From mbaer at cs.tu-berlin.de Tue Mar 10 13:35:23 2009 From: mbaer at cs.tu-berlin.de (=?ISO-8859-1?Q?Matthias_B=E4rwolff?=) Date: Tue, 10 Mar 2009 16:35:23 -0400 Subject: [ih] Baran and arbitrary reliability from arbitrarily unreliable components Message-ID: <49B6CF0B.301@cs.tu-berlin.de> Dear all, I am aware that aspects of this issue have been discussed all over the place, but I still cannot quite figure out what the factual and logical role of Paul Baran was for the development of the internet. His writings strike me as very early and elegant descriptions of the basic notion that a little redundancy in routes (with greater redundancy closer to the edges) together with distributed routing algorithms (even random routing, and routers learning about good routes by counting hops will do) solves the problems of connectivity, congestion avoidance and recovery in shared (multiplexed) packet networks without (virtual) circuits. Had it simply been that this was not immediately relevant for the Arpanet at first (RFNM), and the later developments leading to the internet (multiple ASs, routing based on business considerations)? Thanks for your helping me understanding. Matthias From jnc at mercury.lcs.mit.edu Tue Mar 10 19:27:23 2009 From: jnc at mercury.lcs.mit.edu (Noel Chiappa) Date: Tue, 10 Mar 2009 22:27:23 -0400 (EDT) Subject: [ih] Baran and arbitrary reliability from arbitrarily unreliable components Message-ID: <20090311022723.625A26BE5DE@mercury.lcs.mit.edu> > From: mbaer at cs.tu-berlin.de > I still cannot quite figure out what the factual and logical role of > Paul Baran was for the development of the internet. It's not so much "the Internet", as packet networks in general; the Internet is just the virtually omnipresent network which has subsumed the many small, independent packet-switching networks that used to exist. > His writings strike me as very early and elegant descriptions of the > basic notion that a little redundancy in routes ... solves the problems > of connectivity, congestion avoidance and recovery in shared > (multiplexed) packet networks without (virtual) circuits. That is an important contribution, but not, to me, his most important. I would say that his most important single contribution was the concept of breaking a long "message" up into small, relatively fixed-size, self-describing segments (later named "packets"). I say 'relatively fixed size' because the size range is restricted to a relatively small range, compared to the size of the data objects sent over the network; IPv4 packets ranged from 0 bytes to 576 (or slightly larger now). That is tiny compared to files of many megabytes, which existed even back then. The relatively small maximum size leads to many important consequences, both engineering and fundamental. One advantage is that buffering is much easier with relatively small maximum sizes. Another is that all real links have a non-zero bit error rate, and the chances of segment being corrupted by an error goes up as the segments get large, so an extremely large segment has a small chance of getting through without an error. Yet another is that if you want to have varying precedences, a large low-priority segment can block out other traffic. Sending the segments independently through a _shared_ switching fabric which is shared in a non-pre-scheduled manner among the users is also important, but there are precedents for that idea (e.g. the telegraph network, the road network, etc). But really, his vision was for an overall _system_, and to draw attention to one specific novel aspect - be it the small, self-describing segments, or the on-demand sharing of the fabric (although that has precedents), or the dynamic routing (also with precedents), is to miss the novelty of his entire vision. It's hard to imagine, now, what communications systems looked like before Baran, but to really understand what he did, you need to try and go back, and see the world as it was in, say, 1957. The idea of a modem, of sending digital data over a telephone line, was only about 10 years old (I think the Whirlwind/Cape Cod system had the first ones). Most communication was voice, over networks which dedicated a circuit (some of the links of which perhaps used links shared in a fixed way, through frequency-division, or later time-division, multi-plexing) to the call, and there had to be a set-up, and tear-down, and the circuit was 'up' until the call was completed. Telex was similar, although its data was effectively digital, not analog. Telegraph networks were the closest thing (conceptually) to the packet networks of today. In this environment, Baran's system, described in a series of papers (available online here: http://www.rand.org/about/history/baran.list.html which you should really read if you are i) interested, and ii) have not done so already) was truly radically different. His description of the advantages of his system (see, e.g. pages 22-23 of Volume I) is clear, and far-seeing; the advantages he lists (e.g. sharing a single infrastructure among users with a wide range of bandwidth requirements) are still true today. > Had it simply been that this was not immediately relevant for the > Arpanet at first (RFNM), and the later developments leading to the > internet (multiple ASs, routing based on business considerations)? Baran's specific ideas have little direct relevance to the Internet; many of the specific mechanisms (e.g. congestion control, routing, etc) are very different _in detail_. But the _basic_ concept, of a system for digital communication, based around packets, is the foundation of all data networks since then. Noel From vint at google.com Wed Mar 11 03:18:29 2009 From: vint at google.com (Vint Cerf) Date: Wed, 11 Mar 2009 06:18:29 -0400 Subject: [ih] Baran and arbitrary reliability from arbitrarily unreliable components In-Reply-To: <20090311022723.625A26BE5DE@mercury.lcs.mit.edu> References: <20090311022723.625A26BE5DE@mercury.lcs.mit.edu> Message-ID: Adding to Noel's excellent remarks: Baran was designing a VOICE communication system for command and control. The "message blocks" that his system routed around ("hot potato routing") were chunks of voice. Small packets for rapid delivery (low latency). Radio based. Small packets also reduce likelihood that error correction would fail (this from guesswork and memory, I may be wrong about forward error correction but I believe it was included). a very radical piece of work. It was rejected by the then telecom experts of the Defense Department who "KNEW" that the only way to do telecom was circuit switching. How wrong they were. v Vint Cerf Google 1818 Library Street, Suite 400 Reston, VA 20190 202-370-5637 vint at google.com On Mar 10, 2009, at 10:27 PM, Noel Chiappa wrote: >> From: mbaer at cs.tu-berlin.de > >> I still cannot quite figure out what the factual and logical role of >> Paul Baran was for the development of the internet. > > It's not so much "the Internet", as packet networks in general; the > Internet > is just the virtually omnipresent network which has subsumed the > many small, > independent packet-switching networks that used to exist. > > >> His writings strike me as very early and elegant descriptions of the >> basic notion that a little redundancy in routes ... solves the >> problems >> of connectivity, congestion avoidance and recovery in shared >> (multiplexed) packet networks without (virtual) circuits. > > That is an important contribution, but not, to me, his most > important. I > would say that his most important single contribution was the > concept of > breaking a long "message" up into small, relatively fixed-size, > self-describing segments (later named "packets"). > > I say 'relatively fixed size' because the size range is restricted > to a > relatively small range, compared to the size of the data objects > sent over > the network; IPv4 packets ranged from 0 bytes to 576 (or slightly > larger > now). That is tiny compared to files of many megabytes, which > existed even > back then. The relatively small maximum size leads to many important > consequences, both engineering and fundamental. > > One advantage is that buffering is much easier with relatively small > maximum > sizes. Another is that all real links have a non-zero bit error > rate, and the > chances of segment being corrupted by an error goes up as the > segments get > large, so an extremely large segment has a small chance of getting > through > without an error. Yet another is that if you want to have varying > precedences, a large low-priority segment can block out other traffic. > > Sending the segments independently through a _shared_ switching > fabric which > is shared in a non-pre-scheduled manner among the users is also > important, > but there are precedents for that idea (e.g. the telegraph network, > the road > network, etc). > > > But really, his vision was for an overall _system_, and to draw > attention to > one specific novel aspect - be it the small, self-describing > segments, or the > on-demand sharing of the fabric (although that has precedents), or the > dynamic routing (also with precedents), is to miss the novelty of > his entire > vision. > > It's hard to imagine, now, what communications systems looked like > before > Baran, but to really understand what he did, you need to try and go > back, and > see the world as it was in, say, 1957. > > The idea of a modem, of sending digital data over a telephone line, > was only > about 10 years old (I think the Whirlwind/Cape Cod system had the > first > ones). Most communication was voice, over networks which dedicated a > circuit > (some of the links of which perhaps used links shared in a fixed > way, through > frequency-division, or later time-division, multi-plexing) to the > call, and > there had to be a set-up, and tear-down, and the circuit was 'up' > until the > call was completed. Telex was similar, although its data was > effectively > digital, not analog. Telegraph networks were the closest thing > (conceptually) > to the packet networks of today. > > In this environment, Baran's system, described in a series of papers > (available online here: > > http://www.rand.org/about/history/baran.list.html > > which you should really read if you are i) interested, and ii) have > not done > so already) was truly radically