From bernie at fantasyfarm.com  Tue Sep  1 12:59:14 2009
From: bernie at fantasyfarm.com (Bernie Cosell)
Date: Tue, 01 Sep 2009 15:59:14 -0400
Subject: [ih] ARPAnet and nuclear survivability
Message-ID: <4A9D44D2.20130.1F28B896@bernie.fantasyfarm.com>

On another list the old question came up again and I'm not sure I've ever 
gotten a straight answer.  I have always thought that Licklider's intent 
was to build a collaboration/research network.   The early participants 
were pretty much all universities and I can't recall any military people 
involved in the design or development of it.  [OTOH the IMP *was* 
originally delivered in a fully mil-spec ruggedized case...]

Others have repeated the claim that the justification for the ARPAnet was 
to provide communications that could survive a nuclear attack and of 
course, there was Baran's paper [although he wasn't the only person 
developing ideas about packet switching]

BUT as far as i can remember there were *NO* military folks involved with 
anything to do with the early ARPAnet: I don't recall any military info 
in the reports [at least not the ones that we [BBN] prepared], nor did I 
remember any "requirements" that came from or were reviewed by military 
folk.  Does anyone know that bit of the tale?  [e.g., I have no idea how 
the ARPAnet was presented to Congress]

  /Bernie\

-- 
Bernie Cosell                     Fantasy Farm Fibers
mailto:bernie at fantasyfarm.com     Pearisburg, VA
    -->  Too many people, too few sheep  <--       





From vint at google.com  Tue Sep  1 13:27:08 2009
From: vint at google.com (Vint Cerf)
Date: Tue, 1 Sep 2009 16:27:08 -0400
Subject: [ih] ARPAnet and nuclear survivability
In-Reply-To: <4A9D44D2.20130.1F28B896@bernie.fantasyfarm.com>
References: <4A9D44D2.20130.1F28B896@bernie.fantasyfarm.com>
Message-ID: <AEF5204C-D44F-40BC-A0E9-3DF888691CF8@google.com>

larry roberts confirms that the purpose of arpanet was resource sharing.

the nuclear resilience story came from Paul Baran's work (earlier).

When I managed the Internet project at DARPA, however, I did want
to show resilience so we demonstrated this putting packet radios
in strategic air command aircraft, artificially breaking up the ARPANET
and gluing the pieces together using ground and air packet radios and
TCP/IP.

vint


On Sep 1, 2009, at 3:59 PM, Bernie Cosell wrote:

> On another list the old question came up again and I'm not sure I've  
> ever
> gotten a straight answer.  I have always thought that Licklider's  
> intent
> was to build a collaboration/research network.   The early  
> participants
> were pretty much all universities and I can't recall any military  
> people
> involved in the design or development of it.  [OTOH the IMP *was*
> originally delivered in a fully mil-spec ruggedized case...]
>
> Others have repeated the claim that the justification for the  
> ARPAnet was
> to provide communications that could survive a nuclear attack and of
> course, there was Baran's paper [although he wasn't the only person
> developing ideas about packet switching]
>
> BUT as far as i can remember there were *NO* military folks involved  
> with
> anything to do with the early ARPAnet: I don't recall any military  
> info
> in the reports [at least not the ones that we [BBN] prepared], nor  
> did I
> remember any "requirements" that came from or were reviewed by  
> military
> folk.  Does anyone know that bit of the tale?  [e.g., I have no idea  
> how
> the ARPAnet was presented to Congress]
>
>  /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  Tue Sep  1 14:59:37 2009
From: jnc at mercury.lcs.mit.edu (Noel Chiappa)
Date: Tue,  1 Sep 2009 17:59:37 -0400 (EDT)
Subject: [ih] ARPAnet and nuclear survivability
Message-ID: <20090901215937.417466BE615@mercury.lcs.mit.edu>

    > From: Vint Cerf <vint at google.com>

    > the purpose of arpanet was resource sharing.

Indeed; people seem to have a hard time keeping straight that Baran's work,
the ARPANet, and the Internet were three different systems with different
goals, etc.


    > the Internet project at DARPA, however, I did want to show resilience
    > so we demonstrated this

There were also lots of references in meetings at the time to C3I, military
applications, etc, etc. (As one would expect from DARPA at that point, since
Congress had directed that its work be more immediately military-relevant,
IIRC.)

There was also a contemporaneous proposal which I saw, which some people in
another group in LCS at MIT were associated with, to build a survivable
strategic command-and-control system; part of the concept was for data to be
kept in a distributed database which used replication for reliability in a
high-loss environment, but the communication was all TCP/IP, IIRC.

Between them, that all caused me a certain amount of heartburn at the time
(we're talking circa '79 or so), and pondering whether I really wanted to
work on something that could be used for strategic C+C. I wound up deciding
that it was morally OK to work on it because it was inherently a second-stike
thing (by definition, you pretty much don't need a survivable C+C system for
a first strike :-).

	Noel


From jnc at mercury.lcs.mit.edu  Tue Sep  1 15:03:29 2009
From: jnc at mercury.lcs.mit.edu (Noel Chiappa)
Date: Tue,  1 Sep 2009 18:03:29 -0400 (EDT)
Subject: [ih] ARPAnet and nuclear survivability
Message-ID: <20090901220329.CB2936BE615@mercury.lcs.mit.edu>

    > From: "Bernie Cosell" <bernie at fantasyfarm.com>

    > OTOH the IMP *was* originally delivered in a fully mil-spec ruggedized
    > case...

