25 09:13, Noel Chiappa via Internet-history wrote: > > From: Jack Haverty > > > My recollections, all IIRC of course > > Thanks very much for those; I had previously set out to improve the "TCP and > Internet Meetings" page: > > https://gunkies.org/wiki/TCP_and_Internet_Meetings > > a bit (I'd be very pleased to hear any comments about any remaining errors, > and what else it needs), and I 'borrowed' a few chunks of your message, to > explain the context, and who came - I hope that's OK. > > I too did live through most of this (the first meeting I came to was the > August 1978 one). I just re-read most of the early minutes, looking for > mentions of TCP 2.5 (didn't find much, alas), and most of the discussion > seems to be about topics that later turned out to have been irrelevant, like > EOL/Urgent and fragmentation. > > Perhaps I have missed something, but it seems, in retrospect, that the only > really significant change from TCP 2 to TCP 4 was the TCP/IP split, and the > creation of UDP. > > > Note to future historians: there may be some detail errors in that message > (well, it does say "IIRC"), so cross-check. E.g. "The IETF was formed to > Engineer the operational Internet as it grew. The IRTF was formed to pursue > the Research" - There was an 'InArc' formed at the same time as the IETF: > > https://www.eecis.udel.edu/~mills/database/papers/inarc.pdf > > but it never went anywhere. I think the IRTF was created later - I'm not sure > exactly when, perhaps around the time of RFC 2014? (I have this vague memory > of the IETF and InArc being initially announced at the same West Coast > meeting, around the time of IETF 1.) Maybe the IRTF was created to do what > Inarc should have done? > > Another one: > > > I had the impression that the INWG was part of the group that thought > > the datagram architecture was unworkable. Mentally, I associated it > > with X.25 and X.75 style of interconnecting networks. But perhaps that > > was a mistake. > > I wasn't there, but I get the impression that some considerable part of the > INWG actually was sold on datagrams: Pouzin, whose CYCLADES/CIGALE was the > key step from the ARPANET to internets, was a big player in INWG; and Cerf > and Kahn's "A Protocol for Packet Network Interconnection" originally came > out as an INWG document. See Alexander McKenzie's "INWG and the Conception of > the Internet: An Eyewitness Account": > > https://alexmckenzie.weebly.com/inwg-and-the-conception-of-the-internet-an-eyewitness-account.html > > for more detail. > > Noel -------------- next part -------------- A non-text attachment was scrubbed... Name: OpenPGP_signature.asc Type: application/pgp-signature Size: 665 bytes Desc: OpenPGP digital signature URL: From b_a_denny at yahoo.com Tue Jan 28 14:08:56 2025 From: b_a_denny at yahoo.com (Barbara Denny) Date: Tue, 28 Jan 2025 22:08:56 +0000 (UTC) Subject: [ih] Correct name for early TCP/IP working group? References: <340430285.5621070.1738102136420.ref@mail.yahoo.com> Message-ID: <340430285.5621070.1738102136420@mail.yahoo.com> Just a FYI I think Jill Westcott did some work? on hierarchical routing.? I believe she was looking at it as a way to increase the number of radio nodes in a packet radio network (around 1982/83 but? may have started earlier).? I don't remember her publishing anything on the topic but you might find some info in DTIC, probably in packet radio meeting notes or client reports if interested.? barbara On Tuesday, January 28, 2025 at 04:06:48 AM PST, Scott Bradner via Internet-history wrote: ps - you can find the paper on Kleinrock's website https://www.lk.cs.ucla.edu/data/files/Kamoun/Data%20Communications%20through%20Large%20Packet-Switching%20Networks.pdf > On Jan 28, 2025, at 6:55?AM, Scott Bradner via Internet-history wrote: > > fwiw - the 1976? Kleinrock & Kamoun paper "Hierarchical Routing for Large Networks, Performance evaluation and optimization" is all about scaling of packet switched networks > > Scott > > > >> On Jan 27, 2025, at 8:47?PM, Brian E Carpenter via Internet-history wrote: >> >> Vint, and Noel, >> >> I just glanced through Baran's 1964 paper, and it clearly recognized >> statelessnesss (and a standard packet header) as important for network >> survivability and adaptive routing. But although he mentions networks >> of intercontinental size, I didn't spot any discussion of scalability >> as such. >> >> Interestingly, exactly the same applies to Dave Clark's 1988 "Design >> Philosophy" paper. >> >> In RFC 1958, we did note as principle 3.3 that "All designs must scale >> readily to very many nodes per site and to many millions of sites". >> I guess that by then (1996) this was too obvious to ignore, and it was >> written when IPv4 address exhaustion was considered inevitable. >> >> Maybe somebody who knows the early literature better than me can find >> something. But it's almost as if the intrinsic scalability of stateless >> packet switching was an unnoticed and accidental property. >> >> Regards >>? Brian >> >> On 27-Jan-25 11:16, Vint Cerf via Internet-history wrote: >>> statelessness was an important design choice and was made consciously so >>> that paths were not critical to successful transport. For example we did >>> not want to have to reassemble along a particular path. Even though we >>> deprecated fragmentation, at the time we thought it was important, we did >>> not want gateway (router) state to be necessary to accomplish reassembly >>> regardless of path. I don't know that we recognized the scalability aspect >>> but we definitely cared a lot about statelessness of the gateways. >>> v >>> On Sun, Jan 26, 2025 at 4:25?PM Noel Chiappa via Internet-history < >>> internet-history at elists.isoc.org> wrote: >>>>> From: Jack Haverty jack at 3kitty.org >>>> >>>>> At the time, the "ARPANET crowd" was skeptical that the "datagram" >>>>> nature of TCP could be made to work. Traditional networks, including >>>>> the ARPANET, had elaborate internal mechanisms to provide a "virtual >>>>> circuit" service to its users. >>>> >>>> I was thinkking about this, and wondering if internetworking was a more >>>> fundamental advance than the ARPANET (relegating the latter to a >>>> 'ground-breaking experiment'), and I had another thought. >>>> >>>> >>>> Internetworking (following in the track of CYCLADES) made much of the >>>> fate-sharing aspect - that the data needed to ensure reliable transmission >>>> was co-located was the application. One good reason for that (that we knew >>>> at >>>> the time) was that it made the network itself simpler. >>>> >>>> But there's another side to that, one that was even more important, and >>>> which >>>> I'm not sure was obvious to us at the time (1977-79), which is that because >>>> it means the intermediate packet switches in the overall internet carry no >>>> state about the connections travelling through them, there's no scaling >>>> limit. This, to me, has been the single biggest reason why the Internet has >>>> been able to grow to the stupendous size it has. >>>> >>>> I don't think we could have been thinking 'this aspect of lack of state in >>>> the internet packet switches neans it will scale indefinitely', because I >>>> don't think we had any idea, at that point, about how to do path selection >>>> in >>>> a global-scale internet - so global-scale internets could not have been in >>>> our thinking. >>>> >>>> Did that infinite scalability turn out to be just a happy accident, a >>>> side-effect of good fundamental design (but one whose true complete value >>>> wasn't obvious to us at the time), one that moved state out of the internet >>>> packet switches? >>>> >>>>? ? ? ? Noel >>>> -- >>>> Internet-history mailing list >>>> Internet-history at elists.isoc.org >>>> https://elists.isoc.org/mailman/listinfo/internet-history >>>> >> -- >> Internet-history mailing list >> Internet-history at elists.isoc.org >> https://elists.isoc.org/mailman/listinfo/internet-history > > -- > Internet-history mailing list > Internet-history at elists.isoc.org > https://elists.isoc.org/mailman/listinfo/internet-history -- Internet-history mailing list Internet-history at elists.isoc.org https://elists.isoc.org/mailman/listinfo/internet-history From b_a_denny at yahoo.com Tue Jan 28 14:24:42 2025 From: b_a_denny at yahoo.com (Barbara Denny) Date: Tue, 28 Jan 2025 22:24:42 +0000 (UTC) Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: <340430285.5621070.1738102136420@mail.yahoo.com> References: <340430285.5621070.1738102136420.ref@mail.yahoo.com> <340430285.5621070.1738102136420@mail.yahoo.com> Message-ID: <806421957.5623802.1738103082732@mail.yahoo.com> Oops phone didn't trust my spelling and I didn't see it in time. Jil has one l in her first name. barbara On Tuesday, January 28, 2025 at 02:11:12 PM PST, Barbara Denny via Internet-history wrote: Just a FYI I think Jill Westcott did some work? on hierarchical routing.? I believe she was looking at it as a way to increase the number of radio nodes in a packet radio network (around 1982/83 but? may have started earlier).? I don't remember her publishing anything on the topic but you might find some info in DTIC, probably in packet radio meeting notes or client reports if interested.? barbara ? ? On Tuesday, January 28, 2025 at 04:06:48 AM PST, Scott Bradner via Internet-history wrote:? ps - you can find the paper on Kleinrock's website https://www.lk.cs.ucla.edu/data/files/Kamoun/Data%20Communications%20through%20Large%20Packet-Switching%20Networks.pdf > On Jan 28, 2025, at 6:55?AM, Scott Bradner via Internet-history wrote: > > fwiw - the 1976? Kleinrock & Kamoun paper "Hierarchical Routing for Large Networks, Performance evaluation and optimization" is all about scaling of packet switched networks > > Scott > > > >> On Jan 27, 2025, at 8:47?PM, Brian E Carpenter via Internet-history wrote: >> >> Vint, and Noel, >> >> I just glanced through Baran's 1964 paper, and it clearly recognized >> statelessnesss (and a standard packet header) as important for network >> survivability and adaptive routing. But although he mentions networks >> of intercontinental size, I didn't spot any discussion of scalability >> as such. >> >> Interestingly, exactly the same applies to Dave Clark's 1988 "Design >> Philosophy" paper. >> >> In RFC 1958, we did note as principle 3.3 that "All designs must scale >> readily to very many nodes per site and to many millions of sites". >> I guess that by then (1996) this was too obvious to ignore, and it was >> written when IPv4 address exhaustion was considered inevitable. >> >> Maybe somebody who knows the early literature better than me can find >> something. But it's almost as if the intrinsic scalability of stateless >> packet switching was an unnoticed and accidental property. >> >> Regards >>? Brian >> >> On 27-Jan-25 11:16, Vint Cerf via Internet-history wrote: >>> statelessness was an important design choice and was made consciously so >>> that paths were not critical to successful transport. For example we did >>> not want to have to reassemble along a particular path. Even though we >>> deprecated fragmentation, at the time we thought it was important, we did >>> not want gateway (router) state to be necessary to accomplish reassembly >>> regardless of path. I don't know that we recognized the scalability aspect >>> but we definitely cared a lot about statelessness of the gateways. >>> v >>> On Sun, Jan 26, 2025 at 4:25?PM Noel Chiappa via Internet-history < >>> internet-history at elists.isoc.org> wrote: >>>>> From: Jack Haverty jack at 3kitty.org >>>> >>>>> At the time, the "ARPANET crowd" was skeptical that the "datagram" >>>>> nature of TCP could be made to work. Traditional networks, including >>>>> the ARPANET, had elaborate internal mechanisms to provide a "virtual >>>>> circuit" service to its users. >>>> >>>> I was thinkking about this, and wondering if internetworking was a more >>>> fundamental advance than the ARPANET (relegating the latter to a >>>> 'ground-breaking experiment'), and I had another thought. >>>> >>>> >>>> Internetworking (following in the track of CYCLADES) made much of the >>>> fate-sharing aspect - that the data needed to ensure reliable transmission >>>> was co-located was the application. One good reason for that (that we knew >>>> at >>>> the time) was that it made the network itself simpler. >>>> >>>> But there's another side to that, one that was even more important, and >>>> which >>>> I'm not sure was obvious to us at the time (1977-79), which is that because >>>> it means the intermediate packet switches in the overall internet carry no >>>> state about the connections travelling through them, there's no scaling >>>> limit. This, to me, has been the single biggest reason why the Internet has >>>> been able to grow to the stupendous size it has. >>>> >>>> I don't think we could have been thinking 'this aspect of lack of state in >>>> the internet packet switches neans it will scale indefinitely', because I >>>> don't think we had any idea, at that point, about how to do path selection >>>> in >>>> a global-scale internet - so global-scale internets could not have been in >>>> our thinking. >>>> >>>> Did that infinite scalability turn out to be just a happy accident, a >>>> side-effect of good fundamental design (but one whose true complete value >>>> wasn't obvious to us at the time), one that moved state out of the internet >>>> packet switches? >>>> >>>>? ? ? ? Noel >>>> -- >>>> Internet-history mailing list >>>> Internet-history at elists.isoc.org >>>> https://elists.isoc.org/mailman/listinfo/internet-history >>>> >> -- >> Internet-history mailing list >> Internet-history at elists.isoc.org >> https://elists.isoc.org/mailman/listinfo/internet-history > > -- > Internet-history mailing list > Internet-history at elists.isoc.org > https://elists.isoc.org/mailman/listinfo/internet-history -- Internet-history mailing list Internet-history at elists.isoc.org https://elists.isoc.org/mailman/listinfo/internet-history ? -- Internet-history mailing list Internet-history at elists.isoc.org https://elists.isoc.org/mailman/listinfo/internet-history From gregskinner0 at icloud.com Tue Jan 28 21:38:32 2025 From: gregskinner0 at icloud.com (Greg Skinner) Date: Tue, 28 Jan 2025 21:38:32 -0800 Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: <20250128171348.41F9218C073@mercury.lcs.mit.edu> References: <20250128171348.41F9218C073@mercury.lcs.mit.edu> Message-ID: <34AFAD57-5CB2-416C-A431-41CBB868EDEE@icloud.com> On Jan 28, 2025, at 9:13?AM, Noel Chiappa via Internet-history wrote: > > Note to future historians: there may be some detail errors in that message > (well, it does say "IIRC"), so cross-check. E.g. "The IETF was formed to > Engineer the operational Internet as it grew. The IRTF was formed to pursue > the Research" - There was an 'InArc' formed at the same time as the IETF: > > https://www.eecis.udel.edu/~mills/database/papers/inarc.pdf > > but it never went anywhere. I think the IRTF was