different. His description of the > advantages > of his system (see, e.g. pages 22-23 of Volume I) is clear, and far- > seeing; > the advantages he lists (e.g. sharing a single infrastructure among > users > with a wide range of bandwidth requirements) are still true today. > > >> Had it simply been that this was not immediately relevant for the >> Arpanet at first (RFNM), and the later developments leading to the >> internet (multiple ASs, routing based on business considerations)? > > Baran's specific ideas have little direct relevance to the Internet; > many of > the specific mechanisms (e.g. congestion control, routing, etc) are > very > different _in detail_. > > But the _basic_ concept, of a system for digital communication, > based around > packets, is the foundation of all data networks since then. > > Noel From bernie at fantasyfarm.com Wed Mar 11 05:15:58 2009 From: bernie at fantasyfarm.com (Bernie Cosell) Date: Wed, 11 Mar 2009 08:15:58 -0400 Subject: [ih] Baran and arbitrary reliability from arbitrarily unreliable components In-Reply-To: References: <20090311022723.625A26BE5DE@mercury.lcs.mit.edu>, Message-ID: <49B7733E.2235.1B5B99@bernie.fantasyfarm.com> On 11 Mar 2009 at 6:18, Vint Cerf wrote: > a very radical piece of work. It was rejected by the then telecom > experts of the Defense Department who "KNEW" that the only way to do > telecom was circuit switching. > > How wrong they were. Indeed, didn't they make the same mistake *again*, many years later [and post-ARPAnet] when they let AUTODIN II to Western Union? /Bernie\ -- Bernie Cosell Fantasy Farm Fibers mailto:bernie at fantasyfarm.com Pearisburg, VA --> Too many people, too few sheep <-- From vint at google.com Wed Mar 11 05:50:29 2009 From: vint at google.com (Vint Cerf) Date: Wed, 11 Mar 2009 05:50:29 -0700 Subject: [ih] Baran and arbitrary reliability from arbitrarilyunreliable components Message-ID: Less clear since peter sevcik was architect and very packet oriented. V ----- Original Message ----- From: internet-history-bounces at postel.org To: internet-history at postel.org Sent: Wed Mar 11 05:15:58 2009 Subject: Re: [ih] Baran and arbitrary reliability from arbitrarilyunreliable components On 11 Mar 2009 at 6:18, Vint Cerf wrote: > a very radical piece of work. It was rejected by the then telecom > experts of the Defense Department who "KNEW" that the only way to do > telecom was circuit switching. > > How wrong they were. Indeed, didn't they make the same mistake *again*, many years later [and post-ARPAnet] when they let AUTODIN II to Western Union? /Bernie\ -- Bernie Cosell Fantasy Farm Fibers mailto:bernie at fantasyfarm.com Pearisburg, VA --> Too many people, too few sheep <-- From jeanjour at comcast.net Wed Mar 11 06:08:41 2009 From: jeanjour at comcast.net (John Day) Date: Wed, 11 Mar 2009 09:08:41 -0400 Subject: [ih] Baran and arbitrary reliability from arbitrarily unreliable components In-Reply-To: <49B7733E.2235.1B5B99@bernie.fantasyfarm.com> References: <20090311022723.625A26BE5DE@mercury.lcs.mit.edu>, <49B7733E.2235.1B5B99@bernie.fantasyfarm.com> Message-ID: That was a mistake, but a different one. ;-) They were just incompetent. At least the telephony guys who turned down Baran knew telephony well. I predicted the outcome of DIN II on the day I heard who had gotten the contract. I doubt that I was alone. ;--) >On 11 Mar 2009 at 6:18, Vint Cerf wrote: > >> a very radical piece of work. It was rejected by the then telecom >> experts of the Defense Department who "KNEW" that the only way to do >> telecom was circuit switching. >> >> How wrong they were. > >Indeed, didn't they make the same mistake *again*, many years later [and >post-ARPAnet] when they let AUTODIN II to Western Union? > > /Bernie\ > >-- >Bernie Cosell Fantasy Farm Fibers >mailto:bernie at fantasyfarm.com Pearisburg, VA > --> Too many people, too few sheep <-- At 8:15 -0400 2009/03/11, Bernie Cosell wrote: >On 11 Mar 2009 at 6:18, Vint Cerf wrote: > >> a very radical piece of work. It was rejected by the then telecom >> experts of the Defense Department who "KNEW" that the only way to do >> telecom was circuit switching. >> >> How wrong they were. > >Indeed, didn't they make the same mistake *again*, many years later [and >post-ARPAnet] when they let AUTODIN II to Western Union? > > /Bernie\ > >-- >Bernie Cosell Fantasy Farm Fibers >mailto:bernie at fantasyfarm.com Pearisburg, VA > --> Too many people, too few sheep <-- From jnc at mercury.lcs.mit.edu Wed Mar 11 09:19:41 2009 From: jnc at mercury.lcs.mit.edu (Noel Chiappa) Date: Wed, 11 Mar 2009 12:19:41 -0400 (EDT) Subject: [ih] Baran and arbitrary reliability from arbitrarily unreliable components Message-ID: <20090311161941.A65A96BE5F0@mercury.lcs.mit.edu> > From: Vint Cerf > Baran was designing a VOICE communication system for command and > control. The "message blocks" that his system routed around ... were > chunks of voice. I'm not sure that's correct. In later remarks he definitely includes data, telex, etc; see, for example, his Babbage Institute oral history interview, page 18, "The basic idea that once you went digital, signals from data, teletypewriters, facsimile, and voice would all be digitized." (No doubt he saw telex as filling the application/human-use role now played by email, which of course didn't exist then.) I briefly tried to find the range of applications listed in the original documents (to make sure this was not human memory cross-talk), but so far haven't, although in Volume XI, page vi, he speaks of attaching "telephones and typewriters", so he likely did have the full range in mind back then. This does point to an additional point about Baran's system - he saw that with packets one could support everything, whereas with circuits, the converse is not true. (Volume I, page 22: "While a standardized message block is common in many computer-communication applications, no serious attempt has ever been made to use it as a universal standard.") And of course if one is trying to send voice, limiting the message size is important to minimize end-end delay jitter (hello, ATM :-), once again driving towards breaking large (user-semantics) messages up into segments. In re-reading the documents to find these references, a couple of more points about his system came through: the systems' ability to provide a _reliable system_ with _unreliable components_, again a somewhat novel concept, but now taken for granted. As part of that, his system was entirely distributed, with no central control - something again now taken for granted. He also showed how a certain degree of redundancy would provide a system which was _highly_ immune to _major_ losses of links and nodes; this is not so important in today's Internet, but was also a significant result. In short, I think an old quote from the first Unix paper in CACM also applies to Baran's system (this is from memory, so it's only approximate): 'the strength of Unix is not so much in new ideas, although there are a few new ones, as in the way in which a _carefully-selected set of mechanisms_ work together to provide a very powerful overall system'. > Small packets also reduce likelihood that error correction would fail > (this from guesswork and memory, I may be wrong about forward error > correction but I believe it was included). I'm sure there was some sort of error detection, because the way his system worked was that a node retained a copy of a message until it got an acknowledgement from the neighbouring node that it had been received correctly, only then would it discard it. Of course, many FEC mechanisms also provide 'found error that couldn't be correct' outputs, so perhaps it was FEC, and not just a data-integrity check. The system may have had FEC on the radio links, for the exact same reason as too (BER too high), but I didn't feel like reading the Volume on radio links to find out! :-) Noel From jeanjour at comcast.net Wed Mar 11 09:55:51 2009 From: jeanjour at comcast.net (John Day) Date: Wed, 11 Mar 2009 12:55:51 -0400 Subject: [ih] Baran and arbitrary reliability from arbitrarily unreliable components In-Reply-To: <20090311161941.A65A96BE5F0@mercury.lcs.mit.edu> References: <20090311161941.A65A96BE5F0@mercury.lcs.mit.edu> Message-ID: The idea of "building reliable systems from unreliable parts" was very much in the air during this period, because of John von Neuman's seminal paper of roughly that title. It is von Neuman's paper that first proposes triple modular redundancy and works out the math for it. Everyone was reading it and the turn of phrase was common. But I agree with Noel, that Baran had more than voice on his mind as applications. At 12:19 -0400 2009/03/11, Noel Chiappa wrote: > > From: Vint Cerf > > > Baran was designing a VOICE communication system for command and > > control. The "message blocks" that his system routed around ... were > > chunks of voice. > >I'm not sure that's correct. In later remarks he definitely includes data, >telex, etc; see, for example, his Babbage Institute oral history interview, >page 18, "The basic idea that once you went digital, signals from data, >teletypewriters, facsimile, and voice would all be digitized." (No doubt he >saw telex as filling the application/human-use role now played by email, >which of course didn't exist then.) I briefly tried to find the range of >applications listed in the original documents (to make sure this was not >human memory cross-talk), but so far haven't, although in Volume XI, page >vi, he speaks of attaching "telephones and typewriters", so he likely did >have the full range in mind back then. > >This does point to an additional point about Baran's system - he saw that >with packets one could support everything, whereas with circuits, the >converse is not true. (Volume I, page 22: "While a standardized message block >is common in many computer-communication applications, no serious attempt has >ever been made to use it as a universal standard.") > >And of course if one is trying to send voice, limiting the message size is >important to minimize end-end delay jitter (hello, ATM :-), once again >driving towards breaking large (user-semantics) messages up into segments. > > >In re-reading the documents to find these references, a couple of more points >about his system came through: the systems' ability