I think that was in mostly just for reliability/robustness; I think one
ARPANet history (maybe 'Where Wizards Stay Up Late') discusses the reasoning.

If that old story about the milkshake is true, it might have been a wise
choice, though... :-)

	Noel


From lpress at csudh.edu  Tue Sep  1 16:01:19 2009
From: lpress at csudh.edu (Larry Press)
Date: Tue, 1 Sep 2009 16:01:19 -0700
Subject: [ih] ARPAnet and nuclear survivability
In-Reply-To: <AEF5204C-D44F-40BC-A0E9-3DF888691CF8@google.com>
References: <4A9D44D2.20130.1F28B896@bernie.fantasyfarm.com>
	<AEF5204C-D44F-40BC-A0E9-3DF888691CF8@google.com>
Message-ID: <4A9DA7BF.6090801@csudh.edu>



Vint Cerf wrote:
> larry roberts confirms that the purpose of arpanet was resource sharing.

In 1996, I read over some old papers and wrote an article called 
"Seeding Networks: the Federal Role," Communications of the ACM, pp 
11-18, Vol 39., No. 10, October, 1996, 
http://bpastudio.csudh.edu/fac/lpress/articles/govt.htm.


After reading those papers, I concluded:

"It is often stated that ARPA's interest in networking was motivated by 
the need for a military communication system that could withstand 
attack. While that motivation was clearly stated in a series of Air 
Force reports by Paul Baran [2] outlining the packet-switching 
architecture (analogous to the torn paper-tape telegraph systems of the 
day) that was chosen for the ARPANET, it was not what motivated the 
ARPANET. The goal stated in most ARPANET papers is resource sharing. 
With the exception of Larry Roberts, the importance of email, the killer 
application, was generally unanticipated by the original technicians 
(though not by Licklider). Robustness under attack was a TCP/IP goal, 
but that came later".



For example, Roberts and Wessler wrote this article:

Roberts, Lawrence G., and Wessler, Barry D., "Computer Network 
Development to Achieve Resource Sharing," Proceedings of the 1970 Spring 
Joint Computer Conference, 543-549.

Lar



From dhc2 at dcrocker.net  Wed Sep  2 08:28:17 2009
From: dhc2 at dcrocker.net (Dave CROCKER)
Date: Wed, 02 Sep 2009 08:28:17 -0700
Subject: [ih] ARPAnet and nuclear survivability
In-Reply-To: <4A9D44D2.20130.1F28B896@bernie.fantasyfarm.com>
References: <4A9D44D2.20130.1F28B896@bernie.fantasyfarm.com>
Message-ID: <4A9E8F11.5060307@dcrocker.net>



Bernie Cosell wrote:
> On another list the old question came up again and I'm not sure I've ever 
> gotten a straight answer.  I have always thought that Licklider's intent 
> was to build a collaboration/research network.  


Since the Arpanet had a number of different people in lead positions, it would 
not be surprising if they had different agendas or chose to present things in 
different ways to particular audiences.

When I got involved, in the early 70's, I would periodically hear some folk 
characterize the capability as designed to "survive hostile battlefield 
conditions".  I took that to mean convention battlefields, not nuclear.  (I've 
heard that in the first Iraq War, the US was unable to knock out the Iraque data 
net; survivability was thereby demonstrated...)

I believe I've heard that Lukasik would claim the goal of nuclear survivability 
during presentations to the military or Congress or somesuch, as a funding pitch.

d/

-- 

   Dave Crocker
   Brandenburg InternetWorking
   bbiw.net


From vint at google.com  Wed Sep  2 08:55:54 2009
From: vint at google.com (Vint Cerf)
Date: Wed, 2 Sep 2009 11:55:54 -0400
Subject: [ih] ARPAnet and nuclear survivability
In-Reply-To: <4A9E8F11.5060307@dcrocker.net>
References: <4A9D44D2.20130.1F28B896@bernie.fantasyfarm.com>
	<4A9E8F11.5060307@dcrocker.net>
Message-ID: <97CC4C99-08CE-4DFE-918D-4858570EBA89@google.com>

we'd have to ask steve lukasik. Larry Roberts was adamant that his  
motivation was resource sharing and that comports well with  
Licklider's writings.

vint

On Sep 2, 2009, at 11:28 AM, Dave CROCKER wrote:

>
>
> Bernie Cosell wrote:
>> On another list the old question came up again and I'm not sure  
>> I've ever gotten a straight answer.  I have always thought that  
>> Licklider's intent was to build a collaboration/research network.
>
>
> Since the Arpanet had a number of different people in lead  
> positions, it would not be surprising if they had different agendas  
> or chose to present things in different ways to particular audiences.
>
> When I got involved, in the early 70's, I would periodically hear  
> some folk characterize the capability as designed to "survive  
> hostile battlefield conditions".  I took that to mean convention  
> battlefields, not nuclear.  (I've heard that in the first Iraq War,  
> the US was unable to knock out the Iraque data net; survivability  
> was thereby demonstrated...)
>
> I believe I've heard that Lukasik would claim the goal of nuclear  
> survivability during presentations to the military or Congress or  
> somesuch, as a funding pitch.
>
> d/
>
> -- 
>
>  Dave Crocker
>  Brandenburg InternetWorking
>  bbiw.net