created later - I'm not sure > exactly when, perhaps around the time of RFC 2014? (I have this vague memory > of the IETF and InArc being initially announced at the same West Coast > meeting, around the time of IETF 1.) Maybe the IRTF was created to do what > Inarc should have done? > The INARC announcement is in the IETF 1 (San Diego) proceedings. https://www.ietf.org/proceedings/01.pdf --gregbo From lars at nocrew.org Tue Jan 28 22:04:13 2025 From: lars at nocrew.org (Lars Brinkhoff) Date: Wed, 29 Jan 2025 06:04:13 +0000 Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: (Jack Haverty via Internet-history's message of "Tue, 28 Jan 2025 13:03:00 -0800") References: <20250128171348.41F9218C073@mercury.lcs.mit.edu> Message-ID: <7wjzae17he.fsf@junk.nocrew.org> Jack Haverty wrote: > SATNET, and ARPANET, were both continuously managed by the NOC at > BBN. TCP, and a collection of "core gateways" had become increasingly > important to users, especially in the UK in Peter Kirstein's group, > which had to rely on TCP through SATNET to access computers in the US. This would have been before 1983, correct? I'm curious which (important) computers were avaiable through TCP? It seems to me the transition from NCP to TCP was somewhat gradual. Many histories make a big deal out of the 1/1/83 flag day, as if the entire network switched from NCP-only on one day, to TCP-only the next day. But reading more carefully, I gather some hosts were TCP only or dual TCP/NCP long before that. Is that correct? From jack at 3kitty.org Tue Jan 28 23:33:47 2025 From: jack at 3kitty.org (Jack Haverty) Date: Tue, 28 Jan 2025 23:33:47 -0800 Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: <7wjzae17he.fsf@junk.nocrew.org> References: <20250128171348.41F9218C073@mercury.lcs.mit.edu> <7wjzae17he.fsf@junk.nocrew.org> Message-ID: Short answer -- yes, well before 1983. The planning for 1/1/1983 began before sometime before 21 September 1981.?? That was the date of the first ICCB meeting.? My notes from that meeting contain a list of technical issues that had to be worked out in anticipation of the "Jan 83 System (heavy load)". The ARPANET was managed at that point by DCA, as an operational network.? So the task of getting ready for TCP fell on the guy in charge at DCA, who was Joe Haughney.? I'm not sure of all that he did, but he's the one who set the 1/1/1983 date for the cutover. Joe's daughter Christine recently (2023) put together a podcast "The untold history of how the internet almost didn't happen", which contains more details about who did what during that timeframe. It's still online at: https://www.inc.com/computerfreak? I don't recall everything in there (IIRC I'm on it a little bit too), but it might make a good historical resource, containing a view from the operations perspective.? On the podcast, Joe explains, somewhat gleefully, that he was the one who picked New Year's Eve for the cutover. There was a *lot* of preparatory work before that cutover, which turned out to be somewhat boring.? We had engineers and programmers standing by to fix whatever problems occurred.?? But after "the switch" was thrown --- nothing much happened. For example, NBS (now NIST) created a program for testing new TCP implementations to make sure they followed the spec.?? At BBN, in preparation for the later DDN activity, we set up a service which would run the NBS tests for clients (using a dialup link), and then help them as consultants to fix whatever wasn't working.?? There was lots of work to convert older programs like Telnet, FTP, and mail to use TCP instead of NCP, and to get ancillary, but important, technologies such as SNMP and ICMP widely implemented. Thanks to ARPA, TCP had already been implemented for many different types of computers, so the task was in many cases just getting the various owners of ARPANET machines to put TCP into their systems.? A threat (actually promise) was made that after the cutover date, NCP connections would no longer work at all.? That was apparently enough motivation. It took well over a year, but the cutover went smoothly.? IMHO that was directly due to the advance planning and related work to get ready.? The cutover got all the attention, but the work beforehand made it successful. I don't really remember what machines had TCP over what schedules. But DCA ran the ARPANET and had SRI with Jake Feinler managing the "NIC" (Network Information Center) on SRI-KL.? That's where I'd expect to find lists of machines (e.g., the HOSTS.TXT file that was used before DNS existed), as well as the DCA Newsletters which came out periodically and captured the status of the ARPANET and computers attached to it.?? All that kind of information would have come from the NIC. Jack On 1/28/25 22:04, Lars Brinkhoff wrote: > Jack Haverty wrote: >> SATNET, and ARPANET, were both continuously managed by the NOC at >> BBN. TCP, and a collection of "core gateways" had become increasingly >> important to users, especially in the UK in Peter Kirstein's group, >> which had to rely on TCP through SATNET to access computers in the US. > This would have been before 1983, correct? I'm curious which > (important) computers were avaiable through TCP? > > It seems to me the transition from NCP to TCP was somewhat gradual. > Many histories make a big deal out of the 1/1/83 flag day, as if the > entire network switched from NCP-only on one day, to TCP-only the next > day. But reading more carefully, I gather some hosts were TCP only or > dual TCP/NCP long before that. Is that correct? -------------- next part -------------- A non-text attachment was scrubbed... Name: OpenPGP_signature.asc Type: application/pgp-signature Size: 665 bytes Desc: OpenPGP digital signature URL: From gbuday.irtf at gmail.com Tue Jan 28 23:46:27 2025 From: gbuday.irtf at gmail.com (Gergely Buday) Date: Wed, 29 Jan 2025 07:46:27 +0000 Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: References: <20250128171348.41F9218C073@mercury.lcs.mit.edu> <7wjzae17he.fsf@junk.nocrew.org> Message-ID: https://www.inc.com/computerfreaks is the working link. - Gergely Jack Haverty via Internet-history ezt ?rta (id?pont: 2025. jan. 29., Sze 7:41): > Short answer -- yes, well before 1983. > > The planning for 1/1/1983 began before sometime before 21 September > 1981. That was the date of the first ICCB meeting. My notes from that > meeting contain a list of technical issues that had to be worked out in > anticipation of the "Jan 83 System (heavy load)". > > The ARPANET was managed at that point by DCA, as an operational > network. So the task of getting ready for TCP fell on the guy in charge > at DCA, who was Joe Haughney. I'm not sure of all that he did, but he's > the one who set the 1/1/1983 date for the cutover. > > Joe's daughter Christine recently (2023) put together a podcast "The > untold history of how the internet almost didn't happen", which contains > more details about who did what during that timeframe. It's still online > at: > https://www.inc.com/computerfreak I don't recall everything in there > (IIRC I'm on it a little bit too), but it might make a good historical > resource, containing a view from the operations perspective. On the > podcast, Joe explains, somewhat gleefully, that he was the one who > picked New Year's Eve for the cutover. > > There was a *lot* of preparatory work before that cutover, which turned > out to be somewhat boring. We had engineers and programmers standing by > to fix whatever problems occurred. But after "the switch" was thrown > --- nothing much happened. > > For example, NBS (now NIST) created a program for testing new TCP > implementations to make sure they followed the spec. At BBN, in > preparation for the later DDN activity, we set up a service which would > run the NBS tests for clients (using a dialup link), and then help them > as consultants to fix whatever wasn't working. There was lots of work > to convert older programs like Telnet, FTP, and mail to use TCP instead > of NCP, and to get ancillary, but important, technologies such as SNMP > and ICMP widely implemented. > > Thanks to ARPA, TCP had already been implemented for many different > types of computers, so the task was in many cases just getting the > various owners of ARPANET machines to put TCP into their systems. A > threat (actually promise) was made that after the cutover date, NCP > connections would no longer work at all. That was apparently enough > motivation. > > It took well over a year, but the cutover went smoothly. IMHO that was > directly due to the advance planning and related work to get ready. The > cutover got all the attention, but the work beforehand made it successful. > > I don't really remember what machines had TCP over what schedules. But > DCA ran the ARPANET and had SRI with Jake Feinler managing the "NIC" > (Network Information Center) on SRI-KL. That's where I'd expect to find > lists of machines (e.g., the HOSTS.TXT file that was used before DNS > existed), as well as the DCA Newsletters which came out periodically and > captured the status of the ARPANET and computers attached to it. All > that kind of information would have come from the NIC. > > Jack > > On 1/28/25 22:04, Lars Brinkhoff wrote: > > Jack Haverty wrote: > >> SATNET, and ARPANET, were both continuously managed by the NOC at > >> BBN. TCP, and a collection of "core gateways" had become increasingly > >> important to users, especially in the UK in Peter Kirstein's group, > >> which had to rely on TCP through SATNET to access computers in the US. > > This would have been before 1983, correct? I'm curious which > > (important) computers were avaiable through TCP? > > > > It seems to me the transition from NCP to TCP was somewhat gradual. > > Many histories make a big deal out of the 1/1/83 flag day, as if the > > entire network switched from NCP-only on one day, to TCP-only the next > > day. But reading more carefully, I gather some hosts were TCP only or > > dual TCP/NCP long before that. Is that correct? > > -- > Internet-history mailing list > Internet-history at elists.isoc.org > https://elists.isoc.org/mailman/listinfo/internet-history > From lk at cs.ucla.edu Tue Jan 28 23:53:36 2025 From: lk at cs.ucla.edu (Leonard Kleinrock) Date: Tue, 28 Jan 2025 23:53:36 -0800 Subject: [ih] Some comments on the history In-Reply-To: References: <20250126212518.AC40818C079@mercury.lcs.mit.edu> <0405134f-a905-43d1-991c-49aca2c203c8@gmail.com> <52E30F78-6795-4C7D-A7DA-6EA3FC71F964@comcast.net> Message-ID: <96292BD2-C6CD-4EC7-87D2-81D9FED48263@cs.ucla.edu> Vint, Steve, Scott, et al, Thank you for adding those useful comments in this chain. I would like to comment on two of them: 1. First, regarding the issue of an early focus on large networks, i.e., the scaling effect: As you correctly point out, Farouk Kamoun and I focused strongly on the impact of large networks and the efficiencies to be obtained in large networks as evidenced in our papers (1976 , 1977 , 1979 , 1980 ). Moreover, my early research was to address the networking issues in large communication nets; indeed, the opening line of my 1961 PhD thesis proposal ?Information Flow in Large Communication Nets ? is ?The purpose of this thesis is to investigate the problems associated with information flow in large communication nets.?. I again referred to large networks later in the proposal, namely, ?The two important characteristics of the communication nets that form the subject of this thesis are (1) the number of nodes in the system is large, and ? ? So there we see the recognition that scaling to large networks was a consideration in the earliest days of network research. 2. Second, regarding the history of recognition that packetization of messages provided many benefits and the discussion as to when Larry Roberts learned about the advantages of packetization: I introduced the idea of packetizing messages into short fixed length blocks in my 1962 paper and there I show that the round robin protocol already used in computer timesharing systems could be adapted to data networks and that motivated me to provide the first mathematical model of packetization and analyzed the exact gains to be had for short messages. I clearly pointed out that the notion of chopping units into fixed length blocks was specifically intended for messages in a communication network; for example, in my 1963 PhD dissertation , I say, ?We now explore the manner in which message delay is affected when one introduces a priority structure (or queue discipline) on the set of messages ??. As was pointed out, Larry was my officemate at MIT and we discussed my (and his) research constantly (in fact he wrote a metacompiler for the program I wrote to simulate communication nets). He was intimately familiar with my research and he clearly knew about packetization and its benefits as early as 1961. As a side note, Larry himself stated ?In order to plan to spend millions of dollars and stake my reputation, I needed to understand that it would work. Without Kleinrock's work of Networks and Queuing Theory, I could never have taken such a radical step? . ? I hope this adds more color and background to the discussion of these issues. Len > On Jan 28, 2025, at 5:49?AM, Vint Cerf via Internet-history wrote: > > roberts would also have been aware of Len Kleinrock's queueing theory > analysis of message switching which, mathematically, was not very different > from packet switching. They were both at MIT if memory serves and got their > Ph.D's the same year, 1963. > > v > > > On Tue, Jan 28, 2025 at 7:50?AM John Day > wrote: > >> Brian, >> I agree with you. In the Baran reports, he describes something that sounds >> like a datagram. However, he never explores it much other than to define >> hot-potato routing. His focus is very centered on survivability and >> resilience, which makes sense it was research for the DoD. There is also >> the consideration that so far as I have been able to determine, all of the >> projects Baran was involved in afterwards were virtual-circuit, as were >> Roberts. >> >> OTOH, NPL didn?t do military research, so I guessed that their impetus for >> exploring packet switching had to be different and indeed it was. We have >> found a memo Davies wrote (and a similar one by Derek Barber) that Davies >> had attended the IFIP Congress in the US in 1965 and heard lots of papers >> on timesharing and the time slicing scheduling that timesharing used. The >> advantage being that while batch systems did FCFS and short jobs got stuck >> behind long ones, timesharut ing interleaved jobs and while short jobs were >> still delayed but their *completion* times were shorter. Davies told Derek >> that that was what they should do with communications and they did. Their >> impetus for packet switching was very different. (But there were two >> problems. ;-) 1) Donald got promoted and less time for research, ;-) and 2) >> the GPO got involved and the government directed NPL to concentrate on >> ?practical? projects, so they had to move to virtual-circuits. >> >> Scantlebury told Roberts about packet switching at the Gatlinburg >> conference and convinced him to use it. When Roberts returned to DC, he >> found he had Baran?s reports in a stack of documents but hadn?t read it >> yet. Based on the NPL experience Roger also convinced Roberts not to use >> 2.4Kbps lines but 50Kbps, which was a large part of the ARPANET success. >> (Slower speed would have worked but been so slow people would have said it >> wasn?t practical, etc.) >> >> There is much more to be said about all of this. But that seems to be the >> core of it. I find it very interesting how minor events have major effects. >> >> Take care, >> John Day >> >>> On Jan 27, 2025, at 20:47, Brian E Carpenter via Internet-history < >> internet-history at elists.isoc.org> wrote: >>> >>> Vint, and Noel, >>> >>> I just glanced through Baran's 1964 paper, and it clearly recognized >>> statelessnesss (and a standard packet header) as important for network >>> survivability and adaptive routing. But although he mentions networks >>> of intercontinental size, I didn't spot any discussion of scalability >>> as such. >>> >>> Interestingly, exactly the same applies to Dave Clark's 1988 "Design >>> Philosophy" paper. >>> >>> In RFC 1958, we did note as principle 3.3 that "All designs must scale >>> readily to very many nodes per site and to many millions of sites". >>> I guess that by then (1996) this was too obvious to ignore, and it was >>> written when IPv4 address exhaustion was considered inevitable. >>> >>> Maybe somebody who knows the early literature better than me can find >>> something. But it's almost as if the intrinsic scalability of stateless >>> packet switching was an unnoticed and accidental property. >>> >>> Regards >>> Brian >>> >>> On 27-Jan-25 11:16, Vint Cerf via Internet-history wrote: >>>> statelessness was an important design choice and was made consciously so >>>> that paths were not critical to successful transport. For example we did >>>> not want to have to reassemble along a particular path. Even though we >>>> deprecated fragmentation, at the time we thought it was important, we >> did >>>> not want gateway (router) state to be necessary to accomplish reassembly >>>> regardless of path. I don't know that we recognized the scalability >> aspect >>>> but we definitely cared a lot about statelessness of the gateways. >>>> v >>>> On Sun, Jan 26, 2025 at 4:25?PM Noel Chiappa via Internet-history < >>>> internet-history at elists.isoc.org> wrote: >>>>>> From: Jack Haverty jack at 3kitty.org >>>>> >>>>>> At the time, the "ARPANET crowd" was skeptical that the >> "datagram" >>>>>> nature of TCP could be made to work. Traditional networks, >> including >>>>>> the ARPANET, had elaborate internal mechanisms to provide a >> "virtual >>>>>> circuit" service to its users. >>>>> >>>>> I was thinkking about this, and wondering if internetworking was a more >>>>> fundamental advance than the ARPANET (relegating the latter to a >>>>> 'ground-breaking experiment'), and I had another thought. >>>>> >>>>> >>>>> Internetworking (following in the track of CYCLADES) made much of the >>>>> fate-sharing aspect - that the data needed to ensure reliable >> transmission >>>>> was co-located was the application. One good reason for that (that we >> knew >>>>> at >>>>> the time) was that it made the network itself simpler. >>>>> >>>>> But there's another side to that, one that was even more important, and >>>>> which >>>>> I'm not sure was obvious to us at the time (1977-79), which is that >> because >>>>> it means the intermediate packet switches in the overall internet >> carry no >>>>> state about the connections travelling through them, there's no scaling >>>>> limit. This, to me, has been the single biggest reason why the >> Internet has >>>>> been able to grow to the stupendous size it has. >>>>> >>>>> I don't think we could have been thinking 'this aspect of lack of >> state in >>>>> the internet packet switches neans it will scale indefinitely', >> because I >>>>> don't think we had any idea, at that point, about how to do path >> selection >>>>> in >>>>> a global-scale internet - so global-scale internets could not have >> been in >>>>> our thinking. >>>>> >>>>> Did that infinite scalability turn out to be just a happy accident, a >>>>> side-effect of good fundamental design (but one whose true complete >> value >>>>> wasn't obvious to us at the time), one that moved state out of the >> internet >>>>> packet switches? >>>>> >>>>> Noel >>>>> -- >>>>> Internet-history mailing list >>>>> Internet-history at elists.isoc.org >>>>> https://elists.isoc.org/mailman/listinfo/internet-history >>>>> >>> -- >>> Internet-history mailing list >>> Internet-history at elists.isoc.org >>> https://elists.isoc.org/mailman/listinfo/internet-history >> >> > > -- > Please send any postal/overnight deliveries to: > Vint Cerf > Google, LLC > 1900 Reston Metro Plaza, 16th Floor > Reston, VA 20190 > +1 (571) 213 1346 > > > until further notice > -- > Internet-history mailing list > Internet-history at elists.isoc.org > https://elists.isoc.org/mailman/listinfo/internet-history From b_a_denny at yahoo.com Tue Jan 28 23:57:10 2025 From: b_a_denny at yahoo.com (Barbara Denny) Date: Wed, 29 Jan 2025 07:57:10 +0000 (UTC) Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: <7wjzae17he.fsf@junk.nocrew.org> References: <20250128171348.41F9218C073@mercury.lcs.mit.edu> <7wjzae17he.fsf@junk.nocrew.org> Message-ID: <1314488472.5766718.1738137430770@mail.yahoo.com> TCP existed prior to 1983 in hosts attached to the various networks that comprised the Internet.? ARPAnet was just one of the networks on the Internet during this time period. Flag day marked the day when NCP would no longer be supported in the ARPAnet. Hosts attached to the ARPAnet needed to use TCP or they would essentially be disconnected. Perhaps the network diagrams in this brochure will help you.? Unfortunately I had trouble just grabbing the image of the 1976 2 network internet transmission (Packet Radio and ARPAnet ) so I am including the entire publication.? https://computerhistory.org/wp-content/uploads/2019/08/core-2002-02.pdf barbara On Tuesday, January 28, 2025 at 10:04:22 PM PST, Lars Brinkhoff via Internet-history wrote: Jack Haverty wrote: > SATNET, and ARPANET, were both continuously managed by the NOC at > BBN. TCP, and a collection of "core gateways" had become increasingly > important to users, especially in the UK in Peter Kirstein's group, > which had to rely on TCP through SATNET to access computers in the US. This would have been before 1983, correct?? I'm curious which (important) computers were avaiable through TCP? It seems to me the transition from NCP to TCP was somewhat gradual. Many histories make a big deal out of the 1/1/83 flag day, as if the entire network switched from NCP-only on one day, to TCP-only the next day.? But reading more carefully, I gather some hosts were TCP only or dual TCP/NCP long before that.? Is that correct? -- Internet-history mailing list Internet-history at elists.isoc.org https://elists.isoc.org/mailman/listinfo/internet-history From craig at tereschau.net Wed Jan 29 01:19:38 2025 From: craig at tereschau.net (Craig Partridge) Date: Wed, 29 Jan 2025 02:19:38 -0700 Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: References: <20250128171348.41F9218C073@mercury.lcs.mit.edu> <7wjzae17he.fsf@junk.nocrew.org> Message-ID: On Wed, Jan 29, 2025 at 12:41?AM Jack Haverty via Internet-history < internet-history at elists.isoc.org> wrote: > For example, NBS (now NIST) created a program for testing new TCP > implementations to make sure they followed the spec. At BBN, in > preparation for the later DDN activity, we set up a service which would > run the NBS tests for clients (using a dialup link), and then help them > as consultants to fix whatever wasn't working. There was lots of work > to convert older programs like Telnet, FTP, and mail to use TCP instead > of NCP, and to get ancillary, but important, technologies such as SNMP > and ICMP widely implemented. > > Small nit. SNMP didn't exist until 1988. The Internet did not have a standard management protocol until then. What you're probably thinking of is HMP (the Host Monitoring Protocol), which despite its name, was actually used to monitor the health of routers. It was developed c. 1981 (IEN 197) and was supported on the BBN routers and mailbridges in the 1980s. Craig -- ***** Craig Partridge's email account for professional society activities and mailing lists. From vint at google.com Wed Jan 29 03:48:03 2025 From: vint at google.com (Vint Cerf) Date: Wed, 29 Jan 2025 06:48:03 -0500 Subject: [ih] Some comments on the history In-Reply-To: <96292BD2-C6CD-4EC7-87D2-81D9FED48263@cs.ucla.edu> References: <20250126212518.AC40818C079@mercury.lcs.mit.edu> <0405134f-a905-43d1-991c-49aca2c203c8@gmail.com> <52E30F78-6795-4C7D-A7DA-6EA3FC71F964@comcast.net> <96292BD2-C6CD-4EC7-87D2-81D9FED48263@cs.ucla.edu> Message-ID: thanks Len - that's consistent with what I have understood for many years. v On Wed, Jan 29, 2025 at 2:53?AM Leonard Kleinrock wrote: > Vint, Steve, Scott, et al, > > Thank you for adding those useful comments in this chain. I would like to > comment on two of them: > > 1. First, regarding the issue of an early focus on large networks, i.e., > the scaling effect: As you correctly point out, Farouk Kamoun and I > focused strongly on the impact of large networks and the efficiencies to be > obtained in large networks as evidenced in our papers (1976 > > , 1977 > > , 1979 > , > 1980 > > ). > Moreover, my early research was to address the networking issues in large > communication nets; indeed, the opening line of my 1961 PhD thesis > proposal ?Information Flow in Large Communication Nets > ? > is *?The purpose of this thesis is to investigate the problems associated > with information flow in large communication nets.?. *I again referred > to large networks later in the proposal, namely, *?The two important > characteristics of the communication nets that form the subject of this > thesis are (1) the number of nodes in the system is large, and ? ? * > So there we see the recognition that scaling to large networks was a > consideration in the earliest days of network research. > > 2. Second, regarding the history of recognition that packetization of > messages provided many benefits and the discussion as to when Larry Roberts > learned about the advantages of packetization: I introduced the idea of > packetizing messages into short fixed length blocks in my 1962 > > paper and > there I show that the round robin protocol already used in computer > timesharing systems could be adapted to data networks and that motivated me > to provide the first mathematical model of packetization and analyzed the > exact gains to be had for short messages. I clearly pointed out that the > notion of chopping units into fixed length blocks was specifically intended > for messages in a communication network; for example, in my 1963 PhD > dissertation > , I say, > ?*We now explore the manner in which message delay is affected when one > introduces a priority structure (or queue discipline) on the set of > messages ??. *As was pointed out*, *Larry was my officemate at MIT and > we discussed my (and his) research constantly (in fact he wrote a > metacompiler for the program I wrote to simulate communication nets). He > was intimately familiar with my research and he clearly knew about > packetization and its benefits as early as 1961. As a side note, Larry > himself stated *?In order to plan to spend millions of dollars and stake > my reputation, I needed to understand that it would work. Without > Kleinrock's work of Networks and Queuing Theory, I could never have taken > such a radical step? . ?* > > I hope this adds more color and background to the discussion of these > issues. > > Len > > On Jan 28, 2025, at 5:49?AM, Vint Cerf via Internet-history < > internet-history at elists.isoc.org> wrote: > > roberts would also have been aware of Len Kleinrock's queueing theory > analysis of message switching which, mathematically, was not very different > from packet switching. They were both at MIT if memory serves and got their > Ph.D's the same year, 1963. > > v > > > On Tue, Jan 28, 2025 at 7:50?AM John Day wrote: > > Brian, > I agree with you. In the Baran reports, he describes something that sounds > like a datagram. However, he never explores it much other than to define > hot-potato routing. His focus is very centered on survivability and > resilience, which makes sense it was research for the DoD. There is also > the consideration that so far as I have been able to determine, all of the > projects Baran was involved in afterwards were virtual-circuit, as were > Roberts. > > OTOH, NPL didn?t do military research, so I guessed that their impetus for > exploring packet switching had to be different and indeed it was. We have > found a memo Davies wrote (and a similar one by Derek Barber) that Davies > had attended the IFIP Congress in the US in 1965 and heard lots of papers > on timesharing and the time slicing scheduling that timesharing used. The > advantage being that while batch systems did FCFS and short jobs got stuck > behind long ones, timesharut ing interleaved jobs and while short jobs were > still delayed but their *completion* times were shorter. Davies told Derek > that that was what they should do with communications and they did. Their > impetus for packet switching was very different. (But there were two > problems. ;-) 1) Donald got promoted and less time for research, ;-) and 2) > the GPO got involved and the government directed NPL to concentrate on > ?practical? projects, so they had to move to virtual-circuits. > > Scantlebury told Roberts about packet switching at the Gatlinburg > conference and convinced him to use it. When Roberts returned to DC, he > found he had Baran?s reports in a stack of documents but hadn?t read it > yet. Based on the NPL experience Roger also convinced Roberts not to use > 2.4Kbps lines but 50Kbps, which was a large part of the ARPANET success. > (Slower speed would have worked but been so slow people would have said it > wasn?t practical, etc.) > > There is much more to be said about all of this. But that seems to be the > core of it. I find it very interesting how minor events have major effects. > > Take care, > John Day > > On Jan 27, 2025, at 20:47, Brian E Carpenter via Internet-history < > > internet-history at elists.isoc.org> wrote: > > > Vint, and Noel, > > I just glanced through Baran's 1964 paper, and it clearly recognized > statelessnesss (and a standard packet header) as important for network > survivability and adaptive routing. But although he mentions networks > of intercontinental size, I didn't spot any discussion of scalability > as such. > > Interestingly, exactly the same applies to Dave Clark's 1988 "Design > Philosophy" paper. > > In RFC 1958, we did note as principle 3.3 that "All designs must scale > readily to very many nodes per site and to many millions of sites". > I guess that by then (1996) this was too obvious to ignore, and it was > written when IPv4 address exhaustion was considered inevitable. > > Maybe somebody who knows the early literature better than me can find > something. But it's almost as if the intrinsic scalability of stateless > packet switching was an unnoticed and accidental property. > > Regards > Brian > > On 27-Jan-25 11:16, Vint Cerf via Internet-history wrote: > > statelessness was an important design choice and was made consciously so > that paths were not critical to successful transport. For example we did > not want to have to reassemble along a particular path. Even though we > deprecated fragmentation, at the time we thought it was important, we > > did > > not want gateway (router) state to be necessary to accomplish reassembly > regardless of path. I don't know that we recognized the scalability > > aspect > > but we definitely cared a lot about statelessness of the gateways. > v > On Sun, Jan 26, 2025 at 4:25?PM Noel Chiappa via Internet-history < > internet-history at elists.isoc.org> wrote: > > From: Jack Haverty jack at 3kitty.org > > > At the time, the "ARPANET crowd" was skeptical that the > > "datagram" > > nature of TCP could be made to work. Traditional networks, > > including > > the ARPANET, had elaborate internal mechanisms to provide a > > "virtual > > circuit" service to its users. > > > I was thinkking about this, and wondering if internetworking was a more > fundamental advance than the ARPANET (relegating the latter to a > 'ground-breaking experiment'), and I had another thought. > > > Internetworking (following in the track of CYCLADES) made much of the > fate-sharing aspect - that the data needed to ensure reliable > > transmission > > was co-located was the application. One good reason for that (that we > > knew > > at > the time) was that it made the network itself simpler. > > But there's another side to that, one that was even more important, and > which > I'm not sure was obvious to us at the time (1977-79), which is that > > because > > it means the intermediate packet switches in the overall internet > > carry no > > state about the connections travelling through them, there's no scaling > limit. This, to me, has been the single biggest reason why the > > Internet has > > been able to grow to the stupendous size it has. > > I don't think we could have been thinking 'this aspect of lack of > > state in > > the internet packet switches neans it will scale indefinitely', > > because I > > don't think we had any idea, at that point, about how to do path > > selection > > in > a global-scale internet - so global-scale internets could not have > > been in > > our thinking. > > Did that infinite scalability turn out to be just a happy accident, a > side-effect of good fundamental design (but one whose true complete > > value > > wasn't obvious to us at the time), one that moved state out of the > > internet > > packet switches? > > Noel > -- > Internet-history mailing list > Internet-history at elists.isoc.org > https://elists.isoc.org/mailman/listinfo/internet-history > > -- > Internet-history mailing list > Internet-history at elists.isoc.org > https://elists.isoc.org/mailman/listinfo/internet-history > > > > > -- > Please send any postal/overnight deliveries to: > Vint Cerf > Google, LLC > 1900 Reston Metro Plaza, 16th Floor > Reston, VA 20190 > +1 (571) 213 1346 <(571)%20213-1346> > > > until further notice > -- > Internet-history mailing list > Internet-history at elists.isoc.org > https://elists.isoc.org/mailman/listinfo/internet-history > > > -- Please send any postal/overnight deliveries to: Vint Cerf Google, LLC 1900 Reston Metro Plaza, 16th Floor Reston, VA 20190 +1 (571) 213 1346 until further notice From gregskinner0 at icloud.com Wed Jan 29 10:08:07 2025 From: gregskinner0 at icloud.com (Greg Skinner) Date: Wed, 29 Jan 2025 10:08:07 -0800 Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: <7wjzae17he.fsf@junk.nocrew.org> References: <20250128171348.41F9218C073@mercury.lcs.mit.edu> <7wjzae17he.fsf@junk.nocrew.org> Message-ID: <006B15D0-E482-4EA9-B41C-074CA9AB2CE0@icloud.com> On Jan 28, 2025, at 10:04?PM, Lars Brinkhoff via Internet-history wrote: > > Jack Haverty wrote: >> SATNET, and ARPANET, were both continuously managed by the NOC at >> BBN. TCP, and a collection of "core gateways" had become increasingly >> important to users, especially in the UK in Peter Kirstein's group, >> which had to rely on TCP through SATNET to access computers in the US. > > This would have been before 1983, correct? I'm curious which > (important) computers were avaiable through TCP? RFCs 832-839 and 842-848 have results of tests of TCP availability run by Dave Smallberg around the time of the flag day. As for ?important?, that was a matter of opinion. ;) Some people were unhappy about partial or total loss of some of the ?popular" mailing lists such as human-nets, sf-lovers, and unix-wizards when sites running and/or gatewaying those lists were partially or totally unavailable. --gregbo From jack at 3kitty.org Wed Jan 29 10:29:20 2025 From: jack at 3kitty.org (Jack Haverty) Date: Wed, 29 Jan 2025 10:29:20 -0800 Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: References: <20250128171348.41F9218C073@mercury.lcs.mit.edu> <7wjzae17he.fsf@junk.nocrew.org> Message-ID: <61a6c4c8-e2eb-4876-8af7-69c61a72e6cf@3kitty.org> Right!?? I do remember HMP.? I should have said that the early work led to the later definition of SNMP. There was also lots of discussion of "NMP" - "the" Network Management Protocol".?? But it was considered too elaborate, so the "Simple" NMP was defined as a first step.? I never knew whether it was "simple management" or "simple protocol", or intended for managing only "simple networks". Also in 1980+-, I started a project to implement CMCC - the Catenet Monitoring and Control Center - which I recall David Floodpage built.? IIRC, that was what the NOC at BBN used at first to manage the "core gateways".?? See IEN 105.?? That was part of the plan to "make the Internet a 24x7 reliable service". The early 1980s were a time when the Internet mantra of "rough consensus and running code" was part of the culture.? Historians who rely only on the surviving written records contained in RFCs may get a somewhat warped view of how technology actually evolved. In that era, RFCs documented technology that had usually been well vetted in actual use.? An example is the several RFCs that Jon Postel wrote, documenting TCP and IP so that they could be declared DoD Standards.? Jon interviewed all of us implementors to gather the details of how the current "running code" actually worked, and captured it in the RFCs.?? RFCs were usually not proposals; they were documentation of what was actually operating. Email was a fairly new facility then, so much of the technical discussion and argument was carried out in email, much (most?) of which has been lost.? We all liked to use our new toy - the 'net. Prior to email, such history would have likely been captured in papers, letters, conference proceedings, and such traditional means of publication and collaboration. As we "operated" the core Internet, we learned what worked, what didn't work, what was useful, and how to translate the proven ARPANET operating techniques into the world of the Internet. One operations "incident" I still recall occurred in those early days. Mike Brescia was one of the "Gateway Group".? The ARPANET had a mechanism called "traps", which involved IMPs sending reports of anomalous events to the NOC.?? This had been helpful in noticing things like circuits getting high error rates, which foretold failures before the routing mechanisms would react. A similar capability was implemented in the "core gateways", and Mike watched the reports. One day, Mike noticed that some gateways out in the Internet were reporting unusual numbers of IP checksum errors.? Investigation revealed that the errors all involved traffic from one host computer, located somewhere in the Midwest US (Wisconsin perhaps?). So it wasn't likely to be a gateway problem. Mike used the NIC to look up information about that computer, discovered what it was (a PDP-11 IIRC), and looked at the failed datagrams' headers that the core gateway had included in the error reports.? It was a common problem, where the bytes in the 32-bit fields were out of order, leading to checksum failures.? (I had such a problem in the TCP I wrote for PDP-11 Unix too). Mike found the "technical contact" for the site, and sent an email, advising that the TCP they were trying to get working had a bug, and needed to be changed to swap the bytes involved. Shortly afterwards, he got a reply, something like "Hey, thanks! That fixed it." Somewhat later, he got another reply, something like "Hey!? You're in Cambridge, hundreds of miles from me!? How did you do that????!" It wasn't 1984 yet, but Big Brother was already inside The Internet. Fun times, Jack On 1/29/25 01:19, Craig Partridge wrote: > > > On Wed, Jan 29, 2025 at 12:41?AM Jack Haverty via Internet-history > wrote: > > For example, NBS (now NIST) created a program for testing new TCP > implementations to make sure they followed the spec.?? At BBN, in > preparation for the later DDN activity, we set up a service which > would > run the NBS tests for clients (using a dialup link), and then help > them > as consultants to fix whatever wasn't working.?? There was lots of > work > to convert older programs like Telnet, FTP, and mail to use TCP > instead > of NCP, and to get ancillary, but important, technologies such as > SNMP > and ICMP widely implemented. > > > Small nit.? SNMP didn't exist until 1988.? The Internet did not have a > standard management protocol until then. > > What you're probably thinking of is HMP (the Host Monitoring > Protocol), which despite its name, was actually used to monitor the > health of routers.? It was developed c. 1981 (IEN 197) and was > supported on the BBN routers and mailbridges in the 1980s. > > Craig > > > -- > ***** > Craig Partridge's email account for professional society activities > and mailing lists. -------------- next part -------------- A non-text attachment was scrubbed... Name: OpenPGP_signature.asc Type: application/pgp-signature Size: 665 bytes Desc: OpenPGP digital signature URL: From craig at tereschau.net Wed Jan 29 11:24:35 2025 From: craig at tereschau.net (Craig Partridge) Date: Wed, 29 Jan 2025 12:24:35 -0700 Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: <61a6c4c8-e2eb-4876-8af7-69c61a72e6cf@3kitty.org> References: <20250128171348.41F9218C073@mercury.lcs.mit.edu> <7wjzae17he.fsf@junk.nocrew.org> <61a6c4c8-e2eb-4876-8af7-69c61a72e6cf@3kitty.org> Message-ID: I have a similar Brescia story related to HMP. I wrote the HMP implementation for BSD Unix. It was not the server (BSD didn't respond to HMP) but rather a client that could query the gateways. This was for Jil Wescott's ambitious distributed network management system in the mid-1980s (which did a lot of interesting stuff, Ross Callon and Charlie Lynn were the tech leads, and influenced HEMS and thus SNMP). Once I thought I had everything debugged, I got the HMP password (a 16-bit number! good for all routers! we were so naive!) for the routers, plugged it into my code, and launched an HMP query request from my desktop machine (SUN workstation serial number 201 if I remember correctly) at the main BBN router (128.89.0.1 -- don't recall its net 10 address). No answer came back. So I tried again. As I'm watching for a return packet my phone rings -- it's Mike Brescia "Craig are you sending HMP requests to the BBN router?" "Yes MIke, that's me" "You forgot to swap the bytes in the password field." Craig On Wed, Jan 29, 2025 at 11:29?AM Jack Haverty wrote: > Right! I do remember HMP. I should have said that the early work led to > the later definition of SNMP. > > ... > > Mike Brescia was one of the "Gateway Group". The ARPANET had a mechanism > called "traps", which involved IMPs sending reports of anomalous events to > the NOC. This had been helpful in noticing things like circuits getting > high error rates, which foretold failures before the routing mechanisms > would react. > > A similar capability was implemented in the "core gateways", and Mike > watched the reports. > > One day, Mike noticed that some gateways out in the Internet were > reporting unusual numbers of IP checksum errors. Investigation revealed > that the errors all involved traffic from one host computer, located > somewhere in the Midwest US (Wisconsin perhaps?). So it wasn't likely to > be a gateway problem. > > Mike used the NIC to look up information about that computer, discovered > what it was (a PDP-11 IIRC), and looked at the failed datagrams' headers > that the core gateway had included in the error reports. It was a common > problem, where the bytes in the 32-bit fields were out of order, leading to > checksum failures. (I had such a problem in the TCP I wrote for PDP-11 > Unix too). > > Mike found the "technical contact" for the site, and sent an email, > advising that the TCP they were trying to get working had a bug, and needed > to be changed to swap the bytes involved. > > Shortly afterwards, he got a reply, something like "Hey, thanks! That > fixed it." > > Somewhat later, he got another reply, something like "Hey! You're in > Cambridge, hundreds of miles from me! How did you do that????!" > > It wasn't 1984 yet, but Big Brother was already inside The Internet. > > Fun times, > Jack > > > On 1/29/25 01:19, Craig Partridge wrote: > > > > On Wed, Jan 29, 2025 at 12:41?AM Jack Haverty via Internet-history < > internet-history at elists.isoc.org> wrote: > >> For example, NBS (now NIST) created a program for testing new TCP >> implementations to make sure they followed the spec. At BBN, in >> preparation for the later DDN activity, we set up a service which would >> run the NBS tests for clients (using a dialup link), and then help them >> as consultants to fix whatever wasn't working. There was lots of work >> to convert older programs like Telnet, FTP, and mail to use TCP instead >> of NCP, and to get ancillary, but important, technologies such as SNMP >> and ICMP widely implemented. >> >> > Small nit. SNMP didn't exist until 1988. The Internet did not have a > standard management protocol until then. > > What you're probably thinking of is HMP (the Host Monitoring Protocol), > which despite its name, was actually used to monitor the health of > routers. It was developed c. 1981 (IEN 197) and was supported on the BBN > routers and mailbridges in the 1980s. > > Craig > > > -- > ***** > Craig Partridge's email account for professional society activities and > mailing lists. > > > -- ***** Craig Partridge's email account for professional