to provide a _reliable >system_ with _unreliable components_, again a somewhat novel concept, but now >taken for granted. As part of that, his system was entirely distributed, with >no central control - something again now taken for granted. He also showed >how a certain degree of redundancy would provide a system which was _highly_ >immune to _major_ losses of links and nodes; this is not so important in >today's Internet, but was also a significant result. > >In short, I think an old quote from the first Unix paper in CACM also applies >to Baran's system (this is from memory, so it's only approximate): 'the >strength of Unix is not so much in new ideas, although there are a few new >ones, as in the way in which a _carefully-selected set of mechanisms_ work >together to provide a very powerful overall system'. > > > > Small packets also reduce likelihood that error correction would fail > > (this from guesswork and memory, I may be wrong about forward error > > correction but I believe it was included). > >I'm sure there was some sort of error detection, because the way his system >worked was that a node retained a copy of a message until it got an >acknowledgement from the neighbouring node that it had been received >correctly, only then would it discard it. Of course, many FEC mechanisms also >provide 'found error that couldn't be correct' outputs, so perhaps it was >FEC, and not just a data-integrity check. > >The system may have had FEC on the radio links, for the exact same reason as >too (BER too high), but I didn't feel like reading the Volume on radio links >to find out! :-) > > Noel From mbaer at cs.tu-berlin.de Wed Mar 11 12:32:19 2009 From: mbaer at cs.tu-berlin.de (=?ISO-8859-1?Q?Matthias_B=E4rwolff?=) Date: Wed, 11 Mar 2009 15:32:19 -0400 Subject: [ih] Baran and arbitrary reliability from arbitrarily unreliable components In-Reply-To: References: <20090311161941.A65A96BE5F0@mercury.lcs.mit.edu> Message-ID: <49B811C3.5070705@cs.tu-berlin.de> John Day wrote: > The idea of "building reliable systems from unreliable parts" was very > much in the air during this period, because of John von Neuman's > seminal paper of roughly that title. It is von Neuman's paper that > first proposes triple modular redundancy and works out the math for > it. Everyone was reading it and the turn of phrase was common. Baran refers to the 1956 von Neuman paper in a 1960 RAND report (P-1995, http://rand.org/pubs/papers/2008/P1995.pdf) and also to one by Moore and Shannon on the same subject. On my initial question: I gather from your responses that the intellectual influence of Baran is significant, and that many of the now commonplace ideas about networking are due largely to him (and Pouzin). However, I recall having read in several places that ARPA had not been influenced by Baran's work at all, and Roberts only had been made aware of Baran's work by Davies in the 1967. Just found that odd. Has the RFQ in 1968 been written in complete ignorance of Baran's work? After all, it envisages packet orientation and distributed routing. > > But I agree with Noel, that Baran had more than voice on his mind as > applications. > > At 12:19 -0400 2009/03/11, Noel Chiappa wrote: >> > From: Vint Cerf >> >> > Baran was designing a VOICE communication system for command and >> > control. The "message blocks" that his system routed around ... >> were >> > chunks of voice. >> >> I'm not sure that's correct. In later remarks he definitely includes >> data, >> telex, etc; see, for example, his Babbage Institute oral history >> interview, >> page 18, "The basic idea that once you went digital, signals from data, >> teletypewriters, facsimile, and voice would all be digitized." (No >> doubt he >> saw telex as filling the application/human-use role now played by email, >> which of course didn't exist then.) I briefly tried to find the range of >> applications listed in the original documents (to make sure this was not >> human memory cross-talk), but so far haven't, although in Volume XI, >> page >> vi, he speaks of attaching "telephones and typewriters", so he likely >> did >> have the full range in mind back then. >> >> This does point to an additional point about Baran's system - he saw >> that >> with packets one could support everything, whereas with circuits, the >> converse is not true. (Volume I, page 22: "While a standardized >> message block >> is common in many computer-communication applications, no serious >> attempt has >> ever been made to use it as a universal standard.") >> >> And of course if one is trying to send voice, limiting the message >> size is >> important to minimize end-end delay jitter (hello, ATM :-), once again >> driving towards breaking large (user-semantics) messages up into >> segments. >> >> >> In re-reading the documents to find these references, a couple of >> more points >> about his system came through: the systems' ability to provide a >> _reliable >> system_ with _unreliable components_, again a somewhat novel concept, >> but now >> taken for granted. As part of that, his system was entirely >> distributed, with >> no central control - something again now taken for granted. He also >> showed >> how a certain degree of redundancy would provide a system which was >> _highly_ >> immune to _major_ losses of links and nodes; this is not so important in >> today's Internet, but was also a significant result. >> >> In short, I think an old quote from the first Unix paper in CACM also >> applies >> to Baran's system (this is from memory, so it's only approximate): 'the >> strength of Unix is not so much in new ideas, although there are a >> few new >> ones, as in the way in which a _carefully-selected set of mechanisms_ >> work >> together to provide a very powerful overall system'. >> >> >> > Small packets also reduce likelihood that error correction >> would fail >> > (this from guesswork and memory, I may be wrong about forward >> error >> > correction but I believe it was included). >> >> I'm sure there was some sort of error detection, because the way his >> system >> worked was that a node retained a copy of a message until it got an >> acknowledgement from the neighbouring node that it had been received >> correctly, only then would it discard it. Of course, many FEC >> mechanisms also >> provide 'found error that couldn't be correct' outputs, so perhaps it >> was >> FEC, and not just a data-integrity check. >> >> The system may have had FEC on the radio links, for the exact same >> reason as >> too (BER too high), but I didn't feel like reading the Volume on >> radio links >> to find out! :-) >> >> Noel > From jnc at mercury.lcs.mit.edu Thu Mar 12 00:01:41 2009 From: jnc at mercury.lcs.mit.edu (Noel Chiappa) Date: Thu, 12 Mar 2009 03:01:41 -0400 (EDT) Subject: [ih] Baran and arbitrary reliability from arbitrarily unreliable components Message-ID: <20090312070141.3025F6BE603@mercury.lcs.mit.edu> > From: > I gather from your responses that the intellectual influence of Baran > is significant, and that many of the now commonplace ideas about > networking are due largely to him (and Pouzin). Well, many, but not all. I don't have time right now to accurately assign credit for all of contemporary networking... :-) > I recall having read in several places that ARPA had not been > influenced by Baran's work at all, and Roberts only had been made aware > of Baran's work by Davies in the 1967. This is a complex subject, and I think it's fair to say that most historians of technology who have looked at this think Baran's work _probably_ (more about this below) did have considerable influence on the ARPAnet work, as do I. To start with, I will point you at the two best academic histories (Arthur Norberg, Judy E. O'Neill, "Transforming Computer Technology: Information Processing for the Pentagon" and Janet Abbate, "Inventing the Internet"), both of which give the history in quite some detail. M. Mitchell Waldrop's biography of Licklider, "The Dream Machine", also covers it. There was a lengthy discussion of the topic of credit for the concepts of packet switching at: http://en.wikipedia.org/wiki/Talk:Packet_switching/OriginsArchive and although you can't trust anything you read in Wikipedia, the debate there cites sources you can go read for yourself. You might also want to look at this oral history interview with Baran: http://www.cbi.umn.edu/oh/display.phtml?id=110 where he does things like list entries from his personal calendars showing various ARPA network luminaries visiting him, in the period before the ARPA network was specified. (Human memories are so fallible, _contemporary_ written records are the 'gold standard' for historians.) However, having said all that... I don't think that at this distance we will ever be able to say with _absolute certainty_ what his influence was. That is because we'll never be able to find/prove things like 'A read Baran's document, and mentioned it in the hallway to B, who then said something to C at ARPA', or 'P read the Baran document, forgot the details in his concious mind, but then subconciously reused the ideas later when they started thinking about networking'. All we can say _for sure_ is 'these ideas were laid out by Baran before anyone else that we have definite record of, and his ideas were _widely_ propogated some years before the ARPA network work'. Although that's not necessarily definitive as to influence, because Davies was pretty sure he'd never heard of Baran's work when he started to think about networking some years later. (Davies heard about Baran from someone at the MoD, when he gave a presentation there.) So clearly not everyone heard about them. But maybe he just didn't remember a fifth-hand conversation in a hallway some years before... :-( > Has the RFQ in 1968 been written in complete ignorance of Baran's work? My recollection (from reading the stuff above) some years ago that by the time they got to the stage of writing the RFQ, Baran had been 'discovered' and consulted. But read the histories above, they will have the detail. > After all, it envisages packet orientation and distributed routing. Yes... but the routing mechanism (and many of the other details) of the ARPAnet were rather different. (Read the IMP paper in IFIPS to find out how.) Noel From jeanjour at comcast.net Thu Mar 12 05:49:13 