From vint at google.com  Wed Sep  2 09:22:52 2009
From: vint at google.com (Vint Cerf)
Date: Wed, 2 Sep 2009 12:22:52 -0400
Subject: [ih] ARPAnet and nuclear survivability
In-Reply-To: <alpine.WNT.1.10.0909020914000.5464@teg-HP>
References: <4A9D44D2.20130.1F28B896@bernie.fantasyfarm.com>
	<4A9E8F11.5060307@dcrocker.net>
	<97CC4C99-08CE-4DFE-918D-4858570EBA89@google.com>
	<alpine.WNT.1.10.0909020914000.5464@teg-HP>
Message-ID: <1BD3CD72-FD69-4DA2-BB07-C1738F67B8E6@google.com>

i pushed the concept during my time leading the Internet work and this  
capability was demonstrated by Mike Frankel at SRI International via  
the Strategic Air Command around 1982.

vint


On Sep 2, 2009, at 12:17 PM, Terry Gray wrote:

> I recall asking Bob Braden this same question 25 years ago, and if
> memory serves, his take back then was that the survivability meme was
> not part of the original concept (agreeing with Larry's view), but
> that it started showing up in proposals by the mid-70s, presumably to
> increase their marketability within DOD.
>
> Bob, does that sound right?
>
> -teg
>
> On Wed, 2 Sep 2009, Vint Cerf wrote:
>
>> we'd have to ask steve lukasik. Larry Roberts was adamant that his  
>> motivation
>> was resource sharing and that comports well with Licklider's  
>> writings.
>>
>> vint
>>
>> On Sep 2, 2009, at 11:28 AM, Dave CROCKER wrote:
>>
>>>
>>>
>>> Bernie Cosell wrote:
>>>> On another list the old question came up again and I'm not sure  
>>>> I've ever
>>>> gotten a straight answer.  I have always thought that Licklider's  
>>>> intent
>>>> was to build a collaboration/research network.
>>>
>>>
>>> Since the Arpanet had a number of different people in lead  
>>> positions, it
>>> would not be surprising if they had different agendas or chose to  
>>> present
>>> things in different ways to particular audiences.
>>>
>>> When I got involved, in the early 70's, I would periodically hear  
>>> some folk
>>> characterize the capability as designed to "survive hostile  
>>> battlefield
>>> conditions".  I took that to mean convention battlefields, not  
>>> nuclear.
>>> (I've heard that in the first Iraq War, the US was unable to knock  
>>> out the
>>> Iraque data net; survivability was thereby demonstrated...)
>>>
>>> I believe I've heard that Lukasik would claim the goal of nuclear
>>> survivability during presentations to the military or Congress or  
>>> somesuch,
>>> as a funding pitch.
>>>
>>> d/
>>>
>>> -- 
>>>
>>> Dave Crocker
>>> Brandenburg InternetWorking
>>> bbiw.net
>>



From randy at psg.com  Wed Sep  2 21:50:14 2009
From: randy at psg.com (Randy Bush)
Date: Wed, 02 Sep 2009 21:50:14 -0700
Subject: [ih] codrescu on internet history
Message-ID: <m2ocpspr7d.wl%randy@psg.com>

http://www.npr.org/templates/story/story.php?storyId=112488309


From vint at google.com  Thu Sep  3 05:42:46 2009
From: vint at google.com (Vint Cerf)
Date: Thu, 3 Sep 2009 08:42:46 -0400
Subject: [ih] codrescu on internet history
In-Reply-To: <m2ocpspr7d.wl%randy@psg.com>
References: <m2ocpspr7d.wl%randy@psg.com>
Message-ID: <1065D6D4-6359-4534-9877-5A053F6E547E@google.com>

i liked the third blog comment...

v


On Sep 3, 2009, at 12:50 AM, Randy Bush wrote:

> http://www.npr.org/templates/story/story.php?storyId=112488309



From spencer at mcsr-labs.org  Thu Sep  3 05:51:00 2009
From: spencer at mcsr-labs.org (Spencer Dawkins)
Date: Thu, 3 Sep 2009 07:51:00 -0500
Subject: [ih] codrescu on internet history
References: <m2ocpspr7d.wl%randy@psg.com>
Message-ID: <91DEAF26DDFE4C7E9A174AE330F1CE39@china.huawei.com>

randy,


> http://www.npr.org/templates/story/story.php?storyId=112488309

thank you for sharing - very thought-provoking :-)

Spencer


From randy at psg.com  Thu Sep  3 05:51:10 2009
From: randy at psg.com (Randy Bush)
Date: Thu, 03 Sep 2009 05:51:10 -0700
Subject: [ih] codrescu on internet history
In-Reply-To: <1065D6D4-6359-4534-9877-5A053F6E547E@google.com>
References: <m2ocpspr7d.wl%randy@psg.com>
	<1065D6D4-6359-4534-9877-5A053F6E547E@google.com>
Message-ID: <m2eiqop4xt.wl%randy@psg.com>

> i liked the third blog comment...
>> http://www.npr.org/templates/story/story.php?storyId=112488309

i liked the piece because one could feel the context in the societal
time-line not just in geek time.

randy


From adrian at creative.net.au  Thu Sep  3 08:08:21 2009
From: adrian at creative.net.au (Adrian Chadd)
Date: Thu, 3 Sep 2009 23:08:21 +0800
Subject: [ih] codrescu on internet history
In-Reply-To: <m2eiqop4xt.wl%randy@psg.com>
References: <m2ocpspr7d.wl%randy@psg.com>
	<1065D6D4-6359-4534-9877-5A053F6E547E@google.com>
	<m2eiqop4xt.wl%randy@psg.com>
Message-ID: <20090903150821.GI23682@skywalker.creative.net.au>

On Thu, Sep 03, 2009, Randy Bush wrote:

> i liked the piece because one could feel the context in the societal
> time-line not just in geek time.