society activities and mailing lists. From vgcerf at gmail.com Wed Jan 29 11:27:32 2025 From: vgcerf at gmail.com (vinton cerf) Date: Wed, 29 Jan 2025 14:27:32 -0500 Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: References: <20250128171348.41F9218C073@mercury.lcs.mit.edu> <7wjzae17he.fsf@junk.nocrew.org> <61a6c4c8-e2eb-4876-8af7-69c61a72e6cf@3kitty.org> Message-ID: great story!!! v On Wed, Jan 29, 2025 at 2:24?PM Craig Partridge via Internet-history < internet-history at elists.isoc.org> wrote: > I have a similar Brescia story related to HMP. > > I wrote the HMP implementation for BSD Unix. It was not the server (BSD > didn't respond to HMP) but rather a client that could query the gateways. > This was for Jil Wescott's ambitious distributed network management system > in the mid-1980s (which did a lot of interesting stuff, Ross Callon and > Charlie Lynn were the tech leads, and influenced HEMS and thus SNMP). > > Once I thought I had everything debugged, I got the HMP password (a 16-bit > number! good for all routers! we were so naive!) for the routers, plugged > it into my code, and launched an HMP query request from my desktop machine > (SUN workstation serial number 201 if I remember correctly) at the main BBN > router (128.89.0.1 -- don't recall its net 10 address). No answer came > back. So I tried again. As I'm watching for a return packet my phone > rings -- it's Mike Brescia "Craig are you sending HMP requests to the BBN > router?" "Yes MIke, that's me" "You forgot to swap the bytes in the > password field." > > Craig > > On Wed, Jan 29, 2025 at 11:29?AM Jack Haverty wrote: > > > Right! I do remember HMP. I should have said that the early work led > to > > the later definition of SNMP. > > > > ... > > > > Mike Brescia was one of the "Gateway Group". The ARPANET had a mechanism > > called "traps", which involved IMPs sending reports of anomalous events > to > > the NOC. This had been helpful in noticing things like circuits getting > > high error rates, which foretold failures before the routing mechanisms > > would react. > > > > A similar capability was implemented in the "core gateways", and Mike > > watched the reports. > > > > One day, Mike noticed that some gateways out in the Internet were > > reporting unusual numbers of IP checksum errors. Investigation revealed > > that the errors all involved traffic from one host computer, located > > somewhere in the Midwest US (Wisconsin perhaps?). So it wasn't likely to > > be a gateway problem. > > > > Mike used the NIC to look up information about that computer, discovered > > what it was (a PDP-11 IIRC), and looked at the failed datagrams' headers > > that the core gateway had included in the error reports. It was a common > > problem, where the bytes in the 32-bit fields were out of order, leading > to > > checksum failures. (I had such a problem in the TCP I wrote for PDP-11 > > Unix too). > > > > Mike found the "technical contact" for the site, and sent an email, > > advising that the TCP they were trying to get working had a bug, and > needed > > to be changed to swap the bytes involved. > > > > Shortly afterwards, he got a reply, something like "Hey, thanks! That > > fixed it." > > > > Somewhat later, he got another reply, something like "Hey! You're in > > Cambridge, hundreds of miles from me! How did you do that????!" > > > > It wasn't 1984 yet, but Big Brother was already inside The Internet. > > > > Fun times, > > Jack > > > > > > On 1/29/25 01:19, Craig Partridge wrote: > > > > > > > > On Wed, Jan 29, 2025 at 12:41?AM Jack Haverty via Internet-history < > > internet-history at elists.isoc.org> wrote: > > > >> For example, NBS (now NIST) created a program for testing new TCP > >> implementations to make sure they followed the spec. At BBN, in > >> preparation for the later DDN activity, we set up a service which would > >> run the NBS tests for clients (using a dialup link), and then help them > >> as consultants to fix whatever wasn't working. There was lots of work > >> to convert older programs like Telnet, FTP, and mail to use TCP instead > >> of NCP, and to get ancillary, but important, technologies such as SNMP > >> and ICMP widely implemented. > >> > >> > > Small nit. SNMP didn't exist until 1988. The Internet did not have a > > standard management protocol until then. > > > > What you're probably thinking of is HMP (the Host Monitoring Protocol), > > which despite its name, was actually used to monitor the health of > > routers. It was developed c. 1981 (IEN 197) and was supported on the BBN > > routers and mailbridges in the 1980s. > > > > Craig > > > > > > -- > > ***** > > Craig Partridge's email account for professional society activities and > > mailing lists. > > > > > > > > -- > ***** > Craig Partridge's email account for professional society activities and > mailing lists. > -- > Internet-history mailing list > Internet-history at elists.isoc.org > https://elists.isoc.org/mailman/listinfo/internet-history > From jnc at mercury.lcs.mit.edu Wed Jan 29 14:25:50 2025 From: jnc at mercury.lcs.mit.edu (Noel Chiappa) Date: Wed, 29 Jan 2025 17:25:50 -0500 (EST) Subject: [ih] Anyone have a copy of IEN-33? Message-ID: <20250129222550.530C818C074@mercury.lcs.mit.edu> The IEN repository at the RFC Editor is missing this one; I ask because the IEN's for the minutes of the meetings are all there, except this one. (Bennett, "Internet Meeting Notes - 1&2 May 1978). Not super-important (compared to say, the email archive of the TCP/IP email list at the time, which I think was at DARPA - but maybe at a contractor), but it would be nice to have. Oh, I've found some evidence that I wasn't the only one who called that group the 'Internet Working Group' (r.e. my original query); I don't remember all the instances I've seen, but IEN-60 is entitled "Boston Area Meeting of the Internet Working Group to Discuss Interactions With Gateways". Noel PS: Craig's email reminds me of another password story. Proteon put a field service password in the Proteon routers. So Milo gets the load, and thinks 'Gee, I should try running 'strings' on this'; he does so, and see an odd string (near the 'Password:' prompt, IIRC). He tries it, and it lets him in. He complains. So in the _next_ release, he runs 'strings' over it, and sees the string 'Sorry Milo, it's not so easy this time!' From jack at 3kitty.org Wed Jan 29 16:57:06 2025 From: jack at 3kitty.org (Jack Haverty) Date: Wed, 29 Jan 2025 16:57:06 -0800 Subject: [ih] Anyone have a copy of IEN-33? In-Reply-To: <20250129222550.530C818C074@mercury.lcs.mit.edu> References: <20250129222550.530C818C074@mercury.lcs.mit.edu> Message-ID: <4779e690-9f82-4e8d-8469-b82166228a21@3kitty.org> My book of IENs, collected as they were issued, is also missing IEN 33.? Have you seen any evidence that IEN 33 was ever actually released as an IEN??? I know there were some RFCs and/or IENs that were assigned numbers but never actually produced.?? I authored one RFC in the 700s myself but never actually wrote it. I DO have a copy of "TCP Meeting Notes" sent by Jon Postel to the email address [ISIE]TCP-INTERNET.List?? It documents the meeting of 15-16 June 1978 held at MIT.? There's no indication that it was also released as an IEN.?? The email does say that the file is (was) available online at TCP-MEETING-NOTES.TXT at ISIE. It looks like the typical meeting report of the time, containing a record of Vint's goals for the meeting, followed by status reports from each contractor. Among Vint's goals: "The format of the TCP and INTERNET headers is to be firmly decided at this meeting", "The schedule for implementation of version 4 is to be established.", "The schedule for Telnet and FTP running on TCP is to be established.", and "the whole ARPANET community should expect to move to using TCP". So the planning for the eventual 1/1/1983 Flag Day started sometime before June 1978. Vint's introduction was followed by discussions of various topics, and even some votes: "Shall the Port be part of the Internet Header?" Result: NO. "Shall the Port be part of the TCP Header?" Result: NO. The process of evolving from TCP 2 to TCP 4 resembled sausage making....achieving consensus wasn't easy. I also have the notes I took at the meeting.? Big meetings happened quarterly, so it seems unlikely that there was a meeting in mid-June following one only 6 weeks earlier.? Perhaps the May 1-2 meeting was cancelled and rescheduled into mid June? Jack PS - I don't recall anything being kept "at ARPA"; most stuff was kept at contractors' sites, often SRI or ISI. On 1/29/25 14:25, Noel Chiappa via Internet-history wrote: > The IEN repository at the RFC Editor is missing this one; I ask because the > IEN's for the minutes of the meetings are all there, except this one. > (Bennett, "Internet Meeting Notes - 1&2 May 1978). Not super-important > (compared to say, the email archive of the TCP/IP email list at the time, > which I think was at DARPA - but maybe at a contractor), but it would be nice > to have. > > Oh, I've found some evidence that I wasn't the only one who called that group > the 'Internet Working Group' (r.e. my original query); I don't remember all > the instances I've seen, but IEN-60 is entitled "Boston Area Meeting of the > Internet Working Group to Discuss Interactions With Gateways". > > Noel > > PS: Craig's email reminds me of another password story. Proteon put a field > service password in the Proteon routers. So Milo gets the load, and thinks > 'Gee, I should try running 'strings' on this'; he does so, and see an odd > string (near the 'Password:' prompt, IIRC). He tries it, and it lets him in. > He complains. So in the _next_ release, he runs 'strings' over it, and sees > the string 'Sorry Milo, it's not so easy this time!' -------------- next part -------------- A non-text attachment was scrubbed... Name: OpenPGP_signature.asc Type: application/pgp-signature Size: 665 bytes Desc: OpenPGP digital signature URL: From vint at google.com Wed Jan 29 17:02:58 2025 From: vint at google.com (Vint Cerf) Date: Wed, 29 Jan 2025 20:02:58 -0500 Subject: [ih] Anyone have a copy of IEN-33? In-Reply-To: <4779e690-9f82-4e8d-8469-b82166228a21@3kitty.org> References: <20250129222550.530C818C074@mercury.lcs.mit.edu> <4779e690-9f82-4e8d-8469-b82166228a21@3kitty.org> Message-ID: thanks Jack - little was kept at ARPA except for ARPA Orders. I don't have the IENs - stupid me for not collecting them all religiously. v On Wed, Jan 29, 2025 at 7:57?PM Jack Haverty via Internet-history < internet-history at elists.isoc.org> wrote: > My book of IENs, collected as they were issued, is also missing IEN 33. > Have you seen any evidence that IEN 33 was ever actually released as an > IEN? I know there were some RFCs and/or IENs that were assigned > numbers but never actually produced. I authored one RFC in the 700s > myself but never actually wrote it. > > I DO have a copy of "TCP Meeting Notes" sent by Jon Postel to the email > address [ISIE]TCP-INTERNET.List It documents the meeting of > 15-16 June 1978 held at MIT. There's no indication that it was also > released as an IEN. The email does say that the file is (was) > available online at TCP-MEETING-NOTES.TXT at ISIE. > > It looks like the typical meeting report of the time, containing a > record of Vint's goals for the meeting, followed by status reports from > each contractor. > > Among Vint's goals: "The format of the TCP and INTERNET headers is to be > firmly decided at this meeting", "The schedule for implementation of > version 4 is to be established.", "The schedule for Telnet and FTP > running on TCP is to be established.", and "the whole ARPANET community > should expect to move to using TCP". > > So the planning for the eventual 1/1/1983 Flag Day started sometime > before June 1978. > > Vint's introduction was followed by discussions of various topics, and > even some votes: > > "Shall the Port be part of the Internet Header?" > Result: NO. > "Shall the Port be part of the TCP Header?" > Result: NO. > > The process of evolving from TCP 2 to TCP 4 resembled sausage > making....achieving consensus wasn't easy. > > I also have the notes I took at the meeting. Big meetings happened > quarterly, so it seems unlikely that there was a meeting in mid-June > following one only 6 weeks earlier. Perhaps the May 1-2 meeting was > cancelled and rescheduled into mid June? > > Jack > > PS - I don't recall anything being kept "at ARPA"; most stuff was kept > at contractors' sites, often SRI or ISI. > > On 1/29/25 14:25, Noel Chiappa via Internet-history wrote: > > The IEN repository at the RFC Editor is missing this one; I ask because > the > > IEN's for the minutes of the meetings are all there, except this one. > > (Bennett, "Internet Meeting Notes - 1&2 May 1978). Not super-important > > (compared to say, the email archive of the TCP/IP email list at the time, > > which I think was at DARPA - but maybe at a contractor), but it would be > nice > > to have. > > > > Oh, I've found some evidence that I wasn't the only one who called that > group > > the 'Internet Working Group' (r.e. my original query); I don't remember > all > > the instances I've seen, but IEN-60 is entitled "Boston Area Meeting of > the > > Internet Working Group to Discuss Interactions With Gateways". > > > > Noel > > > > PS: Craig's email reminds me of another password story. Proteon put a > field > > service password in the Proteon routers. So Milo gets the load, and > thinks > > 'Gee, I should try running 'strings' on this'; he does so, and see an odd > > string (near the 'Password:' prompt, IIRC). He tries it, and it lets him > in. > > He complains. So in the _next_ release, he runs 'strings' over it, and > sees > > the string 'Sorry Milo, it's not so easy this time!' > > -- > Internet-history mailing list > Internet-history at elists.isoc.org > https://elists.isoc.org/mailman/listinfo/internet-history > -- Please send any postal/overnight deliveries to: Vint Cerf Google, LLC 1900 Reston Metro Plaza, 16th Floor Reston, VA 20190 +1 (571) 213 1346 until further notice From touch at strayalpha.com Wed Jan 29 20:16:32 2025 From: touch at strayalpha.com (touch at strayalpha.com) Date: Wed, 29 Jan 2025 20:16:32 -0800 Subject: [ih] Anyone have a copy of IEN-33? In-Reply-To: References: <20250129222550.530C818C074@mercury.lcs.mit.edu> <4779e690-9f82-4e8d-8469-b82166228a21@3kitty.org> Message-ID: <161181D9-4677-4D86-9944-16F6C8DAF458@strayalpha.com> Hi, all, I had them scanned for the USC/ISI Postel Center (when such a thing existed; its website has bit-rot). #33 has been missing as long as I have been looking (roughly early 1990s) in every compilation I have checked. If anyone has it, please let me know. For #9, 125, and 126, I have text but the paper could not be scanned at the time (I don?t recall why - might have had severe contrast issues). If anyone has a scan of those, please let me know. I will be re-posting my archive of these at my website (strayalpha.com) shortly. Joe ? Dr. Joe