2009 From: jeanjour at comcast.net (John Day) Date: Thu, 12 Mar 2009 08:49:13 -0400 Subject: [ih] Baran and arbitrary reliability from arbitrarily unreliable components In-Reply-To: <20090312070141.3025F6BE603@mercury.lcs.mit.edu> References: <20090312070141.3025F6BE603@mercury.lcs.mit.edu> Message-ID: I think Noel is dead on. It is reasonably clear that Baran influenced ARPA, but that Davies not having access to RAND reports came up with the idea independently. At the level of ARPA deciding to fund such a thing, the influence is going to be nebulous as he describes after all it is a management level. They are looking at the idea from a high level. They aren't going to worry about the details. If you want to see if Baran's ideas were implemented then it would be ARPA that you want to look at but BBN. Noel also brings up another good point: documentation. Believe it or not, but we have become a much more oral society. With people interacting more using transient communication ranging from more frequent meetings, to email, to messaging. The usual documentation of what happened tends not to exist. The actual events that lead to something are seldom documented. Just as a minor example, I remember how shocked I was when I saw the first "minutes" of a standards meeting. Because none of the social dynamics, the politics, the game playing were of course reflected there. That was clearly the real story, the story that historians would eventually be interested in. Because those were the real reasons things turned out as they did. And in this field, there is a tendency to make it up or argue that "they must have been thinking X" based on where we ended up. When nothing could be further from the case. Every time I read one of these accounts I think of Tom Stoppard's Arcadia. At 3:01 -0400 2009/03/12, Noel Chiappa wrote: > > From: > > > I gather from your responses that the intellectual influence of Baran > > is significant, and that many of the now commonplace ideas about > > networking are due largely to him (and Pouzin). > >Well, many, but not all. I don't have time right now to accurately assign >credit for all of contemporary networking... :-) > > > > I recall having read in several places that ARPA had not been > > influenced by Baran's work at all, and Roberts only had been made aware > > of Baran's work by Davies in the 1967. > >This is a complex subject, and I think it's fair to say that most historians >of technology who have looked at this think Baran's work _probably_ (more >about this below) did have considerable influence on the ARPAnet work, as do >I. > >To start with, I will point you at the two best academic histories (Arthur >Norberg, Judy E. O'Neill, "Transforming Computer Technology: Information >Processing for the Pentagon" and Janet Abbate, "Inventing the Internet"), >both of which give the history in quite some detail. M. Mitchell Waldrop's >biography of Licklider, "The Dream Machine", also covers it. > >There was a lengthy discussion of the topic of credit for the concepts of >packet switching at: > > http://en.wikipedia.org/wiki/Talk:Packet_switching/OriginsArchive > >and although you can't trust anything you read in Wikipedia, the debate there >cites sources you can go read for yourself. You might also want to look at >this oral history interview with Baran: > > http://www.cbi.umn.edu/oh/display.phtml?id=110 > >where he does things like list entries from his personal calendars showing >various ARPA network luminaries visiting him, in the period before the ARPA >network was specified. (Human memories are so fallible, _contemporary_ >written records are the 'gold standard' for historians.) > > >However, having said all that... I don't think that at this distance we will >ever be able to say with _absolute certainty_ what his influence was. > >That is because we'll never be able to find/prove things like 'A read Baran's >document, and mentioned it in the hallway to B, who then said something to C >at ARPA', or 'P read the Baran document, forgot the details in his concious >mind, but then subconciously reused the ideas later when they started thinking >about networking'. > >All we can say _for sure_ is 'these ideas were laid out by Baran before >anyone else that we have definite record of, and his ideas were _widely_ >propogated some years before the ARPA network work'. > >Although that's not necessarily definitive as to influence, because Davies was >pretty sure he'd never heard of Baran's work when he started to think about >networking some years later. (Davies heard about Baran from someone at the >MoD, when he gave a presentation there.) So clearly not everyone heard about >them. But maybe he just didn't remember a fifth-hand conversation in a hallway >some years before... :-( > > > > Has the RFQ in 1968 been written in complete ignorance of Baran's work? > >My recollection (from reading the stuff above) some years ago that by the >time they got to the stage of writing the RFQ, Baran had been 'discovered' >and consulted. But read the histories above, they will have the detail. > > > After all, it envisages packet orientation and distributed routing. > >Yes... but the routing mechanism (and many of the other details) of the >ARPAnet were rather different. (Read the IMP paper in IFIPS to find out >how.) > > Noel From craig at aland.bbn.com Thu Mar 12 06:34:00 2009 From: craig at aland.bbn.com (Craig Partridge) Date: Thu, 12 Mar 2009 09:34:00 -0400 Subject: [ih] Baran and arbitrary reliability from arbitrarily unreliable components In-Reply-To: Your message of "Thu, 12 Mar 2009 08:49:13 EDT." Message-ID: <20090312133400.7F17E28E155@aland.bbn.com> In message , John Day writes: >I think Noel is dead on. It is reasonably clear that Baran >influenced ARPA, but that Davies not having access to RAND reports >came up with the idea independently. > >At the level of ARPA deciding to fund such a thing, the influence is >going to be nebulous as he describes after all it is a management >level. They are looking at the idea from a high level. They aren't >going to worry about the details. If you want to see if Baran's ideas >were implemented then it would be ARPA that you want to look at but >BBN. Actually, ARPA apparently did worry about some details -- per Bob Taylor's note about central vs. distributed control (embedded in this a larger note which I attach for context). Craig > Begin forwarded message: > > From: Bob Taylor > Date: October 6, 2004 2:45:03 AM EDT > To: David Farber > Subject: [IP] more on 35th Anniversary of the Internet (well the start=20 > of the Arpanet anyway djf) > > Hello Dave.=A0 I agree with you that Rick Adams was "right to the=20 > point".=A0 Here is some more ARPAnet history background. > > In February of 1966 I initiated the ARPAnet project.=A0 I was Director=20 > of ARPA's Information Processing Techniques Office (IPTO) from late '65=20 > to late '69.=A0 There were only two people involved in the decision to=20 > launch the ARPAnet:=A0 my boss, the Director of ARPA Charles Herzfeld,=20 > and me. > > From 1962 to 1970, beginning with J.C.R. Licklider, Ivan Sutherland,=20 > and then me, IPTO funded several of the first projects devoted to the=20 > creation of interactive computing -- then referred to as time-sharing.=A0= > =20 > In '64 - '65, I witnessed that within each local site when users were=20 > first connected by a time-sharing system, a community of people with=20 > common interests began to discover one another and interact through the=20 > medium of the computer.=A0 I was struck by the fact that this was a=20 > wonderfully new and powerful phenomenon.=A0 > > The next obvious step was to connect those sites with an interactive=20 > network.=A0 To me, computing was about communication, not arithmetic.=A0=20 > Hence the ARPAnet. > This theme is elaborated in a paper Lick and I wrote in 1968 entitled,=20 > "The Computer as a Communications Device".=A0 Google can find it for=20 > you.=A0 On the last couple of pages there is a scenario that is=20 > reminiscent of today's Internet. > > Numerous untruths have been disseminated about events surrounding the=20 > origins of the ARPAnet.=A0 Here are some facts: > =A0=A0=A0=A0=A0=A0=A0=A0=A0 The creation of the ARPAnet was not motivated= > by=20 > considerations of war.=A0 The ARPAnet was created to enable folks with=20 > common interests to connect to one another through interactive=20 > computing even when widely separated by geography. > > =A0=A0=A0=A0=A0=A0=A0=A0=A0 The singularly most important contribution to t= > he=20 > architectural design of the ARPAnet/Internet came from Wesley Clark:=A0=20 > the interface message processor (IMP).=A0 Wes is the designer of the LINC= > =20 > which was arguably the first personal computer.=A0 Wes' ARPAnet concept=20 > ensured the critically valuable distributed architecture of the=20 > ARPAnet.=A0 Prior to Wes' contribution, Larry Roberts, whom I hired in=20 > Dec '66 to be ARPAnet's program manager, was considering a single,=20 > central ARPAnet control computer at a military base in Nebraska.=A0=20 > Fortunately, Wes quickly disabused Roberts of this notion. > > > =A0=A0=A0 The most significant role in actually building the ARPAnet was=20 > played by Frank Heart and his Bolt, Beranek & Newman team:=A0 Severo=20 > Ornstein, Will Crowther, Bob Barker, Bernie Cosell, Dave Walden, and=20 > Bob Kahn.=A0 > > =A0=A0=A0 Two suspicious claims relating to the ARPAnet were an important= > =20 > part of the case for awarding the 2001 Draper Prize to Kahn and=20 > Kleinrock.=A0 > =A0=A0=A0=A0 1. Kahn has claimed far and wide to be "responsible for the= > =20 > systems=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 > =A0=A0=A0=A0=A0=A0=A0=A0 design of the ARPAnet" while a member of the BB&= > N team.=A0 Since > =A0=A0=A0=A0=A0=A0=A0=A0 no other team member agrees, I doubt the validit= > y of this=20 > claim.=A0 > =A0=A0=A0 2.=A0 Roberts and Kleinrock (close friends since college) began= > to=20 > claim > =A0=A0=A0=A0=A0=A0=A0=A0 in 1995, more than 30 years after the fact, that= > Kleinrock=20 > invented=A0=A0 > =A0=A0=A0=A0=A0=A0=A0=A0 packet switching.=A0 Most of us believe that Don= > ald Davies in=20 > England > =A0=A0=A0=A0=A0=A0=A0=A0 and Paul Baran in the U.S. independently invente= > d packet=20 > switching in > =A0=A0=A0=A0=A0=A0=A0=A0 the early '60s. > I believe these two claims are false but they are recorded as facts on=20 > the web sites of the National Academy of Engineering and the Computer=20 > History Museum.