I liked that I could link the development of the internet pre being
personally aware of it to things I -was- aware of as a child.

:)


Adrian



From mbaer at cs.tu-berlin.de  Wed Sep  9 04:46:52 2009
From: mbaer at cs.tu-berlin.de (=?ISO-8859-1?Q?Matthias_B=E4rwolff?=)
Date: Wed, 09 Sep 2009 13:46:52 +0200
Subject: [ih] ARPAnet and nuclear survivability
In-Reply-To: <20090901220329.CB2936BE615@mercury.lcs.mit.edu>
References: <20090901220329.CB2936BE615@mercury.lcs.mit.edu>
Message-ID: <4AA795AC.3000001@cs.tu-berlin.de>



Noel Chiappa wrote:
>     > From: "Bernie Cosell" <bernie at fantasyfarm.com>
>
>     > OTOH the IMP *was* originally delivered in a fully mil-spec ruggedized
>     > case...
>
> I think that was in mostly just for reliability/robustness; I think one
> ARPANet history (maybe 'Where Wizards Stay Up Late') discusses the reasoning.
>   
Indeed, Hafner and Lyon discuss this at page 97:

"Heart's choices were limited. ... From the start, the reliability issue
led Heart to favor the new Honeywell DDP-516, .... The 516 also helped
to settle Heart's fear that inquisitive graduate students might bring
down the network with their tinkering. He could rest much easier knowing
the IMPs would be housed in a box built to withstand war."

That account in turn is based on various personal interviews with Heart
and the IMP guys, plus previous CBI interviews with the same (cf. page 268).

The "BBN versus graduate students at host sites" issue comes up once or
twice more in the Hafner and Lyon book. And given Heart's well-known
initial adamancy in reliability it sounds plausible to me.

Matthias
> If that old story about the milkshake is true, it might have been a wise
> choice, though... :-)
>
> 	Noel
>   

-- 
Matthias B?rwolff
www.b?rwolff.de



From jack at 3kitty.org  Wed Sep  9 08:03:44 2009
From: jack at 3kitty.org (Jack Haverty)
Date: Wed, 09 Sep 2009 08:03:44 -0700
Subject: [ih] ARPAnet and nuclear survivability
In-Reply-To: <4AA795AC.3000001@cs.tu-berlin.de>
References: <20090901220329.CB2936BE615@mercury.lcs.mit.edu>
	<4AA795AC.3000001@cs.tu-berlin.de>
Message-ID: <1252508624.3321.14.camel@localhost>

On Wed, 2009-09-09 at 13:46 +0200, Matthias B?rwolff wrote:
> The "BBN versus graduate students at host sites" issue 

This was a real issue.  I remember clearly the consternation on the DEC
field service guy's face when he came to MIT to service the AI Lab
PDP-10 and discovered that the instruction set had been changed.  It was
helpful to have a "rotate" instruction that went "the other way" for use
by the Chess program, so someone (Tom Knight?  Jack Holloway?) simply
added the hardware to make an unused opcode do something useful.

BBN was effectively an MIT spinoff, just up the road from MIT, and
heavily populated by ex-graduate students.  The IMPs were being
delivered to MIT and other similar universities.

I'm surprised the cabinet wasn't welded shut.

Later on, when the IMPs were out in military sites, we had one machine
that was consistently unreliable - rebooting every few days for no
apparent reason.  After field service had replaced everything except the
cabinet, a field service guy was assigned to sit in the computer room
and wait for the IMP to reboot to see if he could learn anything.

After a while, a door slammed somewhere, and a older colonel stormed
into the computer room, cursing and complaining, went to the IMP, opened
the cabinet door, opened the top of the computer itself, reached inside,
and pushed the internal reset switch on the main board.

Turns out that he knew that when comms didn't work, he knew from
experience that all you had to do was recycle the box, so when his email
didn't work...he'd reboot the IMP.

Must have been a graduate student before Arpanet time.

I think the cabinets were all locked after that.  Frank's concern was
well-founded.

/Jack Haverty
Point Arena, CA
ex-MIT (1966-1978)
ex-BBN (1978-1990)



From dhc2 at dcrocker.net  Wed Sep  9 08:38:09 2009
From: dhc2 at dcrocker.net (Dave CROCKER)
Date: Wed, 09 Sep 2009 08:38:09 -0700
Subject: [ih] ARPAnet and nuclear survivability
In-Reply-To: <1252508624.3321.14.camel@localhost>
References: <20090901220329.CB2936BE615@mercury.lcs.mit.edu>	<4AA795AC.3000001@cs.tu-berlin.de>
	<1252508624.3321.14.camel@localhost>
Message-ID: <4AA7CBE1.8060200@dcrocker.net>



Jack Haverty wrote:
> Turns out that he knew that when comms didn't work, he knew from
> experience that all you had to do was recycle the box, so when his email
> didn't work...he'd reboot the IMP.

Perhaps he was right.

Any chance we can fix spam by simultaneously rebooting all the routers on the net?

(Don't answer.  I realize we would also need to reboot all the hosts, but didn't 
want to suggest something unreasonable.)

d/
-- 

   Dave Crocker
   Brandenburg InternetWorking
   bbiw.net


From jnc at mercury.lcs.mit.edu  Wed Sep  9 08:41:10 2009
From: jnc at mercury.lcs.mit.edu (Noel Chiappa)
Date: Wed,  9 Sep 2009 11:41:10 -0400 (EDT)
Subject: [ih] ARPAnet and nuclear survivability
Message-ID: <20090909154110.141926BE619@mercury.lcs.mit.edu>

    > From: Dave CROCKER <dhc2 at dcrocker.net>

    > I realize we would also need to reboot all the hosts

Don't you mean 'replace the OSs on them all with something that's a little
more resistant to viruses, so we can get rid of the bot-nets that are
the source of a lot of spam'? :-) (Ducks!)