Touch, temporal epistemologist www.strayalpha.com > On Jan 29, 2025, at 5:02?PM, Vint Cerf via Internet-history wrote: > > thanks Jack - little was kept at ARPA except for ARPA Orders. > I don't have the IENs - stupid me for not collecting them all religiously. > > v > > > On Wed, Jan 29, 2025 at 7:57?PM Jack Haverty via Internet-history < > internet-history at elists.isoc.org> wrote: > >> My book of IENs, collected as they were issued, is also missing IEN 33. >> Have you seen any evidence that IEN 33 was ever actually released as an >> IEN? I know there were some RFCs and/or IENs that were assigned >> numbers but never actually produced. I authored one RFC in the 700s >> myself but never actually wrote it. >> >> I DO have a copy of "TCP Meeting Notes" sent by Jon Postel to the email >> address [ISIE]TCP-INTERNET.List It documents the meeting of >> 15-16 June 1978 held at MIT. There's no indication that it was also >> released as an IEN. The email does say that the file is (was) >> available online at TCP-MEETING-NOTES.TXT at ISIE. >> >> It looks like the typical meeting report of the time, containing a >> record of Vint's goals for the meeting, followed by status reports from >> each contractor. >> >> Among Vint's goals: "The format of the TCP and INTERNET headers is to be >> firmly decided at this meeting", "The schedule for implementation of >> version 4 is to be established.", "The schedule for Telnet and FTP >> running on TCP is to be established.", and "the whole ARPANET community >> should expect to move to using TCP". >> >> So the planning for the eventual 1/1/1983 Flag Day started sometime >> before June 1978. >> >> Vint's introduction was followed by discussions of various topics, and >> even some votes: >> >> "Shall the Port be part of the Internet Header?" >> Result: NO. >> "Shall the Port be part of the TCP Header?" >> Result: NO. >> >> The process of evolving from TCP 2 to TCP 4 resembled sausage >> making....achieving consensus wasn't easy. >> >> I also have the notes I took at the meeting. Big meetings happened >> quarterly, so it seems unlikely that there was a meeting in mid-June >> following one only 6 weeks earlier. Perhaps the May 1-2 meeting was >> cancelled and rescheduled into mid June? >> >> Jack >> >> PS - I don't recall anything being kept "at ARPA"; most stuff was kept >> at contractors' sites, often SRI or ISI. >> >> On 1/29/25 14:25, Noel Chiappa via Internet-history wrote: >>> The IEN repository at the RFC Editor is missing this one; I ask because >> the >>> IEN's for the minutes of the meetings are all there, except this one. >>> (Bennett, "Internet Meeting Notes - 1&2 May 1978). Not super-important >>> (compared to say, the email archive of the TCP/IP email list at the time, >>> which I think was at DARPA - but maybe at a contractor), but it would be >> nice >>> to have. >>> >>> Oh, I've found some evidence that I wasn't the only one who called that >> group >>> the 'Internet Working Group' (r.e. my original query); I don't remember >> all >>> the instances I've seen, but IEN-60 is entitled "Boston Area Meeting of >> the >>> Internet Working Group to Discuss Interactions With Gateways". >>> >>> Noel >>> >>> PS: Craig's email reminds me of another password story. Proteon put a >> field >>> service password in the Proteon routers. So Milo gets the load, and >> thinks >>> 'Gee, I should try running 'strings' on this'; he does so, and see an odd >>> string (near the 'Password:' prompt, IIRC). He tries it, and it lets him >> in. >>> He complains. So in the _next_ release, he runs 'strings' over it, and >> sees >>> the string 'Sorry Milo, it's not so easy this time!' >> >> -- >> Internet-history mailing list >> Internet-history at elists.isoc.org >> https://elists.isoc.org/mailman/listinfo/internet-history >> > > > -- > Please send any postal/overnight deliveries to: > Vint Cerf > Google, LLC > 1900 Reston Metro Plaza, 16th Floor > Reston, VA 20190 > +1 (571) 213 1346 > > > until further notice > -- > Internet-history mailing list > Internet-history at elists.isoc.org > https://elists.isoc.org/mailman/listinfo/internet-history From jack at 3kitty.org Wed Jan 29 20:55:12 2025 From: jack at 3kitty.org (Jack Haverty) Date: Wed, 29 Jan 2025 20:55:12 -0800 Subject: [ih] Anyone have a copy of IEN-33? In-Reply-To: <161181D9-4677-4D86-9944-16F6C8DAF458@strayalpha.com> References: <20250129222550.530C818C074@mercury.lcs.mit.edu> <4779e690-9f82-4e8d-8469-b82166228a21@3kitty.org> <161181D9-4677-4D86-9944-16F6C8DAF458@strayalpha.com> Message-ID: <1ed9575d-2805-45d1-9af5-8123e7c3c8ad@3kitty.org> Joe, There were apparently lots of materials kept in computer files, such as the directory Jon's notes mention.? Did ISI keep backups on magtape??? If so, those old directories may still be recoverable and may contain historical records. MIT kept backup tapes back from the 1970s and recently (last ten years or so) made them available to researchers and/or their original owners.? Lars and the ITS-Hackers group have made good use of the MIT tapes to resurrect ancient PDP-10 code and actually get it running again. If ISI and/or SRI similarly kept backup tapes from half-a-century ago, they might be a useful resource for historians.?? ISIA was where many ARPA people had their mailboxes; SRI had all of the NIC materials. But we still have the problem of where to put such stuff for "perpetual" archival storage..... Jack On 1/29/25 20:16, touch at strayalpha.com wrote: > Hi, all, > > I had them scanned for the USC/ISI Postel Center (when such a thing > existed; its website has bit-rot). > > #33 has been missing as long as I have been looking (roughly early > 1990s) in every compilation I have checked. If anyone has it, please > let me know. > > For #9, 125, and 126, I have text but the paper could not be scanned > at the time (I don?t recall why - might have had severe contrast > issues). If anyone has a scan of those, please let me know. > > I will be re-posting my archive of these at my website > (strayalpha.com) shortly. > > Joe > > ? > Dr. Joe Touch, temporal epistemologist > www.strayalpha.com > >> On Jan 29, 2025, at 5:02?PM, Vint Cerf via Internet-history >> wrote: >> >> thanks Jack - little was kept at ARPA except for ARPA Orders. >> I don't have the IENs - stupid me for not collecting them all >> religiously. >> >> v >> >> >> On Wed, Jan 29, 2025 at 7:57?PM Jack Haverty via Internet-history < >> internet-history at elists.isoc.org> wrote: >> >>> My book of IENs, collected as they were issued, is also missing IEN 33. >>> Have you seen any evidence that IEN 33 was ever actually released as an >>> IEN? ??I know there were some RFCs and/or IENs that were assigned >>> numbers but never actually produced. ??I authored one RFC in the 700s >>> myself but never actually wrote it. >>> >>> I DO have a copy of "TCP Meeting Notes" sent by Jon Postel to the email >>> address [ISIE]TCP-INTERNET.List ??It documents the meeting of >>> 15-16 June 1978 held at MIT. ?There's no indication that it was also >>> released as an IEN. ??The email does say that the file is (was) >>> available online at TCP-MEETING-NOTES.TXT at ISIE. >>> >>> It looks like the typical meeting report of the time, containing a >>> record of Vint's goals for the meeting, followed by status reports from >>> each contractor. >>> >>> Among Vint's goals: "The format of the TCP and INTERNET headers is to be >>> firmly decided at this meeting", "The schedule for implementation of >>> version 4 is to be established.", "The schedule for Telnet and FTP >>> running on TCP is to be established.", and "the whole ARPANET community >>> should expect to move to using TCP". >>> >>> So the planning for the eventual 1/1/1983 Flag Day started sometime >>> before June 1978. >>> >>> Vint's introduction was followed by discussions of various topics, and >>> even some votes: >>> >>> "Shall the Port be part of the Internet Header?" >>> Result: NO. >>> "Shall the Port be part of the TCP Header?" >>> Result: NO. >>> >>> The process of evolving from TCP 2 to TCP 4 resembled sausage >>> making....achieving consensus wasn't easy. >>> >>> I also have the notes I took at the meeting. ?Big meetings happened >>> quarterly, so it seems unlikely that there was a meeting in mid-June >>> following one only 6 weeks earlier. ?Perhaps the May 1-2 meeting was >>> cancelled and rescheduled into mid June? >>> >>> Jack >>> >>> PS - I don't recall anything being kept "at ARPA"; most stuff was kept >>> at contractors' sites, often SRI or ISI. >>> >>> On 1/29/25 14:25, Noel Chiappa via Internet-history wrote: >>>> The IEN repository at the RFC Editor is missing this one; I ask because >>> the >>>> IEN's for the minutes of the meetings are all there, except this one. >>>> (Bennett, "Internet Meeting Notes - 1&2 May 1978). Not super-important >>>> (compared to say, the email archive of the TCP/IP email list at the >>>> time, >>>> which I think was at DARPA - but maybe at a contractor), but it >>>> would be >>> nice >>>> to have. >>>> >>>> Oh, I've found some evidence that I wasn't the only one who called that >>> group >>>> the 'Internet Working Group' (r.e. my original query); I don't remember >>> all >>>> the instances I've seen, but IEN-60 is entitled "Boston Area Meeting of >>> the >>>> Internet Working Group to Discuss Interactions With Gateways". >>>> >>>> ?????Noel >>>> >>>> PS: Craig's email reminds me of another password story. Proteon put a >>> field >>>> service password in the Proteon routers. So Milo gets the load, and >>> thinks >>>> 'Gee, I should try running 'strings' on this'; he does so, and see >>>> an odd >>>> string (near the 'Password:' prompt, IIRC). He tries it, and it >>>> lets him >>> in. >>>> He complains. So in the _next_ release, he runs 'strings' over it, and >>> sees >>>> the string 'Sorry Milo, it's not so easy this time!' >>> >>> -- >>> Internet-history mailing list >>> Internet-history at elists.isoc.org >>> https://elists.isoc.org/mailman/listinfo/internet-history >>> >> >> >> -- >> Please send any postal/overnight deliveries to: >> Vint Cerf >> Google, LLC >> 1900 Reston Metro Plaza, 16th Floor >> Reston, VA 20190 >> +1 (571) 213 1346 >> >> >> until further notice >> -- >> Internet-history mailing list >> Internet-history at elists.isoc.org >> https://elists.isoc.org/mailman/listinfo/internet-history > -------------- next part -------------- A non-text attachment was scrubbed... Name: OpenPGP_signature.asc Type: application/pgp-signature Size: 665 bytes Desc: OpenPGP digital signature URL: From touch at strayalpha.com Wed Jan 29 22:10:55 2025 From: touch at strayalpha.com (touch at strayalpha.com) Date: Wed, 29 Jan 2025 22:10:55 -0800 Subject: [ih] Anyone have a copy of IEN-33? In-Reply-To: <1ed9575d-2805-45d1-9af5-8123e7c3c8ad@3kitty.org> References: <20250129222550.530C818C074@mercury.lcs.mit.edu> <4779e690-9f82-4e8d-8469-b82166228a21@3kitty.org> <161181D9-4677-4D86-9944-16F6C8DAF458@strayalpha.com> <1ed9575d-2805-45d1-9af5-8123e7c3c8ad@3kitty.org> Message-ID: <51893DC6-9A61-4F75-A0CF-0C74481855CB@strayalpha.com> Hi, Jack, > On Jan 29, 2025, at 8:55?PM, Jack Haverty wrote: > > Joe, > > There were apparently lots of materials kept in computer files, such as the directory Jon's notes mention. Did ISI keep backups on magtape? If so, those old directories may still be recoverable and may contain historical records. They did not. I know because I had tried to retrieve the archive of the E2E-interest list after a virus took out active system and found out there were no backups and they had discarded the tapes from the previous system (TOPS-20?), which is why the online repository goes back only to 2001. They did the same with their entire tech report library in 2006. NB: if anyone here cares about the stuff at postel.org, I encourage you to grab it and back it up somewhere. I moved the entire Internet-history list - archives and all - here exactly so it would be more stable. > MIT kept backup tapes back from the 1970s and recently (last ten years or so) made them available to researchers and/or their original owners. Lars and the ITS-Hackers group have made good use of the MIT tapes to resurrect ancient PDP-10 code and actually get it running again. I?m aware; Rick Shiffman has been busy lately restoring muddle to operational status from those tapes. > If ISI and/or SRI similarly kept backup tapes from half-a-century ago, they might be a useful resource for historians. ISIA was where many ARPA people had their mailboxes; SRI had all of the NIC materials. > > But we still have the problem of where to put such stuff for "perpetual" archival storage..... Yes, that?s an additional problem. Joe > > Jack > > On 1/29/25 20:16, touch at strayalpha.com wrote: >> Hi, all, >> >> I had them scanned for the USC/ISI Postel Center (when such a thing existed; its website has bit-rot). >> >> #33 has been missing as long as I have been looking (roughly early 1990s) in every compilation I have checked. If anyone has it, please let me know. >> >> For #9, 125, and 126, I have text but the paper could not be scanned at the time (I don?t recall why - might have had severe contrast issues). If anyone has a scan of those, please let me know. >> >> I will be re-posting my archive of these at my website (strayalpha.com) shortly. >> >> Joe >> >> ? >> Dr. Joe Touch, temporal epistemologist >> www.strayalpha.com >> >>> On Jan 29, 2025, at 5:02?PM, Vint Cerf via Internet-history wrote: >>> >>> thanks Jack - little was kept at ARPA except for ARPA Orders. >>> I don't have the IENs - stupid me for not collecting them all religiously. >>> >>> v >>> >>> >>> On Wed, Jan 29, 2025 at 7:57?PM Jack Haverty via Internet-history < >>> internet-history at elists.isoc.org > wrote: >>> >>>> My book of IENs, collected as they were issued, is also missing IEN 33. >>>> Have you seen any evidence that IEN 33 was ever actually released as an >>>> IEN? I know there were some RFCs and/or IENs that were assigned >>>> numbers but never actually produced. I authored one RFC in the 700s >>>> myself but never actually wrote it. >>>> >>>> I DO have a copy of "TCP Meeting Notes" sent by Jon Postel to the email >>>> address [ISIE]TCP-INTERNET.List It documents the meeting of >>>> 15-16 June 1978 held at MIT. There's no indication that it was also >>>> released as an IEN. The email does say that the file is (was) >>>> available online at TCP-MEETING-NOTES.TXT at ISIE. >>>> >>>> It looks like the typical meeting report of the time, containing a >>>> record of Vint's goals for the meeting, followed by status reports from >>>> each contractor. >>>> >>>> Among Vint's goals: "The format of the TCP and INTERNET headers is to be >>>> firmly decided at this meeting", "The schedule for implementation of >>>> version 4 is to be established.", "The schedule for Telnet and FTP >>>> running on TCP is to be established.", and "the whole ARPANET community >>>> should expect to move to using TCP". >>>> >>>> So the planning for the eventual 1/1/1983 Flag Day started sometime >>>> before June 1978. >>>> >>>> Vint's introduction was followed by discussions of various topics, and >>>> even some votes: >>>> >>>> "Shall the Port be part of the Internet Header?" >>>> Result: NO. >>>> "Shall the Port be part of the TCP Header?" >>>> Result: NO. >>>> >>>> The process of evolving from TCP 2 to TCP 4 resembled sausage >>>> making....achieving consensus wasn't easy. >>>> >>>> I also have the notes I took at the meeting. Big meetings happened >>>> quarterly, so it seems unlikely that there was a meeting in mid-June >>>> following one only 6 weeks earlier. Perhaps the May 1-2 meeting was >>>> cancelled and rescheduled into mid June? >>>> >>>> Jack >>>> >>>> PS - I don't recall anything being kept "at ARPA"; most stuff was kept >>>> at contractors' sites, often SRI or ISI. >>>> >>>> On 1/29/25 14:25, Noel Chiappa via Internet-history wrote: >>>>> The IEN repository at the RFC Editor is missing this one; I ask because >>>> the >>>>> IEN's for the minutes of the meetings are all there, except this one. >>>>> (Bennett, "Internet Meeting Notes - 1&2 May 1978). Not super-important >>>>> (compared to say, the email archive of the TCP/IP email list at the time, >>>>> which I think was at DARPA - but maybe at a contractor), but it would be >>>> nice >>>>> to have. >>>>> >>>>> Oh, I've found some evidence that I wasn't the only one who called that >>>> group >>>>> the 'Internet Working Group' (r.e. my original query); I don't remember >>>> all >>>>> the instances I've seen, but IEN-60 is entitled "Boston Area Meeting of >>>> the >>>>> Internet Working Group to Discuss Interactions With Gateways". >>>>> >>>>> Noel >>>>> >>>>> PS: Craig's email reminds me of another password story. Proteon put a >>>> field >>>>> service password in the Proteon routers. So Milo gets the load, and >>>> thinks >>>>> 'Gee, I should try running 'strings' on this'; he does so, and see an odd >>>>> string (near the 'Password:' prompt, IIRC). He tries it, and it lets him >>>> in. >>>>> He complains. So in the _next_ release, he runs 'strings' over it, and >>>> sees >>>>> the string 'Sorry Milo, it's not so easy this time!' >>>> >>>> -- >>>> Internet-history mailing list >>>> Internet-history at elists.isoc.org >>>> https://elists.isoc.org/mailman/listinfo/internet-history >>>> >>> >>> >>> -- >>> Please send any postal/overnight deliveries to: >>> Vint Cerf >>> Google, LLC >>> 1900 Reston Metro Plaza, 16th Floor >>> Reston, VA 20190 >>> +1 (571) 213 1346 >>> >>> >>> until further notice >>> -- >>> Internet-history mailing list >>> Internet-history at elists.isoc.org >>> https://elists.isoc.org/mailman/listinfo/internet-history >> > > From lars at nocrew.org Wed Jan 29 22:41:21 2025 From: lars at nocrew.org (Lars Brinkhoff) Date: Thu, 30 Jan 2025 06:41:21 +0000 Subject: [ih] Anyone have a copy of IEN-33? In-Reply-To: <1ed9575d-2805-45d1-9af5-8123e7c3c8ad@3kitty.org> (Jack Haverty via Internet-history's message of "Wed, 29 Jan 2025 20:55:12 -0800") References: <20250129222550.530C818C074@mercury.lcs.mit.edu> <4779e690-9f82-4e8d-8469-b82166228a21@3kitty.org> <161181D9-4677-4D86-9944-16F6C8DAF458@strayalpha.com> <1ed9575d-2805-45d1-9af5-8123e7c3c8ad@3kitty.org> Message-ID: <7w1pwk248e.fsf@junk.nocrew.org> Jack Haverty wrote: > If ISI and/or SRI similarly kept backup tapes from half-a-century ago, > they might be a useful resource for historians.?? ISIA was where many > ARPA people had their mailboxes; SRI had all of the NIC materials. There are some tapes from SRI-NIC: https://bitsavers.org/bits/DEC/pdp10/sri-nic/ https://github.com/PDP-10/sri-nic/tree/master/files But it's just binaries and source code for programs, no email archives or the like. I believe there are tapes from SRI-ARC as well, but not available to the public. As Joe noted, ISI has been particularly industrious in eradicating their own historical materials. From lars at nocrew.org Thu Jan 30 04:58:15 2025 From: lars at nocrew.org (Lars Brinkhoff) Date: Thu, 30 Jan 2025 12:58:15 +0000 Subject: [ih] Some random thoughts about ICCC 72 In-Reply-To: <7w7c6i2fvi.fsf@junk.nocrew.org> (Lars Brinkhoff via Internet-history's message of "Sun, 26 Jan 2025 07:28:33 +0000") References: <7w7c6i2fvi.fsf@junk.nocrew.org> Message-ID: <7wo6zozcew.fsf@junk.nocrew.org> Lars Brinkhoff > Alexander McKenzie wrote: >> Bob and AL had arranged to have a film made that would explain the >> ARPAnet and why ARPA had built it > https://www.youtube.com/watch?v=Cdf5fMAL994 Here's a new one, on a much smaller scale in every sense. "Scenarios for using Arpanet Computers in 2025" https://youtu.be/FQnE-nAxzmY From b_a_denny at yahoo.com Thu Jan 30 21:25:41 2025 From: b_a_denny at yahoo.com (Barbara Denny) Date: Fri, 31 Jan 2025 05:25:41 +0000 (UTC) Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: <1314488472.5766718.1738137430770@mail.yahoo.com> References: <20250128171348.41F9218C073@mercury.lcs.mit.edu> <7wjzae17he.fsf@junk.nocrew.org> <1314488472.5766718.1738137430770@mail.yahoo.com> Message-ID: <1776225359.6704266.1738301141392@mail.yahoo.com> Found a better copy of the network diagrams so you can see both TCP and NCP shown in the ARPAnet in 1976. https://ed-thelen.org/comp-hist/CORE-3-1-SRI-TCP-IP.html When I have more time, I will see if I can find the SRI progress report that covers this activity. barbara On Tuesday, January 28, 2025 at 11:59:57 PM PST, Barbara Denny via Internet-history wrote: TCP existed prior to 1983 in hosts attached to the various networks that comprised the Internet.? ARPAnet was just one of the networks on the Internet during this time period. Flag day marked the day when NCP would no longer be supported in the ARPAnet. Hosts attached to the ARPAnet needed to use TCP or they would essentially be disconnected. Perhaps the network diagrams in this brochure will help you.? Unfortunately I had trouble just grabbing the image of the 1976 2 network internet transmission (Packet Radio and ARPAnet ) so I am including the entire publication.? https://computerhistory.org/wp-content/uploads/2019/08/core-2002-02.pdf barbara ? ? On Tuesday, January 28, 2025 at 10:04:22 PM PST, Lars Brinkhoff via Internet-history wrote:? Jack Haverty wrote: > SATNET, and ARPANET, were both continuously managed by the NOC at > BBN. TCP, and a collection of "core gateways" had become increasingly > important to users, especially in the UK in Peter Kirstein's group, > which had to rely on TCP through SATNET to access computers in the US. This would have been before 1983, correct?? I'm curious which (important) computers were avaiable through TCP? It seems to me the transition from NCP to TCP was somewhat gradual. Many histories make a big deal out of the 1/1/83 flag day, as if the entire network switched from NCP-only on one day, to TCP-only the next day.? But reading more carefully, I gather some hosts were TCP only or dual TCP/NCP long before that.? Is that correct? -- Internet-history mailing list Internet-history at elists.isoc.org https://elists.isoc.org/mailman/listinfo/internet-history ? -- Internet-history mailing list Internet-history at elists.isoc.org https://elists.isoc.org/mailman/listinfo/internet-history From jack at 3kitty.org Thu Jan 30 23:27:21 2025 From: jack at 3kitty.org (Jack Haverty) Date: Thu, 30 Jan 2025 23:27:21 -0800 Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: <1776225359.6704266.1738301141392@mail.yahoo.com> References: <20250128171348.41F9218C073@mercury.lcs.mit.edu> <7wjzae17he.fsf@junk.nocrew.org> <1314488472.5766718.1738137430770@mail.yahoo.com> <1776225359.6704266.1738301141392@mail.yahoo.com> Message-ID: [trying again... furst try was rejected "Message too big."?? The Internet can now handle gigabit speeds, but apparently not emails more than 400 kilobytes?] Nice find!? Here's a picture of the commemorative plaque, It's located "on the front of Rosotti?s just to the right of the front door".??? A friend found it there and took the picture.? /Jack On 1/30/25 21:25, Barbara Denny via Internet-history wrote: > Found a better copy of the network diagrams so you can see both TCP and NCP shown in the ARPAnet in 1976. > > https://ed-thelen.org/comp-hist/CORE-3-1-SRI-TCP-IP.html > When I have more time, I will see if I can find the SRI progress report that covers this activity. > > barbara > > > On Tuesday, January 28, 2025 at 11:59:57 PM PST, Barbara Denny via Internet-history wrote: > > TCP existed prior to 1983 in hosts attached to the various networks that comprised the Internet.? ARPAnet was just one of the networks on the Internet during this time period. Flag day marked the day when NCP would no longer be supported in the ARPAnet. Hosts attached to the ARPAnet needed to use TCP or they would essentially be disconnected. > Perhaps the network diagrams in this brochure will help you.? Unfortunately I had trouble just grabbing the image of the 1976 2 network internet transmission (Packet Radio and ARPAnet ) so I am including the entire publication. > https://computerhistory.org/wp-content/uploads/2019/08/core-2002-02.pdf > > barbara > ? ? On Tuesday, January 28, 2025 at 10:04:22 PM PST, Lars Brinkhoff via Internet-history wrote: > > Jack Haverty wrote: >> SATNET, and ARPANET, were both continuously managed by the NOC at >> BBN. TCP, and a collection of "core gateways" had become increasingly >> important to users, especially in the UK in Peter Kirstein's group, >> which had to rely on TCP through SATNET to access computers in the US. > This would have been before 1983, correct?? I'm curious which > (important) computers were avaiable through TCP? > > It seems to me the transition from NCP to TCP was somewhat gradual. > Many histories make a big deal out of the 1/1/83 flag day, as if the > entire network switched from NCP-only on one day, to TCP-only the next > day.? But reading more carefully, I gather some hosts were TCP only or > dual TCP/NCP long before that.? Is that correct? -------------- next part -------------- A non-text attachment was scrubbed... Name: OpenPGP_signature.asc Type: application/pgp-signature Size: 665 bytes Desc: OpenPGP digital signature URL: From touch at strayalpha.com Fri Jan 31 07:40:27 2025 From: touch at strayalpha.com (touch at strayalpha.com) Date: Fri, 31 Jan 2025 07:40:27 -0800 Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: References: <20250128171348.41F9218C073@mercury.lcs.mit.edu> <7wjzae17he.fsf@junk.nocrew.org> <1314488472.5766718.1738137430770@mail.yahoo.com> <1776225359.6704266.1738301141392@mail.yahoo.com> Message-ID: > On Jan 30, 2025, at 11:27?PM, Jack Haverty via Internet-history wrote: > > [trying again... furst try was rejected "Message too big." The Internet can now handle gigabit speeds, but apparently not emails more than 400 kilobytes?] That?s correct; as has been noted before, this list is for discussions but is not a storage archive. Large items should be posted via links to other storage sites. Joe (list admin) From jack at 3kitty.org Fri Jan 31 11:10:34 2025 From: jack at 3kitty.org (Jack Haverty) Date: Fri, 31 Jan 2025 11:10:34 -0800 Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: References: <20250128171348.41F9218C073@mercury.lcs.mit.edu> <7wjzae17he.fsf@junk.nocrew.org> <1314488472.5766718.1738137430770@mail.yahoo.com> <1776225359.6704266.1738301141392@mail.yahoo.com> Message-ID: <59418f5e-0104-49f6-aac3-78c2627be4e2@3kitty.org> Thanks, Joe.? I didn't remember ISOC's specific limitations until I got the rejection report, which said the message was too big.?? So I quickly converted the photo into a smaller size of 80KB, to fit well within the 400KB constraint, and resent it.? The second try made it through the list server, but the image was stripped away with no indication that it had ever been there.? I realize you can't do anything about it and sympathize. Apparently the ISOC service silently censors and alters messages as they pass through.?? The recipients don't get what I sent.? It also breaks my digital signature.?? I'm disappointed that ISOC, as parent of the Engineering arm of the Internet, doesn't use its own services as showcase models of "best practice" to demonstrate how to "do it right", as ARPA, NSF, et al did back in the early days of the Internet. Jack Haverty On 1/31/25 07:40, touch at strayalpha.com wrote: > >> On Jan 30, 2025, at 11:27?PM, Jack Haverty via Internet-history >> wrote: >> >> [trying again... furst try was rejected "Message too big."?? The >> Internet can now handle gigabit speeds, but apparently not emails >> more than 400 kilobytes?] > > That?s correct; as has been noted before, this list is for discussions > but is not a storage archive. > > Large items should be posted via links to other storage sites. > > Joe (list admin) -------------- next part -------------- A non-text attachment was scrubbed... Name: OpenPGP_signature.asc Type: application/pgp-signature Size: 665 bytes Desc: OpenPGP digital signature URL: From dhc at dcrocker.net Fri Jan 31 11:22:14 2025 From: dhc at dcrocker.net (Dave Crocker) Date: Fri, 31 Jan 2025 19:22:14 +0000 (UTC) Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: <59418f5e-0104-49f6-aac3-78c2627be4e2@3kitty.org> References: <20250128171348.41F9218C073@mercury.lcs.mit.edu> <7wjzae17he.fsf@junk.nocrew.org> <1314488472.5766718.1738137430770@mail.yahoo.com> <1776225359.6704266.1738301141392@mail.yahoo.com> <59418f5e-0104-49f6-aac3-78c2627be4e2@3kitty.org> Message-ID: <7f065e97-6fc1-44d9-8dbb-e20479b5b1b3@dcrocker.net> On 1/31/2025 11:10 AM, Jack Haverty via Internet-history wrote: > Apparently the ISOC service silently censors and alters messages as > they pass through.?? The recipients don't get what I sent.? It also > breaks my digital signature.?? I'm disappointed that ISOC, as parent > of the Engineering arm of the Internet, doesn't use its own services > as showcase models of "best practice" to demonstrate how to "do it > right", as ARPA, NSF, et al did back in the early days of the Internet. Arguably, in this space, about these issues, there are common practices, but few or no best practices. Or a different view:? ISOC demonstrates the real world, and in this space, the real world is painful, inconsistent, and constraining.? ISOC is therefore doing quite a good job at communicating how things actually are... d/ -- Dave Crocker Brandenburg InternetWorking bbiw.net bluesky: @dcrocker.bsky.social mast: @dcrocker at mastodon.social From jnc at mercury.lcs.mit.edu Fri Jan 31 12:09:06 2025 From: jnc at mercury.lcs.mit.edu (Noel Chiappa) Date: Fri, 31 Jan 2025 15:09:06 -0500 (EST) Subject: [ih] Correct name for early TCP/IP working group? Message-ID: <20250131200906.623A218C074@mercury.lcs.mit.edu> > From: John Shoch > Thanks for putting up a page on the Internet Experiment Notes, and > early meetings. My contribution was minimal; the latter is mostly just a list of meetings, and links to the minutes. (Again, Jon gets all the credit for recording our travails in detail.) As far as I know, that era of internetting work, which was fairly important technically (seeing the TCP-IP split, among other important steps) is very poorly covered in later histories. E.g. Abbate's otherwise excellent book skims over that stage in a page or two. In part, that seems to be because her book focuses on the organizational history, and also on what the users saw from the outside, with not much focus on the technical details. The organizational and user history _are_ important, and should be covered in such a work. However, without the underlying technology, there would have been no project, so it should be covered too - in addition to being of inherent intellectual value/interest. This is a common fault I find with histories of technology by professional historians. (I'm not trying to give them grief, here.) They don't understand the technology well enough to point out the key technical points - unless they can find a piece of paper from a technologist that makes the point. (I will pass over a gripe I have about that trait of historians, around the argument about dropping the bombs on Japan; it's irrelevant to us.) And sometimes the point isn't obvious to many technologists! We had a really great example of this recently, with the observation that packet-based systems are inherently better at scaling to gargantuan sizes; as Brian pointed out: https://elists.isoc.org/pipermail/internet-history/2025-January/010120.html "it's almost as if the intrinsic scalability of stateless packet switching was an unnoticed and accidental property". But it's an even better example of my general observation about 'the point not being obvious to many technologists in the field'. As I mentioned recently, on the Unix Heritage Society mailing list (which seems to contain a lot of mindless Unix fans who don't know anything about other OS's), they were talking about how great Datakit was, and bemoaning the fact that it was bypassed by history. (Which forced me to write a note: https://www.tuhs.org/pipermail/tuhs/2022-June/026001.html explaining that virtual circuit networks wouldn't scale to the size of today's Internet.) A wonderful example of this is "Crystal Fire", by Hoddeson and and Riordan - to me, by some ways, the best history of the creation of semiconductor electronics. I think it is not an accident that Hoddeson and and Riordan were themselves physicists, before they turned to history. Someday someone with write a technical history of the development of internetting. I'm not sure there is one, yet. Noel From vint at google.com Fri Jan 31 12:19:09 2025 From: vint at google.com (Vint Cerf) Date: Fri, 31 Jan 2025 15:19:09 -0500 Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: <20250131200906.623A218C074@mercury.lcs.mit.edu> References: <20250131200906.623A218C074@mercury.lcs.mit.edu> Message-ID: how much of the technical history might be found in Dave Clark's book on Internet design? v On Fri, Jan 31, 2025 at 3:09?PM Noel Chiappa via Internet-history < internet-history at elists.isoc.org> wrote: > > From: John Shoch > > > Thanks for putting up a page on the Internet Experiment Notes, and > > early meetings. > > My contribution was minimal; the latter is mostly just a list of meetings, > and links to the minutes. (Again, Jon gets all the credit for recording our > travails in detail.) > > As far as I know, that era of internetting work, which was fairly important > technically (seeing the TCP-IP split, among other important steps) is very > poorly covered in later histories. E.g. Abbate's otherwise excellent book > skims over that stage in a page or two. > > > In part, that seems to be because her book focuses on the organizational > history, and also on what the users saw from the outside, with not much > focus > on the technical details. The organizational and user history _are_ > important, and should be covered in such a work. However, without the > underlying technology, there would have been no project, so it should be > covered too - in addition to being of inherent intellectual value/interest. > > This is a common fault I find with histories of technology by professional > historians. (I'm not trying to give them grief, here.) They don't > understand > the technology well enough to point out the key technical points - unless > they > can find a piece of paper from a technologist that makes the point. (I > will > pass over a gripe I have about that trait of historians, around the > argument > about dropping the bombs on Japan; it's irrelevant to us.) > > And sometimes the point isn't obvious to many technologists! We had a > really > great example of this recently, with the observation that packet-based > systems are inherently better at scaling to gargantuan sizes; as Brian > pointed out: > > > https://elists.isoc.org/pipermail/internet-history/2025-January/010120.html > > "it's almost as if the intrinsic scalability of stateless packet switching > was an unnoticed and accidental property". > > But it's an even better example of my general observation about 'the point > not being obvious to many technologists in the field'. As I mentioned > recently, on the Unix Heritage Society mailing list (which seems to > contain a > lot of mindless Unix fans who don't know anything about other OS's), they > were talking about how great Datakit was, and bemoaning the fact that it > was bypassed by history. > > (Which forced me to write a note: > > https://www.tuhs.org/pipermail/tuhs/2022-June/026001.html > > explaining that virtual circuit networks wouldn't scale to the size of > today's Internet.) > > > A wonderful example of this is "Crystal Fire", by Hoddeson and and Riordan > - > to me, by some ways, the best history of the creation of semiconductor > electronics. I think it is not an accident that Hoddeson and and Riordan > were > themselves physicists, before they turned to history. > > Someday someone with write a technical history of the development of > internetting. I'm not sure there is one, yet. > > Noel > -- > Internet-history mailing list > Internet-history at elists.isoc.org > https://elists.isoc.org/mailman/listinfo/internet-history > -- Please send any postal/overnight deliveries to: Vint Cerf Google, LLC 1900 Reston Metro Plaza, 16th Floor Reston, VA 20190 +1 (571) 213 1346 until further notice From sob at sobco.com Fri Jan 31 12:29:59 2025 From: sob at sobco.com (Scott Bradner) Date: Fri, 31 Jan 2025 15:29:59 -0500 Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: References: <20250128171348.41F9218C073@mercury.lcs.mit.edu> <7wjzae17he.fsf@junk.nocrew.org> <1314488472.5766718.1738137430770@mail.yahoo.com> <1776225359.6704266.1738301141392@mail.yahoo.com> Message-ID: <9B1E0373-A66D-4A39-A3F8-1DBE674587B5@sobco.com> a number of people on the list are complaining about ISOC's email rules but based on this letter the choices may be the list moderator's (rules that make sense to me fwiw) Scott > On Jan 31, 2025, at 10:40?AM, touch--- via Internet-history wrote: > > >> On Jan 30, 2025, at 11:27?PM, Jack Haverty via Internet-history wrote: >> >> [trying again... furst try was rejected "Message too big." The Internet can now handle gigabit speeds, but apparently not emails more than 400 kilobytes?] > > That?s correct; as has been noted before, this list is for discussions but is not a storage archive. > > Large items should be posted via links to other storage sites. > > Joe (list admin) > -- > Internet-history mailing list > Internet-history at elists.isoc.org > https://elists.isoc.org/mailman/listinfo/internet-history From johnl at iecc.com Fri Jan 31 15:42:00 2025 From: johnl at iecc.com (John Levine) Date: 31 Jan 2025 18:42:00 -0500 Subject: [ih] mail hosting, Correct name for early TCP/IP working group? In-Reply-To: <9B1E0373-A66D-4A39-A3F8-1DBE674587B5@sobco.com> References: <7wjzae17he.fsf@junk.nocrew.org> <20250128171348.41F9218C073@mercury.lcs.mit.edu> <1314488472.5766718.1738137430770@mail.yahoo.com> <1776225359.6704266.1738301141392@mail.yahoo.com> <9B1E0373-A66D-4A39-A3F8-1DBE674587B5@sobco.com> Message-ID: <20250131234201.41D4EBAC3B89@ary.qy> It appears that Scott Bradner via Internet-history said: >a number of people on the list are complaining about ISOC's email rules but based on >this letter the choices may be the list moderator's (rules that make sense to me fwiw) ISOC is hosting this list basically as a favor to me, having asked them to do it when its previous host was no longer available. It's free, with a 200% money back guarantee. I'm on other lists hosted at groups.io, who provide exellent service and file hosting and all sorts of other goodies, for $220/yr. If people would prefer that, we could pass the hat and move it. Or we could figure that free is adequate. R's, John PS: "This list is very important but not so important that I'm willing to pay" is not a winning argument. From brian.e.carpenter at gmail.com Fri Jan 31 16:46:45 2025 From: brian.e.carpenter at gmail.com (Brian E Carpenter) Date: Sat, 1 Feb 2025 13:46:45 +1300 Subject: [ih] Correct name for early TCP/IP working group? In-Reply-To: <59418f5e-0104-49f6-aac3-78c2627be4e2@3kitty.org> References: <20250128171348.41F9218C073@mercury.lcs.mit.edu> <7wjzae17he.fsf@junk.nocrew.org> <1314488472.5766718.1738137430770@mail.yahoo.com> <1776225359.6704266.1738301141392@mail.yahoo.com> <59418f5e-0104-49f6-aac3-78c2627be4e2@3kitty.org> Message-ID: <37b9b85b-647a-4bd7-9423-b56c9174abfa@gmail.com> Jack, Back when 95% (or whatever the exact fraction is) of email wasn't spam, mailing list operators didn't have to do anything special. But today, every mailing list operator has to either do a number of things that involve munging messages in one way or another, to avoid anti-spam mechanisms used by all the major email provders, or give up and close the lists. An expert on this such as John Levine could explain many of those munging mechanisms, so I won't try. But ISOC's choice is to rewrite the nominal sender of the mail to match the actual sender, i.e. Jack Haverty via Internet-history for your messages, so naturally they will not be signed by you when they reach subscribers. That's "doing it right" in the era of pervasive spam. As for: >>> Large items should be posted via links to other storage sites. Surely people here of all people are aware that mailing list archives are a very poor method of digital conservation. For example, many (probably most) IETF WG mail archives prior to the lists being hosted at ietf.org are incomplete or lost. Regards Brian Carpenter On 01-Feb-25 08:10, Jack Haverty via Internet-history wrote: > Thanks, Joe.? I didn't remember ISOC's specific limitations until I got > the rejection report, which said the message was too big.?? So I quickly > converted the photo into a smaller size of 80KB, to fit well within the > 400KB constraint, and resent it.? The second try made it through the > list server, but the image was stripped away with no indication that it > had ever been there.? I realize you can't do anything about it and > sympathize. > > Apparently the ISOC service silently censors and alters messages as they > pass through.?? The recipients don't get what I sent.? It also breaks my > digital signature.?? I'm disappointed that ISOC, as parent of the > Engineering arm of the Internet, doesn't use its own services as > showcase models of "best practice" to demonstrate how to "do it right", > as ARPA, NSF, et al did back in the early days of the Internet. > > Jack Haverty > > On 1/31/25 07:40, touch at strayalpha.com wrote: >> >>> On Jan 30, 2025, at 11:27?PM, Jack Haverty via Internet-history >>> wrote: >>> >>> [trying again... furst try was rejected "Message too big."?? The >>> Internet can now handle gigabit speeds, but apparently not emails >>> more than 400 kilobytes?] >> >> That?s correct; as has been noted before, this list is for discussions >> but is not a storage archive. >> >> Large items should be posted via links to other storage sites. >> >> Joe (list admin) > > From j at shoch.com Fri Jan 31 17:38:34 2025 From: j at shoch.com (John Shoch) Date: Fri, 31 Jan 2025 17:38:34 -0800 Subject: [ih] Internet-history Digest, Vol 62, Issue 32 In-Reply-To: References: Message-ID: Noel, Your concern about the technical content of computer history puts you in good company -- with Don Knuth. --In 2021 the CACM published a transcript of a talk Don had given 7 years before: "Let?s Not Dumb Down the History of Computer Science." From the CACM's introduction: "On May 7, 2014, Don Knuth delivered that year's Kailath Lecture at Stanford University to a packed auditorium. In it he decried the absence of technical content from the histories of computer science being written, and he made an impassioned plea for historians of computer science to get back on track, as the historians of mathematics have always been." https://cacm.acm.org/opinion/lets-not-dumb-down-the-history-of-computer-science/ --I had the good fortune to be at that lecture 10 years ago, which is available on Youtube: https://www.youtube.com/watch?v=gAXdDEQveKw [There is an amusing introduction by John Hennesey.] --A year after the talk, in 2015, the CACM ran a response from Thomas Haigh, arguing the other side: "In this column I will be looking at the changing relationship between the discipline of computer science and the growing body of scholarly work on the history of computing, beginning with a recent plea made by renowned computer scientist Donald Knuth." ... "Computing is much bigger than computer science, and so the history of computing is much bigger than the history of computer science. Yet Knuth treated Campbell-Kelly?s book on the business history of the software industry (accurately subtitled ?a history of the software industry?) and all the rest of the history of computing as part of ?the history of computer science.? ... "To call such work ?dumbed down? history of computer science, rather than smart history of many other things, is to misunderstand both the intentions and the accomplishments of its authors." https://cacm.acm.org/opinion/the-tears-of-donald-knuth/ I commend both articles to you. John PS: Personally, I tend to come down on the Haigh side of the discussion. "History" can take many forms when looking at any subject area -- political history, economic history, business history, social history, technical history, architectural history, etc. An example: some of you may know that Robert Garner is undertaking a prodigious effort to dig into the technical evolution of the Ethernet (he's looking at original board designs, simulation equations, timing issues, etc. Yet that still leaves room for different historical work on the techno/political battles of the standardization process (Ethernet vs. token ring, Xerox/Dec/Intel vs. IBM, IEEE vs. ECMA in Europe, etc.). I think we need both. [Full disclosure: I served for about two decades on the Board of the Computer History Museum, which informed my broader view of the opportunity. The technical history of computing (and networking) is important, but will probably serve a narrower audience.]