=A0 The worst property of self-promotion is that it takes=20 > credit away from the people who actually made the contributions.=A0=20 > Roberts, Kahn, and Kleinrock have, however, made other important=20 > contributions.=A0 These can only be tarnished by extravagant claims. > > =A0=A0=A0 Packet switching is an important part of modern networking, but i= > t=20 > is not the only key piece.=A0 The multiplicity of the applications and=20 > the openness of the standards also played critical roles in ARPAnet=20 > development, as did Steve Crocker's initiation and management of the=20 > RFC process. > > > =A0=A0=A0=A0=A0=A0=A0=A0=A0 I believe the first internet was created at Xer= > ox PARC, circa=20 > '75, when we connected, via PUP, the Ethernet with the ARPAnet.=A0 PUP=20 > (PARC Universal Protocol) was instrumental later in defining TCP (ask=20 > Metcalfe or Shoch, they were there).=A0 > > =A0=A0=A0=A0=A0=A0=A0=A0=A0 For the internet to grow, it also needed a ne= > tworked=20 > personal computer, a graphical user interface with WYSIWYG properties,=20 > modern word processing, and desktop publishing.=A0 These, along with the=20 > Ethernet, all came out of my lab at Xerox PARC in the '70s, and were=20 > commercialized over the next 20 years by Adobe, Apple, Cisco,=20 > Microsoft, Novell, Sun and other companies that were necessary to the=20 > development of the Internet. > > > =A0=A0=A0=A0=A0=A0=A0=A0The ARPAnet was not an internet.=A0 An internet is = > a connection=20 > between two or more computer networks.=A0 The ARPAnet, with help from=20 > thousands of people, slowly evolved into the Internet.=A0 Without the=20 > ARPAnet, the Internet would have been a much longer time in coming. > > > rwt From lpress at csudh.edu Thu Mar 12 06:57:09 2009 From: lpress at csudh.edu (Larry Press) Date: Thu, 12 Mar 2009 06:57:09 -0700 Subject: [ih] Baran and arbitrary reliability from arbitrarily unreliable components In-Reply-To: References: <20090312070141.3025F6BE603@mercury.lcs.mit.edu> Message-ID: <49B914B5.7070507@csudh.edu> I have a teaching note page on Internet history has a broader focus than just packet switching(http://bpastudio.csudh.edu/fac/lpress/471/hout/netHistory/) Do you think any of these folks were inspired by the "torn tape" system used to route From lpress at csudh.edu Thu Mar 12 07:01:44 2009 From: lpress at csudh.edu (Larry Press) Date: Thu, 12 Mar 2009 07:01:44 -0700 Subject: [ih] Whoops In-Reply-To: <49B914B5.7070507@csudh.edu> References: <20090312070141.3025F6BE603@mercury.lcs.mit.edu> <49B914B5.7070507@csudh.edu> Message-ID: <49B915C8.7040508@csudh.edu> Sorry -- I sent the last message accidentally. Do you think any of these folks were inspired by the "torn tape" system used to route telegrams? There is a bit on torn tape at: http://bpastudio.csudh.edu/fac/lpress/471/hout/netHistory/ Larry From jnc at mercury.lcs.mit.edu Thu Mar 12 07:29:30 2009 From: jnc at mercury.lcs.mit.edu (Noel Chiappa) Date: Thu, 12 Mar 2009 10:29:30 -0400 (EDT) Subject: [ih] Baran and arbitrary reliability from arbitrarily unreliable components Message-ID: <20090312142930.C290C6BE594@mercury.lcs.mit.edu> > From: Craig Partridge > Actually, ARPA apparently did worry about some details It's clear if you read the DARPA RFQ that they had put an enormous amount of thought/design into the network before the RFQ went out; the RFQ is very detailed (although some changes were made as BBN actually implemented the network). In addition to the books I already mentioned, one of the best sources I know of for detail about the creation of the ARPANet is probably Katie Hafner's book, "Where Wizards Stay up Late". I know she went to an enormous amount of trouble to research it, including many, many interviews. (Alas, for a variety of reasons the resulting book did not match other great histories of technology - her publisher pushed for a more 'popular' treatment - but her research materials, including interviews, were all deposited in an academic institution; I think the University of Texas.) Looking quickly, the best secondary source for detail on the work at DARPA just before the RFQ went out appears to be Hafner, although none have a great deal of detail on that phase of the work (inside DARPA, creating the RFQ). Noel From jeanjour at comcast.net Thu Mar 12 07:58:34 2009 From: jeanjour at comcast.net (John Day) Date: Thu, 12 Mar 2009 10:58:34 -0400 Subject: [ih] Baran and arbitrary reliability from arbitrarily unreliable components In-Reply-To: <20090312133400.7F17E28E155@aland.bbn.com> References: <20090312133400.7F17E28E155@aland.bbn.com> Message-ID: See what I get for assuming! ;-) I should have known they couldn't keep their hands out of it. ;-0 But then reading it closely, this is really what I had in mind, levels of abstraction (a seeming a lost concept in CS these days). The ARPA people were outlining the general properties: distributed control, packet switching, etc. But not specifying details like what routing algorithm, how to do error correction, etc. Thanks Craig, good to see this. John At 9:34 -0400 2009/03/12, Craig Partridge wrote: >In message , John Day writes: > >>I think Noel is dead on. It is reasonably clear that Baran >>influenced ARPA, but that Davies not having access to RAND reports >>came up with the idea independently. >> >>At the level of ARPA deciding to fund such a thing, the influence is >>going to be nebulous as he describes after all it is a management >>level. They are looking at the idea from a high level. They aren't >>going to worry about the details. If you want to see if Baran's ideas >>were implemented then it would be ARPA that you want to look at but >>BBN. > >Actually, ARPA apparently did worry about some details -- per Bob >Taylor's note about central vs. distributed control (embedded in this >a larger note which I attach for context). > >Craig > >> Begin forwarded message: >> >> From: Bob Taylor >> Date: October 6, 2004 2:45:03 AM EDT >> To: David Farber >> Subject: [IP] more on 35th Anniversary of the Internet (well the start=20 >> of the Arpanet anyway djf) >> >> Hello Dave.=A0 I agree with you that Rick Adams was "right to the=20 >> point".=A0 Here is some more ARPAnet history background. >> >> In February of 1966 I initiated the ARPAnet project.=A0 I was Director=20 >> of ARPA's Information Processing Techniques Office (IPTO) from late '65=20 >> to late '69.=A0 There were only two people involved in the decision to=20 >> launch the ARPAnet:=A0 my boss, the Director of ARPA Charles Herzfeld,=20 >> and me. >> >> From 1962 to 1970, beginning with J.C.R. Licklider, Ivan Sutherland,=20 >> and then me, IPTO funded several of the first projects devoted to the=20 >> creation of interactive computing -- then referred to as time-sharing.=A0= >> =20 >> In '64 - '65, I witnessed that within each local site when users were=20 >> first connected by a time-sharing system, a community of people with=20 >> common interests began to discover one another and interact through the=20 >> medium of the computer.=A0 I was struck by the fact that this was a=20 >> wonderfully new and powerful phenomenon.=A0 >> >> The next obvious step was to connect those sites with an interactive=20 >> network.=A0 To me, computing was about communication, not arithmetic.=A0=20 >> Hence the ARPAnet. >> This theme is elaborated in a paper Lick and I wrote in 1968 entitled,=20 >> "The Computer as a Communications Device".=A0 Google can find it for=20 >> you.=A0 On the last couple of pages there is a scenario that is=20 >> reminiscent of today's Internet. >> >> Numerous untruths have been disseminated about events surrounding the=20 >> origins of the ARPAnet.=A0 Here are some facts: >> =A0=A0=A0=A0=A0=A0=A0=A0=A0 The creation of the ARPAnet was not motivated= >> by=20 >> considerations of war.=A0 The ARPAnet was created to enable folks with=20 >> common interests to connect to one another through interactive=20 >> computing even when widely separated by geography. >> >> =A0=A0=A0=A0=A0=A0=A0=A0=A0 The singularly most important contribution to t= >> he=20 >> architectural design of the ARPAnet/Internet came from Wesley Clark:=A0=20 >> the interface message processor (IMP).=A0 Wes is the designer of the LINC= >> =20 >> which was arguably the first personal computer.=A0 Wes' ARPAnet concept=20 >> ensured the critically valuable distributed architecture of the=20 > > ARPAnet.=A0 Prior to Wes' contribution, Larry Roberts, whom I hired in=20 >> Dec '66 to be ARPAnet's program manager, was considering a single,=20 >> central ARPAnet control computer at a military base in Nebraska.=A0=20 >> Fortunately, Wes quickly disabused Roberts of this notion. >> >> >> =A0=A0=A0 The most significant role in actually building the ARPAnet was=20 >> played by Frank Heart and his Bolt, Beranek & Newman team:=A0 Severo=20 >> Ornstein, Will Crowther, Bob Barker, Bernie Cosell, Dave Walden, and=20 >> Bob Kahn.=A0 >> >> =A0=A0=A0 Two suspicious claims relating to the ARPAnet were an important= >> =20 >> part of the case for awarding the 2001 Draper Prize to Kahn and=20 >> Kleinrock.=A0 >> =A0=A0=A0=A0 1. Kahn has claimed far and wide to be "responsible for the= >> =20 >> systems=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0 >> =A0=A0=A0=A0=A0=A0=A0=A0 design of the ARPAnet" while a member of the BB&= >> N team.=A0 Since >> =A0=A0=A0=A0=A0=A0=A0=A0 no other team member agrees, I doubt the validit= >> y of this=20 >> claim.=A0 >> =A0=A0=A0 2.=A0 Roberts and Kleinrock (close friends since college) began= >> to=20 >> claim >> =A0=A0=A0=A0=A0=A0=A0=A0 in 1995, more than 30 years after the fact, that= >> Kleinrock=20 >> invented=A0=A0 >> =A0=A0=A0=A0=A0=A0=A0=A0 packet switching.=A0 Most of us believe that Don= >> ald Davies in=20 >> England >> =A0=A0=A0=A0=A0=A0=A0=A0 and Paul Baran in the U.S. independently invente= >> d packet=20 >> switching in >> =A0=A0=A0=A0=A0=A0=A0=A0 the early '60s. >> I believe these two claims are false but they are recorded as facts on=20 >> the web sites of the National Academy of Engineering and the Computer=20 >> History Museum.=A0 The worst property of self-promotion is that it takes=20 >> credit away from the people who actually made the contributions.