	Noel


From dhc2 at dcrocker.net  Wed Sep  9 09:16:08 2009
From: dhc2 at dcrocker.net (Dave CROCKER)
Date: Wed, 09 Sep 2009 09:16:08 -0700
Subject: [ih] ARPAnet and nuclear survivability
In-Reply-To: <20090909154110.141926BE619@mercury.lcs.mit.edu>
References: <20090909154110.141926BE619@mercury.lcs.mit.edu>
Message-ID: <4AA7D4C8.8050107@dcrocker.net>



Noel Chiappa wrote:
>     > I realize we would also need to reboot all the hosts
> 
> Don't you mean 'replace the OSs on them all with something that's a little
> more resistant to viruses, 


No, Noel, I don't.

That might smack of trying to be serious...

d/

-- 

   Dave Crocker
   Brandenburg InternetWorking
   bbiw.net


From craig at aland.bbn.com  Wed Sep  9 09:50:48 2009
From: craig at aland.bbn.com (Craig Partridge)
Date: Wed, 09 Sep 2009 12:50:48 -0400
Subject: [ih] ARPAnet and nuclear survivability
Message-ID: <20090909165048.1990E28E137@aland.bbn.com>


In the fun memories of the past department -- recall Van used to do this.
He had a tool that knew what mangled datagram would cause each OS and
router code build to crash.  As I recall, he used to use it in a Bay
Area research network to force the network to function.  Made sense,
he often did his work late at night and this was before 24-hour operations
support became standard.  (Which leads to a story of an early NSFNET
regional network that, if memory serves, used to take their network down
to perform maintenance once a week between 9 and 11am Eastern....)

Craig

> Jack Haverty wrote:
> > Turns out that he knew that when comms didn't work, he knew from
> > experience that all you had to do was recycle the box, so when his email
> > didn't work...he'd reboot the IMP.
> 
> Perhaps he was right.
> 
> Any chance we can fix spam by simultaneously rebooting all the routers on the
>  net?
> 
> (Don't answer.  I realize we would also need to reboot all the hosts, but did
> n't 
> want to suggest something unreasonable.)
> 
> d/
> -- 
> 
>    Dave Crocker
>    Brandenburg InternetWorking
>    bbiw.net
********************
Craig Partridge
Chief Scientist, BBN Technologies
E-mail: craig at aland.bbn.com or craig at bbn.com
Phone: +1 517 324 3425


From sherry.griddine at inetinteractive.com  Wed Sep  9 14:56:15 2009
From: sherry.griddine at inetinteractive.com (Sherry Griddine)
Date: Wed, 9 Sep 2009 16:56:15 -0500
Subject: [ih] Unsubscribe (2nd notice)
In-Reply-To: <20090909165048.1990E28E137@aland.bbn.com>
References: <20090909165048.1990E28E137@aland.bbn.com>
Message-ID: <AADC425642A548E1BCA3F977BFABA91B@inetintesherry>

Please unsubscribe - thank you.

Sherry Griddine 


-----Original Message-----
From: internet-history-bounces at postel.org
[mailto:internet-history-bounces at postel.org] On Behalf Of Craig Partridge
Sent: Wednesday, September 09, 2009 11:51 AM
To: dcrocker at bbiw.net; Dave CROCKER
Cc: internet-history at postel.org
Subject: Re: [ih] ARPAnet and nuclear survivability


In the fun memories of the past department -- recall Van used to do this.
He had a tool that knew what mangled datagram would cause each OS and
router code build to crash.  As I recall, he used to use it in a Bay
Area research network to force the network to function.  Made sense,
he often did his work late at night and this was before 24-hour operations
support became standard.  (Which leads to a story of an early NSFNET
regional network that, if memory serves, used to take their network down
to perform maintenance once a week between 9 and 11am Eastern....)

Craig

> Jack Haverty wrote:
> > Turns out that he knew that when comms didn't work, he knew from
> > experience that all you had to do was recycle the box, so when his email
> > didn't work...he'd reboot the IMP.
> 
> Perhaps he was right.
> 
> Any chance we can fix spam by simultaneously rebooting all the routers on
the
>  net?
> 
> (Don't answer.  I realize we would also need to reboot all the hosts, but
did
> n't 
> want to suggest something unreasonable.)
> 
> d/
> -- 
> 
>    Dave Crocker
>    Brandenburg InternetWorking
>    bbiw.net
********************
Craig Partridge
Chief Scientist, BBN Technologies
E-mail: craig at aland.bbn.com or craig at bbn.com
Phone: +1 517 324 3425




From mbaer at cs.tu-berlin.de  Thu Sep 10 06:47:12 2009
From: mbaer at cs.tu-berlin.de (=?ISO-8859-1?Q?Matthias_B=E4rwolff?=)
Date: Thu, 10 Sep 2009 15:47:12 +0200
Subject: [ih] That Host-Host Protocol Document No. 1
In-Reply-To: <F7A74DC2-1C73-435B-AEF5-0A6B1444ADA8@google.com>
References: <4A7A033C.9040702@cs.tu-berlin.de> <m2y6pxryp8.wl%randy@psg.com>
	<4A7BF5AB.6070800@cs.tu-berlin.de>
	<F7A74DC2-1C73-435B-AEF5-0A6B1444ADA8@google.com>
Message-ID: <4AA90360.9090606@cs.tu-berlin.de>