=A0=20 >> Roberts, Kahn, and Kleinrock have, however, made other important=20 >> contributions.=A0 These can only be tarnished by extravagant claims. >> >> =A0=A0=A0 Packet switching is an important part of modern networking, but i= >> t=20 >> is not the only key piece.=A0 The multiplicity of the applications and=20 >> the openness of the standards also played critical roles in ARPAnet=20 >> development, as did Steve Crocker's initiation and management of the=20 >> RFC process. >> >> >> =A0=A0=A0=A0=A0=A0=A0=A0=A0 I believe the first internet was created at Xer= >> ox PARC, circa=20 >> '75, when we connected, via PUP, the Ethernet with the ARPAnet.=A0 PUP=20 >> (PARC Universal Protocol) was instrumental later in defining TCP (ask=20 >> Metcalfe or Shoch, they were there).=A0 >> >> =A0=A0=A0=A0=A0=A0=A0=A0=A0 For the internet to grow, it also needed a ne= >> tworked=20 >> personal computer, a graphical user interface with WYSIWYG properties,=20 >> modern word processing, and desktop publishing.=A0 These, along with the=20 >> Ethernet, all came out of my lab at Xerox PARC in the '70s, and were=20 >> commercialized over the next 20 years by Adobe, Apple, Cisco,=20 >> Microsoft, Novell, Sun and other companies that were necessary to the=20 >> development of the Internet. >> >> >> =A0=A0=A0=A0=A0=A0=A0=A0The ARPAnet was not an internet.=A0 An internet is = >> a connection=20 >> between two or more computer networks.=A0 The ARPAnet, with help from=20 >> thousands of people, slowly evolved into the Internet.=A0 Without the=20 >> ARPAnet, the Internet would have been a much longer time in coming. >> >> >> rwt From jnc at mercury.lcs.mit.edu Thu Mar 12 08:31:17 2009 From: jnc at mercury.lcs.mit.edu (Noel Chiappa) Date: Thu, 12 Mar 2009 11:31:17 -0400 (EDT) Subject: [ih] Baran and arbitrary reliability from arbitrarily unreliable components Message-ID: <20090312153117.582296BE5E4@mercury.lcs.mit.edu> > From: John Day > The ARPA people were outlining the general properties: distributed > control, packet switching, etc. But not specifying details like what > routing algorithm, how to do error correction, etc. Actually, if you read the RFQ, that's not quite accurate. The main body of the proposal is mostly written in fairly high-level terms, but Appendixes cover a great deal of detail, including IMP-IMP packet formats, the checksum algorithm to be used on the IMP-IMP links, a briefly sketched routing algorithm, a network connectivity map, etc. One Appendix starts out by saying: "This section, like the others, is designed to provide guidlines to the contractor who may, if he [sic] wishes, modify any part of the description in his proposal. Each modification will be evaluated, and will be considered in the final contractor selection." So the Appendixes weren't hard-and-fast requirements, but they weren't 'just for illustration' either. Interesting aside, from the body of the RFQ: "Since the HOST's staff can program the IMP [!! - JNC], it is recommended that physical memory protection be provided." That one got changed pretty quickly! :-) Noel From adrian.j.hooke at jpl.nasa.gov Wed Mar 11 06:30:08 2009 From: adrian.j.hooke at jpl.nasa.gov (Adrian J. Hooke) Date: Wed, 11 Mar 2009 09:30:08 -0400 Subject: [ih] Baran and arbitrary reliability from arbitrarily unreliable components In-Reply-To: <20090311022723.625A26BE5DE@mercury.lcs.mit.edu> References: <20090311022723.625A26BE5DE@mercury.lcs.mit.edu> Message-ID: <6.2.3.4.2.20090311090031.029b7008@mail.jpl.nasa.gov> At 10:27 PM 3/10/2009, Noel Chiappa wrote: > That is an important contribution, but not, to me, his most important. I >would say that his most important single contribution was the concept of >breaking a long "message" up into small, relatively fixed-size, >self-describing segments (later named "packets"). As an aside, this is a concept which - totally independently - was being developed by some folk in the space communications community in the early 1970s (attached). "Block" telemetry subsequently morphed into "Packet Telemetry" and "Packet Telecommand" around 1976 and has become the underpinning of most current international space communications: http://public.ccsds.org/publications/SLS.aspx. These space communications protocols have now re-converged with Internet history and provide a platform on which we are developing the "Interplanetary Internet": http://www.nasa.gov/home/hqnews/2008/nov/HQ_08-298_Deep_space_internet.html Adrian Hooke -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: Block TLM.pdf Type: application/octet-stream Size: 881711 bytes Desc: not available URL: From jnc at mercury.lcs.mit.edu Wed Mar 18 15:51:07 2009 From: jnc at mercury.lcs.mit.edu (Noel Chiappa) Date: Wed, 18 Mar 2009 18:51:07 -0400 (EDT) Subject: [ih] Baran and arbitrary reliability from arbitrarily unreliable components Message-ID: <20090318225107.EE9146BE599@mercury.lcs.mit.edu> > From: John Day > Davies not having access to RAND reports Oh, one minor point for the record on this. A summary of Baran's work was published in: Paul Baran, "On Distributed Communications Networks", ('IEEE Transactions on Communications Systems', March 1964) Note the date - in Baran's oral history interview he notes (pp. 40-41) that one work, a book (by Donald Davies, from context) had a typo and listed the paper as being in 1969, an error that has since been widely copied (with the obvious erroneous conclusions drawn) - to Baran's amusement! ("It's fun to see many people refer to that paper with the 1969 date year after year in footnotes and in bibliographies. It's obvious that they haven't read the paper, only the reference to it.") I don't know how widespread ToCS was back then, but to reply to your observation above, it wasn't necessary to have had access to Baran's reports to find out about his work back in '64 in pretty fair detail (I have read the ToCS paper, and it covers the main points). Even more problematic, an abstract of Baran's '64 IEEE ToCS paper had been published in IEEE Spectrum in August '64. I seem to recall doing some research to find out the circulation figures for that journal back then, and I apparently discovered (no notes as to where I found this) that it was about 160,000 in those days! I haven't seen that abstract though, to know how detailed it was. Mind, I have no reason to doubt Davies' memory that he did not recall seeing anything about Baran's work - but as I pointed out before, it's almost certainly impossible to prove, at this remove, that _some inkling_ of it did not reach him somehow, and he simply didn't remember. Still, Baran's ideas had been circulated quite openly and widely through professional journals years before. Noel From dhc2 at dcrocker.net Wed Mar 18 18:32:52 2009 From: dhc2 at dcrocker.net (Dave CROCKER) Date: Wed, 18 Mar 2009 18:32:52 -0700 Subject: [ih] Baran and arbitrary reliability from arbitrarily unreliable components In-Reply-To: <49B914B5.7070507@csudh.edu> References: <20090312070141.3025F6BE603@mercury.lcs.mit.edu> <49B914B5.7070507@csudh.edu> Message-ID: <49C1A0C4.7080805@dcrocker.net> Larry Press wrote: > I have a teaching note page on Internet history has a broader focus than > just packet > switching(http://bpastudio.csudh.edu/fac/lpress/471/hout/netHistory/) Nice list. I've found it useful to distinguish 3 major paradigm increments in the history of the Internet's data-handling basics (separate from the history of its apps): Packet network - Heterogeneous machines form a single network; Common canonical formats and basic handling; Arpanet, Irvine ring, Alohanet, Cyclades, et al. Interconnected networks Heterogeneous networks; Meta-formats and independent administrations; Arpa Internet Interconnected services Multiple backbones & regional operations - NSFnet Internet Paul and the suite of early contributors gave us that first insight. My understanding from my brother is that the mid-60s had a number of focused, experimental efforts to explore this space, before the Arpanet contract was finally let. (I tend to view that NSFNet step as having core technical impact that is generally under-appreciated, since it sowed the seeds for the richly competitive infrastructure, without which we might have a single-operator backbone...) Noel Chiappa wrote: > In addition to the books I already mentioned, one of the best sources I know > of for detail about the creation of the ARPANet is probably Katie Hafner's > book, "Where Wizards Stay up Late". I know she went to an enormous amount of > trouble to research it, including many, many interviews. What we are missing is Volume 2, about the creation of the Internet. It only sits around in isolated pieces, as anecdotes, personal files, and fading memories. What we need is something with the richness of Katie's book, that covers the remarkable history that moved a relatively obscure technical mechanism into a global infrastructure. This was as much an innovation in collaborative culture as it was computer technology. In 1997, a police sergeant who was chatting with me, on the eastern side of the the Malaysian peninsula, said that he knew that we had the Internet in the U.S., but wasn't sure whether it had been invented in the U.S. or in Malaysia. It's difficult to think of a better indication of successful diffusion of innovation. Both the formal, historical steps and the rich array of anecdotes really ought to be recorded. (BTW, Katie said that sales of Wizards wasn't great and Wired wouldn't even do an article about it: It was only about the past...) d/ -- Dave Crocker Brandenburg InternetWorking bbiw.net From vint at google.com Wed Mar 18 18:48:34 2009 From: vint at google.com (Vint Cerf) Date: Wed, 18 Mar 2009 21:48:34 -0400 Subject: [ih] Baran and arbitrary reliability from arbitrarily unreliable components In-Reply-To: <49C1A0C4.7080805@dcrocker.net> References: <20090312070141.3025F6BE603@mercury.lcs.mit.edu> <49B914B5.7070507@csudh.edu> <49C1A0C4.7080805@dcrocker.net> Message-ID: <3DEFF784-5DD4-4760-9621-16542A236065@google.com> +1 Vint Cerf Google 1818 Library Street, Suite 400 Reston, VA 20190 202-370-5637 vint at google.com On Mar 18, 2009, at 9:32 PM, Dave CROCKER wrote: > > > Larry Press wrote: >> I have a teaching note page on Internet history has a broader focus >> than just packet switching(http://bpastudio.csudh.edu/fac/lpress/471/hout/netHistory/ >> ) > > Nice list. > > I've found it useful to distinguish 3 major paradigm increments in > the history of the Internet's data-handling basics (separate from > the history of its apps): > > Packet network - > Heterogeneous machines form a single network; > Common canonical formats and basic handling; > Arpanet, Irvine ring, Alohanet, Cyclades, et al. > > Interconnected networks > Heterogeneous networks; > Meta-formats and independent administrations; > Arpa Internet > > Interconnected services > Multiple backbones & regional operations - NSFnet Internet > > Paul and the suite of early contributors gave us that first > insight. My understanding from my brother is that the mid-60s had a > number of focused, experimental efforts to explore this space, > before the Arpanet contract was finally let. > > (I tend to view that NSFNet step as having core technical impact > that is generally under-appreciated, since it sowed the seeds for > the richly competitive infrastructure, without which we might have a > single-operator backbone...) > > > > Noel Chiappa wrote: > > In addition to the books I already mentioned, one of the best > sources I know > > of for detail about the creation of the ARPANet is probably Katie > Hafner's > > book, "Where Wizards Stay up Late". I know she went to an enormous > amount of > > trouble to research it, including many, many interviews. > > What we are missing is Volume 2, about the creation of the > Internet. It only sits around in isolated pieces, as anecdotes, > personal files, and fading memories. What we need is something with > the richness of Katie's book, that covers the remarkable history > that moved a relatively obscure technical mechanism into a global > infrastructure. This was as much an innovation in collaborative > culture as it was computer technology. > > In 1997, a police sergeant who was chatting with me, on the eastern > side of the the Malaysian peninsula, said that he knew that we had > the Internet in the U.S., but wasn't sure whether it had been > invented in the U.S. or in Malaysia. It's difficult to think of a > better indication of successful diffusion of innovation. > > Both the formal, historical steps and the rich array of anecdotes > really ought to be recorded. > > (BTW, Katie said that sales of Wizards wasn't great and Wired > wouldn't even do an article about it: It was only about the past...) > > d/ > -- > > Dave Crocker > Brandenburg InternetWorking > bbiw.net From sbrim at cisco.com Thu Mar 19 04:43:34 2009 From: sbrim at cisco.com (Scott Brim) Date: Thu, 19 Mar 2009 07:43:34 -0400 Subject: [ih] Baran and arbitrary reliability from arbitrarily unreliable components In-Reply-To: <49C1A0C4.7080805@dcrocker.net> References: <20090312070141.3025F6BE603@mercury.lcs.mit.edu> <49B914B5.7070507@csudh.edu> <49C1A0C4.7080805@dcrocker.net> Message-ID: <20090319114334.GF44014@cisco.com> Excerpts from Dave CROCKER on Wed, Mar 18, 2009 06:32:52PM -0700: > I've found it useful to distinguish 3 major paradigm increments in the > history of the Internet's data-handling basics (separate from the history > of its apps): > > Packet network - > Heterogeneous machines form a single network; > Common canonical formats and basic handling; > Arpanet, Irvine ring, Alohanet, Cyclades, et al. > > Interconnected networks > Heterogeneous networks; > Meta-formats and independent administrations; > Arpa Internet > > Interconnected services > Multiple backbones & regional operations - NSFnet Internet > > Paul and the suite of early contributors gave us that first insight. My > understanding from my brother is that the mid-60s had a number of > focused, experimental efforts to explore this space, before the Arpanet > contract was finally let. > > (I tend to view that NSFNet step as having core technical impact that is > generally under-appreciated, since it sowed the seeds for the richly > competitive infrastructure, without which we might have a single-operator > backbone...) Does this imply an opinion about, e.g., CSnet? I think NSFNet was not in itself the big step, rather it was the unexpected creation of the regional networks as separate from the backbone and in many ways more significant. Scott From cls at rkey.com Thu Mar 19 05:37:51 2009 From: cls at rkey.com (Craig Simon) Date: Thu, 19 Mar 2009 08:37:51 -0400 Subject: [ih] Baran and arbitrary reliability from arbitrarily unreliable components In-Reply-To: <49C1A0C4.7080805@dcrocker.net> References: <20090312070141.3025F6BE603@mercury.lcs.mit.edu> <49B914B5.7070507@csudh.edu> <49C1A0C4.7080805@dcrocker.net> Message-ID: <49C23C9F.3080805@rkey.com> Dave CROCKER wrote: > What we are missing is Volume 2, about the creation of the Internet. It > only sits around in isolated pieces, as anecdotes, personal files, and > fading memories. What we need is something with the richness of Katie's > book, that covers the remarkable history that moved a relatively obscure > technical mechanism into a global infrastructure. This was as much an > innovation in collaborative culture as it was computer technology. It certainly doesn't set the standard for a Volume 2, but I can offer my Ph.D. dissertation as a useful resource to anyone moving in that direction. http://www.rkey.com/essays/diss.pdf Craig Simon From dhc2 at dcrocker.net Thu Mar 19 08:22:21 2009 From: dhc2 at dcrocker.net (Dave CROCKER) Date: Thu, 19 Mar 2009 08:22:21 -0700 Subject: [ih] Baran and arbitrary reliability from arbitrarily unreliable components In-Reply-To: <20090319114334.GF44014@cisco.com> References: <20090312070141.3025F6BE603@mercury.lcs.mit.edu> <49B914B5.7070507@csudh.edu> <49C1A0C4.7080805@dcrocker.net> <20090319114334.GF44014@cisco.com> Message-ID: <49C2632D.6090504@dcrocker.net> Scott Brim wrote: > Excerpts from Dave CROCKER on Wed, Mar 18, 2009 06:32:52PM -0700: >> (I tend to view that NSFNet step as having core technical impact that is >> generally under-appreciated, since it sowed the seeds for the richly >> competitive infrastructure, without which we might have a single-operator >> backbone...) > > Does this imply an opinion about, e.g., CSnet? Probably. For example, see: But, then, I'm not exactly objective about the role of CSNet... > I think NSFNet was not in itself the big step, rather it was the > unexpected creation of the regional networks as separate from the > backbone and in many ways more significant. My impression is that it was, itself, very much the big step in setting the final global architecture stone for the current, competitive Internet. I believe there were some other backbones lurking around, in addition to the BBN-run official one, but that they were dealt with in a very ad hoc manner, but that NSFNet's creation forced development of BGP, to explicitly accomodate multiple (ie, competing...) backbones. And the creation of the regionals rather explicitly set the stage for local ISPs, since the funding model for NSFNet (which was test-marketed by CSNet) was that funding from the government had a sunset date, by which other sources of funding needed to be found. d/ -- Dave Crocker Brandenburg InternetWorking bbiw.net From ian.peter at ianpeter.com Thu Mar 19 14:35:44 2009 From: ian.peter at ianpeter.com (Ian Peter) Date: Fri, 20 Mar 2009 08:35:44 +1100 Subject: [ih] Baran and Davies and respetive roles in Apra design In-Reply-To: Message-ID: >From what I have read here and elsewhere, it would appear that both Baran and Davies (via Peter Kerstein) met at times in the Arpanet design phase with various members of the Arpanet team to explain their similar theories on packet switching. I have also read that it was the Davies design which was adopted, not Baran's. Can anyone shed more light on this, who met who, and why if both designs were known Davies was preferred? Ian Peter PO Box 429 Bangalow NSW 2479 Australia Tel (+614) 1966 7772 or (+612) 6687 0773 www.ianpeter.com www.nethistory.info From vint at google.com Thu Mar 19 15:11:00 2009 From: vint at google.com (Vint Cerf) Date: Thu, 19 Mar 2009 18:11:00 -0400 Subject: [ih] Baran and Davies and respetive roles in Apra design In-Reply-To: References: Message-ID: + bob kahn bob was a key player in the design of the IMP and the ARPANET Paul's principal conceptual ideas related to packetization (message blocks) and hot-potato routing, I believe Donald Davies had a multi-node idea but implemented only one node owing to funding limits. He or his team via Roger Scantlebury influenced the ARPANET design be convincing Larry Roberts to use 50 kb/s circuits rather than 2.4 kb/s circuits. Donald also contributed the term "packet" into the literature. Vint Cerf Google 1818 Library Street, Suite 400 Reston, VA 20190 202-370-5637 vint at google.com On Mar 19, 2009, at 5:35 PM, Ian Peter wrote: >> From what I have read here and elsewhere, it would appear that both >> Baran > and Davies (via Peter Kerstein) met at times in the Arpanet design > phase > with various members of the Arpanet team to explain their similar > theories > on packet switching. I have also read that it was the Davies design > which > was adopted, not Baran's. > > Can anyone shed more light on this, who met who, and why if both > designs > were known Davies was preferred? > > > > > Ian Peter > PO Box 429 > Bangalow NSW 2479 > Australia > Tel (+614) 1966 7772 or (+612) 6687 0773 > www.ianpeter.com > www.nethistory.info > > > From jnc at mercury.lcs.mit.edu Thu Mar 19 16:02:28 2009 From: jnc at mercury.lcs.mit.edu (Noel Chiappa) Date: Thu, 19 Mar 2009 19:02:28 -0400 (EDT) Subject: [ih] Baran and Davies and respetive roles in Apra design Message-ID: <20090319230228.BC1236BE572@mercury.lcs.mit.edu> > From: "Ian Peter" > it would appear that both Baran and Davies (via Peter Kerstein) met at > times in the Arpanet design phase with various members of the Arpanet > team Davies' 'courier' was actually Roger Scantlebury, not Peter Kirstein (note correct spelling), who was later (and most heavily involved in the early Internet work, and was not involved in the early ARPANET). > I have also read that it was the Davies design which was adopted, not > Baran's. I'm not at all sure that's accurate. I don't recall the details of Davies' design, and I don't have time to research it, but I have seen Baran's and Davies' designs described as