This is to wrap up the thread for the public record with the results of
an extensive off-list discussion with various people, resulting in the
nice people from the Computer History Museum
(http://www.*computerhistory*.org) digging out the Host-Host Protocol
Document No. 1 (NIC 5143
<http://www.b%C3%A4rwolff.de/public/Crocker-1970-Host-Host-Protocol-Document-No-1-NIC5143.pdf>)
from their garages and sending me a scan. For the time being I have put
it up at
<http://www.baerwolff.de/public/Crocker-1970-Host-Host-Protocol-Document-No-1-NIC5143.pdf>.

It turned out that Crocker's "Host-Host Protocol Document #1" (dated
August 3, 1970; also referred to as NIC 5143) was an effort along the
lines laid in RFC 53 (dated June 9, 1970) which tried to establish a
protocol series that would have a slightly more formal feel to it than
the then still very young RFC series. (In fact, NIC published Protocol
Handbooks from 1970 to 1990, all of which are available at the Computer
Museum.) It was, so to speak, the first "official" version of the
protocol that the NCPs at the host sites were to implement. Of course,
the protocol specification changed quite a lot over time, and many early
RFCs commented on "document #1". By early 1971 at the latest, with RFC
107, <http://www.rfc-editor.org/rfc/rfc107.txt> "document #1" was
officially obsolete. A July 1971 version of the host-host protocol
("Host-Host Protocol for the ARPA Network", NIC 7141) can be found at
<http://ia300210.us.archive.org/1/items/HostHostProtocolForTheArpaNetwork/>.

Matthias



Vint Cerf wrote:
> RFC 1 was written to establish the series. We thrashed around with a
> number of ideas in the late 1969 and early 1970 period. NIL, DEL, and
> eventually the H-H protocol. Steve Crocker led that effort and is
> surely the expert on events of the day.
>
> v
>
> On Aug 7, 2009, at 5:36 AM, Matthias B?rwolff wrote:
>
>> Thanks for the hint. But actually I was under strong impression that RFC
>> 1 was just that: a request for comments with more questions and problem
>> statements than bold proposals; and the first go at really specifying a
>> generic host-host protocol happened some time in 1970 rather than early
>> 1969 (when the BBN 1822 report was not even out yet). I took it that the
>> references and discussion in RFC 65 are about this document (which is
>> therein referred to as Host-Host Protocol Document No. 1; and this term
>> is referred to in various other accounts at the time, too, so I was just
>> curious as to what exactly that document was; I frankly doubt it's RFC
>> 1). (I may go and ask Dave Walden or Steve Crocker directly on this.)
>>
>> Randy Bush wrote:
>>>     http://www.rfc-editor.org/rfc/rfc1.txt
>>>
>>
>> -- 
>> Matthias B?rwolff
>> www.b?rwolff.de
>>
>

-- 
Matthias B?rwolff
www.b?rwolff.de



From vint at google.com  Thu Sep 10 06:53:24 2009
From: vint at google.com (Vint Cerf)
Date: Thu, 10 Sep 2009 09:53:24 -0400
Subject: [ih] That Host-Host Protocol Document No. 1
In-Reply-To: <4AA90360.9090606@cs.tu-berlin.de>
References: <4A7A033C.9040702@cs.tu-berlin.de> <m2y6pxryp8.wl%randy@psg.com>
	<4A7BF5AB.6070800@cs.tu-berlin.de>
	<F7A74DC2-1C73-435B-AEF5-0A6B1444ADA8@google.com>
	<4AA90360.9090606@cs.tu-berlin.de>
Message-ID: <C9C1B60F-B3E4-4F29-9D4D-260743AB77EA@google.com>

this is a nice piece of historical research, rescue and recovery!

v

On Sep 10, 2009, at 9:47 AM, Matthias B?rwolff wrote:

> This is to wrap up the thread for the public record with the results  
> of
> an extensive off-list discussion with various people, resulting in the
> nice people from the Computer History Museum
> (http://www.*computerhistory*.org) digging out the Host-Host Protocol
> Document No. 1 (NIC 5143
> <http://www.b%C3%A4rwolff.de/public/Crocker-1970-Host-Host-Protocol-Document-No-1-NIC5143.pdf 
> >)
> from their garages and sending me a scan. For the time being I have  
> put
> it up at
> <http://www.baerwolff.de/public/Crocker-1970-Host-Host-Protocol-Document-No-1-NIC5143.pdf 
> >.
>
> It turned out that Crocker's "Host-Host Protocol Document #1" (dated
> August 3, 1970; also referred to as NIC 5143) was an effort along the
> lines laid in RFC 53 (dated June 9, 1970) which tried to establish a
> protocol series that would have a slightly more formal feel to it than
> the then still very young RFC series. (In fact, NIC published Protocol
> Handbooks from 1970 to 1990, all of which are available at the  
> Computer
> Museum.) It was, so to speak, the first "official" version of the
> protocol that the NCPs at the host sites were to implement. Of course,
> the protocol specification changed quite a lot over time, and many  
> early
> RFCs commented on "document #1". By early 1971 at the latest, with RFC
> 107, <http://www.rfc-editor.org/rfc/rfc107.txt> "document #1" was
> officially obsolete. A July 1971 version of the host-host protocol
> ("Host-Host Protocol for the ARPA Network", NIC 7141) can be found at
> <http://ia300210.us.archive.org/1/items/HostHostProtocolForTheArpaNetwork/ 
> >.
>
> Matthias
>
>
>
> Vint Cerf wrote:
>> RFC 1 was written to establish the series. We thrashed around with a
>> number of ideas in the late 1969 and early 1970 period. NIL, DEL, and
>> eventually the H-H protocol. Steve Crocker led that effort and is
>> surely the expert on events of the day.
>>
>> v
>>
>> On Aug 7, 2009, at 5:36 AM, Matthias B?rwolff wrote:
>>
>>> Thanks for the hint. But actually I was under strong impression  
>>> that RFC
>>> 1 was just that: a request for comments with more questions and  
>>> problem
>>> statements than bold proposals; and the first go at really  
>>> specifying a
>>> generic host-host protocol happened some time in 1970 rather than  
>>> early
>>> 1969 (when the BBN 1822 report was not even out yet). I took it  
>>> that the
>>> references and discussion in RFC 65 are about this document (which  
>>> is
>>> therein referred to as Host-Host Protocol Document No. 1; and this  
>>> term
>>> is referred to in various other accounts at the time, too, so I  
>>> was just
>>> curious as to what exactly that document was; I frankly doubt it's  
>>> RFC
>>> 1). (I may go and ask Dave Walden or Steve Crocker directly on  
>>> this.)
>>>
>>> Randy Bush wrote:
>>>>    http://www.rfc-editor.org/rfc/rfc1.txt
>>>>
>>>
>>> -- 
>>> Matthias B?rwolff
>>> www.b?rwolff.de
>>>
>>
>
> -- 
> Matthias B?rwolff
> www.b?rwolff.de
>




From bortzmeyer at nic.fr  Thu Sep 10 12:33:41 2009
From: bortzmeyer at nic.fr (Stephane Bortzmeyer)
Date: Thu, 10 Sep 2009 21:33:41 +0200
Subject: [ih] What is the origin of the private network address 192.168.*.*?
Message-ID: <20090910193341.GA28251@nic.fr>

The excellent Q&A site Server Fault has an interesting question, which
may interest some persons here. Answer directly on Server Fault (you
need to log in with an OpenID identifier) or post here and I'll write
it back to Server Fault.

http://serverfault.com/questions/64013/what-is-the-origin-of-the-private-network-address-192-168

I had a smart (but non-technical) user ask me today:

Why did they pick 192.168.. for a private network address?

The only answer I could come up with is because 192 = 11000000 in binary. And 168 is 10101000 in binary. Both of which are kind of cool looking.

Is there a real historical reason for that particular choice of numbers? Why not 127.127..? Or 128.128..?

Similar question for 10.0.0.0 and 172.16.0.0


From craig at aland.bbn.com  Thu Sep 10 12:45:08 2009
From: craig at aland.bbn.com (Craig Partridge)
Date: Thu, 10 Sep 2009 15:45:08 -0400
Subject: [ih] What is the origin of the private network address
	192.168.*.*?
Message-ID: <20090910194508.9F1BD28E137@aland.bbn.com>

> The excellent Q&A site Server Fault has an interesting question, which
> may interest some persons here. Answer directly on Server Fault (you
> need to log in with an OpenID identifier) or post here and I'll write
> it back to Server Fault.
> 
> http://serverfault.com/questions/64013/what-is-the-origin-of-the-private-netw
> ork-address-192-168
> 
> I had a smart (but non-technical) user ask me today:
> 
> Why did they pick 192.168.. for a private network address?
> The only answer I could come up with is because 192 = 11000000 in binary. And
>  168 is 10101000 in binary. Both of which are kind of cool looking.
> 
> Is there a real historical reason for that particular choice of numbers? Why 
> not 127.127..? Or 128.128..?


Dunno, though the fact it is what used to be a class C address is a hint
but I think the convention was established after we went to CIDR.

> Similar question for 10.0.0.0 and 172.16.0.0

10.0.0.0 is easy.  For folks who needed LARGE private networks the only
large space available by the early 1990s was the old ARPANET network
number (the ARPANET was net 10 and was decommissioned around 1991).

********************
Craig Partridge
Chief Scientist, BBN Technologies
E-mail: craig at aland.bbn.com or craig at bbn.com
Phone: +1 517 324 3425


From vint at google.com  Thu Sep 10 13:08:19 2009
From: vint at google.com (Vint Cerf)
Date: Thu, 10 Sep 2009 16:08:19 -0400
Subject: [ih] What is the origin of the private network address
	192.168.*.*?
In-Reply-To: <20090910194508.9F1BD28E137@aland.bbn.com>
References: <20090910194508.9F1BD28E137@aland.bbn.com>
Message-ID: <4143E51A-2929-459B-BC7B-C6DE2C8C53FF@google.com>

arpanet was decommissioned july 1990 i believe.

v

On Sep 10, 2009, at 3:45 PM, Craig Partridge wrote:

>> The excellent Q&A site Server Fault has an interesting question,  
>> which
>> may interest some persons here. Answer directly on Server Fault (you
>> need to log in with an OpenID identifier) or post here and I'll write
>> it back to Server Fault.
>>
>> http://serverfault.com/questions/64013/what-is-the-origin-of-the-private-netw
>> ork-address-192-168
>>
>> I had a smart (but non-technical) user ask me today:
>>
>> Why did they pick 192.168.. for a private network address?
>> The only answer I could come up with is because 192 = 11000000 in  
>> binary. And
>> 168 is 10101000 in binary. Both of which are kind of cool looking.
>>
>> Is there a real historical reason for that particular choice of  
>> numbers? Why
>> not 127.127..? Or 128.128..?
>
>
> Dunno, though the fact it is what used to be a class C address is a  
> hint
> but I think the convention was established after we went to CIDR.
>
>> Similar question for 10.0.0.0 and 172.16.0.0
>
> 10.0.0.0 is easy.  For folks who needed LARGE private networks the  
> only
> large space available by the early 1990s was the old ARPANET network
> number (the ARPANET was net 10 and was decommissioned around 1991).
>
> ********************
> Craig Partridge
> Chief Scientist, BBN Technologies
> E-mail: craig at aland.bbn.com or craig at bbn.com
> Phone: +1 517 324 3425



From louie at transsys.com  Thu Sep 10 14:28:45 2009
From: louie at transsys.com (Louis Mamakos)
Date: Thu, 10 Sep 2009 17:28:45 -0400
Subject: [ih] What is the origin of the private network address
	192.168.*.*?
In-Reply-To: <20090910193341.GA28251@nic.fr>
References: <20090910193341.GA28251@nic.fr>
Message-ID: <7D73FCA0-F656-48D8-B27A-53574FF7A8B2@transsys.com>

I think that I recall at the time there was desire to have class-A,  
class-B and class-C private address space for private networks.  I'm  
not sure where the lined up against the transition to CIDR-style  
prefixes, but there were a considerable number of "legacy" systems at  
the time that had classfull IP addresses wired into their code.

As Craig pointed out, net 10 was easy, as it had just been recently  
decommissioned.  Perhaps the class-B and class-C prefixes just  
happened to be where the IANA was in the allocation process at the time?

Louis Mamakos

On Sep 10, 2009, at 3:33 PM, Stephane Bortzmeyer wrote:

> The excellent Q&A site Server Fault has an interesting question, which
> may interest some persons here. Answer directly on Server Fault (you
> need to log in with an OpenID identifier) or post here and I'll write
> it back to Server Fault.
>
> http://serverfault.com/questions/64013/what-is-the-origin-of-the-private-network-address-192-168
>
> I had a smart (but non-technical) user ask me today:
>
> Why did they pick 192.168.. for a private network address?
>
> The only answer I could come up with is because 192 = 11000000 in  
> binary. And 168 is 10101000 in binary. Both of which are kind of  
> cool looking.
>
> Is there a real historical reason for that particular choice of  
> numbers? Why not 127.127..? Or 128.128..?
>
> Similar question for 10.0.0.0 and 172.16.0.0
>



From tony.li at tony.li  Thu Sep 10 19:52:18 2009
From: tony.li at tony.li (Tony Li)
Date: Thu, 10 Sep 2009 19:52:18 -0700
Subject: [ih] What is the origin of the private network
	address	192.168.*.*?
In-Reply-To: <7D73FCA0-F656-48D8-B27A-53574FF7A8B2@transsys.com>
References: <20090910193341.GA28251@nic.fr>
	<7D73FCA0-F656-48D8-B27A-53574FF7A8B2@transsys.com>
Message-ID: <4AA9BB62.3030303@tony.li>


And using 127.0.0.0 or any of other other well known 'martians' would 
have required a number of legacy systems to change, as the martians 
table was well established by then.

Tony


Louis Mamakos wrote:
> I think that I recall at the time there was desire to have class-A, 
> class-B and class-C private address space for private networks.  I'm not 
> sure where the lined up against the transition to CIDR-style prefixes, 
> but there were a considerable number of "legacy" systems at the time 
> that had classfull IP addresses wired into their code.
> 
> As Craig pointed out, net 10 was easy, as it had just been recently 
> decommissioned.  Perhaps the class-B and class-C prefixes just happened 
> to be where the IANA was in the allocation process at the time?
> 
> Louis Mamakos
> 
> On Sep 10, 2009, at 3:33 PM, Stephane Bortzmeyer wrote:
> 
>> The excellent Q&A site Server Fault has an interesting question, which
>> may interest some persons here. Answer directly on Server Fault (you
>> need to log in with an OpenID identifier) or post here and I'll write
>> it back to Server Fault.
>>
>> http://serverfault.com/questions/64013/what-is-the-origin-of-the-private-network-address-192-168 
>>
>>
>> I had a smart (but non-technical) user ask me today:
>>
>> Why did they pick 192.168.. for a private network address?
>>
>> The only answer I could come up with is because 192 = 11000000 in 
>> binary. And 168 is 10101000 in binary. Both of which are kind of cool 
>> looking.
>>
>> Is there a real historical reason for that particular choice of 
>> numbers? Why not 127.127..? Or 128.128..?
>>
>> Similar question for 10.0.0.0 and 172.16.0.0
>>
> 
> 


From sbrim at cisco.com  Fri Sep 11 03:29:10 2009
From: sbrim at cisco.com (Scott Brim)
Date: Fri, 11 Sep 2009 06:29:10 -0400
Subject: [ih] What is the origin of the private network address
 192.168.*.*?
In-Reply-To: <4AA9BB62.3030303@tony.li>
References: <20090910193341.GA28251@nic.fr>
	<7D73FCA0-F656-48D8-B27A-53574FF7A8B2@transsys.com>
	<4AA9BB62.3030303@tony.li>
Message-ID: <20090911102910.GF17769@cisco.com>

Yes but why 192.168 specifically?  I remember when it happened but not
why.