having "striking ... technical similarities". My _guess_, without researching it, is that the ARPANet design probably drew on both, plus ideas from others as well (e.g. the idea of IMPs was due to Wesley Clark). > Can anyone shed more light on this, who met who I think you will find that all covered in decent detail in the three works I've already listed: "Where Wizards Stay up Late", "Inventing the Internet", and "Transforming Computer Technology". I'd recommending getting ahold of them if you're really interested in this. Noel From lpress at csudh.edu Thu Mar 19 21:40:27 2009 From: lpress at csudh.edu (Larry Press) Date: Thu, 19 Mar 2009 21:40:27 -0700 Subject: [ih] Baran and Davies and respetive roles in Apra design In-Reply-To: References: Message-ID: <49C31E3B.9030808@csudh.edu> > Donald also contributed the term "packet" into the literature. I am still curious -- did you all know about routing of telegrams using torn paper tape, and, if so, did that have anything to do with the idea of packet switching? Larry From vint at google.com Fri Mar 20 04:06:00 2009 From: vint at google.com (Vint Cerf) Date: Fri, 20 Mar 2009 07:06:00 -0400 Subject: [ih] Baran and Davies and respetive roles in Apra design In-Reply-To: <49C31E3B.9030808@csudh.edu> References: <49C31E3B.9030808@csudh.edu> Message-ID: <3F3C0604-DF17-49A0-8FA6-B7EF4CE56CFB@google.com> yes of course we did know about "torn tape" and also about message switching in the form of AUTODIN. If you look at Len Kleinrock's book that emerged from his dissertation it was about stochastic flow and delay in message switched systems. An ARPANET innovation was to break messages up into "packets" for purposes of transmission to reduce transmission delay which was significant over low speed backbone trunks available at the time. Also, for purposes of noise resistance, the shorter packets had a higher probability of arriving intact than long messages. Communication Nets: Stochastic Message Flow and Delay This text develops a queuing theory model of communications nets, with realistic assessments that will benefit those working with computers and other communications systems. Topics include optimal channel capacity assignment, effect of priority and other queue disciplines, choice of routine procedure, fixed-cost restraint, and design of topological structures. 1964 edition. Publisher: Dover Publications ISBN: 0486458806 EAN: 9780486458809 No. of Pages: 209 vint Vint Cerf Google 1818 Library Street, Suite 400 Reston, VA 20190 202-370-5637 vint at google.com On Mar 20, 2009, at 12:40 AM, Larry Press wrote: >> Donald also contributed the term "packet" into the literature. > > > I am still curious -- did you all know about routing of telegrams > using torn paper tape, and, if so, did that have anything to do with > the idea of packet switching? > > Larry > From mbaer at cs.tu-berlin.de Fri Mar 20 08:48:58 2009 From: mbaer at cs.tu-berlin.de (=?ISO-8859-1?Q?Matthias_B=E4rwolff?=) Date: Fri, 20 Mar 2009 11:48:58 -0400 Subject: [ih] Baran and Davies and respetive roles in Apra design In-Reply-To: <3F3C0604-DF17-49A0-8FA6-B7EF4CE56CFB@google.com> References: <49C31E3B.9030808@csudh.edu> <3F3C0604-DF17-49A0-8FA6-B7EF4CE56CFB@google.com> Message-ID: <49C3BAEA.2040901@cs.tu-berlin.de> An aside: The Kleinrock book also features a little section somewhere in the beginning on how message switching in the telegraph system worked, including priority schemes and error recovery. Basically, they just had huge "buffers" and the internal switching speed was a lot faster than the line / line out speeds. And if I recall correctly, priority messages triggered an alarm upon which a human operator would simply take the message and put it right in the front of the proper outgoing queue. The processes were gradually automated. The reference he in turn points to is Gilbert S. Vernam, Automatic Telegraph Switching System Plan 55-A, Western Union Technical Review, Vol. 12 No. 2 April 1958, Pages 37--50. Matthias Vint Cerf wrote: > yes of course we did know about "torn tape" and also about message > switching in the form of AUTODIN. > > If you look at Len Kleinrock's book that emerged from his dissertation > it was about stochastic flow and delay in message switched systems. An > ARPANET innovation was to break messages up into "packets" for > purposes of transmission to reduce transmission delay which was > significant over low speed backbone trunks available at the time. > Also, for purposes of noise resistance, the shorter packets had a > higher probability of arriving intact than long messages. > > Communication Nets: Stochastic Message Flow and Delay > > This text develops a queuing theory model of communications nets, with > realistic assessments that will benefit those working with computers > and other communications systems. Topics include optimal channel > capacity assignment, effect of priority and other queue disciplines, > choice of routine procedure, fixed-cost restraint, and design of > topological structures. 1964 edition. > > Publisher: Dover Publications > ISBN: 0486458806 > EAN: 9780486458809 > No. of Pages: 209 > > vint > > Vint Cerf > Google > 1818 Library Street, Suite 400 > Reston, VA 20190 > 202-370-5637 > vint at google.com > > > > > On Mar 20, 2009, at 12:40 AM, Larry Press wrote: > >>> Donald also contributed the term "packet" into the literature. >> >> >> I am still curious -- did you all know about routing of telegrams >> using torn paper tape, and, if so, did that have anything to do with >> the idea of packet switching? >> >> Larry >> > -- -- Matthias B?rwolff mbaer at csail.mit.edu From vint at google.com Sun Mar 22 12:11:43 2009 From: vint at google.com (Vint Cerf) Date: Sun, 22 Mar 2009 15:11:43 -0400 Subject: [ih] Fwd: Baran and arbitrary reliability from arbitrarily unreliable components References: <000001c9ab1e$9c74f910$0402a8c0@Integral> Message-ID: <9A895941-4139-4408-B330-1CBEAEE4BA19@google.com> clears up some of the uncertainties Vint Cerf Google 1818 Library Street, Suite 400 Reston, VA 20190 202-370-5637 vint at google.com Begin forwarded message: > From: "roger scantlebury" > Date: March 22, 2009 2:47:08 PM EDT > To: "'Vint Cerf'" > Subject: RE: [ih] Baran and arbitrary reliability from arbitrarily > unreliable components > > Hi Vint > > We referenced Baran?s paper in our 1967 Gatlinburg ACM paper. You > will find it in the References. Therefore I am sure that we > introduced Baran?s work to Larry (and hence the BBN guys). > > We were unaware of Baran?s work when we started our own design work > in 1965, but were given a copy of his paper by one of our colleagues > in the UK Ministry of Defense (in 1966) while we were writing the > 1967 paper. Clearly Donald and Paul Baran had independently come to > a similar idea ? albeit for different purposes ? Paul for a > survivable voice/telex network, ours for a high-speed computer > network. > > I hope this explains the time-line?. > > Best Regards > > Roger > > -----Original Message----- > From: Vint Cerf [mailto:vint at google.com] > Sent: 19 March 2009 09:01 > To: Roger Scantlebury > Subject: Fwd: [ih] Baran and arbitrary reliability from arbitrarily > unreliable components > > Roger, > > Is it your understanding that Donald and the NPL team were unaware > of the Baran work at RAND during the period of development of the > network at NPL? I don't think this has to do with anyone fighting > over paternity. It is just a question about when the various > "packet" efforts became aware of each other. > > For example, your interaction with Larry Roberts in 1967 is the key > link that drew ARPA and NPL groups into mutual awareness I think. > > I had once thought that you might have drawn Roberts' attention to > Baran's work but I suppose not, if you were unaware of it in 1967. > In fact, it would be of interest to know when and how you (or > Donald) might have learned of it? > > thanks, hope this finds you well! > > v > > > Vint Cerf > Google > 1818 Library Street, Suite 400 > Reston, VA 20190 > 202-370-5637 > vint at google.com > > > > > > Begin forwarded message: > > > From: jnc at mercury.lcs.mit.edu (Noel Chiappa) > Date: March 18, 2009 6:51:07 PM EDT > To: internet-history at postel.org > Cc: jnc at mercury.lcs.mit.edu > Subject: Re: [ih] Baran and arbitrary reliability from arbitrarily > unreliable components > >> From: John Day > > > Davies not having access to RAND reports > > Oh, one minor point for the record on this. > > A summary of Baran's work was published in: > > Paul Baran, "On Distributed Communications Networks", ('IEEE > Transactions on > Communications Systems', March 1964) > > Note the date - in Baran's oral history interview he notes (pp. > 40-41) that > one work, a book (by Donald Davies, from context) had a typo and > listed the > paper as being in 1969, an error that has since been widely copied > (with the > obvious erroneous conclusions drawn) - to Baran's amusement! ("It's > fun to > see many people refer to that paper with the 1969 date year after > year in > footnotes and in bibliographies. It's obvious that they haven't read > the > paper, only the reference to it.") > > I don't know how widespread ToCS was back then, but to reply to your > observation above, it wasn't necessary to have had access to Baran's > reports > to find out about his work back in '64 in pretty fair detail (I have > read the > ToCS paper, and it covers the main points). > > Even more problematic, an abstract of Baran's '64 IEEE ToCS paper > had been > published in IEEE Spectrum in August '64. I seem to recall doing some > research to find out the circulation figures for that journal back > then, and > I apparently discovered (no notes as to where I found this) that it > was about > 160,000 in those days! I haven't seen that abstract though, to know > how > detailed it was. > > > Mind, I have no reason to doubt Davies' memory that he did not > recall seeing > anything about Baran's work - but as I pointed out before, it's almost > certainly impossible to prove, at this remove, that _some inkling_ > of it did > not reach him somehow, and he simply didn't remember. > > Still, Baran's ideas had been circulated quite openly and widely > through > professional journals years before. > > Noel > -------------- next part -------------- An HTML attachment was scrubbed... URL: