[ih] Nit-picking an origin story

[John Day]

> On Aug 18, 2025, at 10:21, Steve Crocker <steve at shinkuro.com> wrote:
> 
> John,
> 
> See inline for comments.
> 
> Steve
> 
> 
> On Mon, Aug 18, 2025 at 10:00 AM John Day via Internet-history <internet-history at elists.isoc.org <mailto:internet-history at elists.isoc.org>> wrote:
>> Apologies for going back so far but just for the record, a couple of things.
>> 
>> I have been working with David Hutchison at Univ of Lancaster and he has been digging through old archives.  He found a1965 memo by Davies who had just returned from a US conference where he heard several papers on about timesharing systems, and had the idea that they should do to networking what time-slicing did in OSs. He called it packet switching. (What is fun, David also found a memo by Derek Barber relating Donald coming back from the conference and excited about the idea.)
>> 
>> What seem to me to be the two seminal events leading to the ARPANET were:
>> 1) The Ann Arbor PI meeting in 67, where Roberts got a lot of pushback for the idea of putting them on a network and Wes Clark’s suggestion to put a minicomputer in front of each host. Hence the IMPs.
> 
> Agreed.  And I think it's relevant to note that Moore's Law played a pivotal role here in the sense that the cost of minicomputers had come down to the point where paying for a minicomputer for each site was (just barely) feasible.  Reliability was also crucial.  We all remember that time-shared systems rarely stayed up longer than several hours before needing a reboot.

Totally agree.
>> 
>> 2) Later that same year, Roberts gave a talk at the Gatlinburg OS conference on the ARPA’s network plans. In one of those after the talk discussions, i.e., in the bar, there were two major discussions: 1)    Roger Scantlebury from NPL (along with others) convinced Roberts to use packet switching. (Roberts had never heard of it, but later found Baran’s report in a stack of documents in his office.) ;-) and 2)     In his paper at the conference, Roberts had said they were going to use 2.4Kbps lines for the network.  Again Scantlebury convinced Roberts that that was no where near fast enough and that they had found that at least 50Kbps was needed.
> 
> Well, Len Kleinrock and Larry Roberts were office mates and worked together at Lincoln Lab.  Len focused on message switching.  Larry was definitely aware of the idea, and it played a central role in his thinking.
> 
> Re the use of 50Kbps lines, Larry discovered the government could lease these lines at a special rate, which made it feasible within the budget he had.

Yes, you mentioned that before. When I first read that the first thing I thought of the difference in budget between NPL and the ARPANET!  ;-)  But in the Gatlinburg paper he still had 2.4.

(BTW, that must have been some conference. Gatlinburg is at the entrance to Smoky Mountain National Park and in 1965 (when  I was there) was a pretty cheesy tourist trap. What a place for a conference!)
> 
>> This last one I think doesn’t get enough credit. It is a very small thing, but I think was a major contribution to the success of the ARPANET. It would have worked at 2.4 or 9.6, but been so glacially slow as to have been considered not successful. At 50Kbps, we could do real work that was way beyond what people expected. Not to take anything away from the great software development that went into the IMPs and the NCPs, etc. I really think this gets too little credit for the success.
> 
> I agree.  John McCarthy, well known for his AI work but also a prime supporter of time-sharing, argued against the Arpanet, pushing instead for dial-up email forwarding service, i.e., the UUNET architecture.

I remember at the 4th Data Comm in Quebec, Robert Fano was the keynote speaker and said they built timesharing to make more efficient use of the hardware, but found it encouraged collaboration and that was the real benefit. He predicted we would find the same thing about networks and I thought then he was dead on.

>   UUNET was quite successful in its own right, so McCarthy wasn't wrong, but I agree that the 50Kbps lines in the Arpanet made a big difference.  I would add that the reliability of the Arpanet was also a major contributor to its success.  Frank Heart was adamant about it, and Ben Barker has a colorful story about the lights on the IMP panel being a major source of outages.  The IMPs had a 98% percent uptime at first.  98% was astonishingly good compared to other machines of the day, but intolerably poor in terms of providing an always available service.  Ben re-engineered the lights and brought the reliability up to 99.98%.  How's that for a small thing having a big effect!

Not bad.
> 
> 
>  
>> 
>> 
>> (Also I have to admit, I kinda like the idea when small things have a major effect.)
>> 
>> > On Aug 16, 2025, at 13:16, John Day via Internet-history <internet-history at elists.isoc.org <mailto:internet-history at elists.isoc.org>> wrote:
>> > 
>> > The NPL network already existed and had for awhile, a couple of years but I will have to go look at sources to be exact.
>> > 
>> > Of course, what this should say is the first messages exchanged on the ARPANET.
>> > 
>> > I am sure BBN tested it before they delivered it, but I don’t remember now what Hafner says about that.
>> > 
>> > Take care,
>> > John
>> > 
>> >> On Aug 16, 2025, at 12:41, Dave Crocker via Internet-history <internet-history at elists.isoc.org <mailto:internet-history at elists.isoc.org>> wrote:
>> >> 
>> >> My Facebook feed just delivered a tidbit from UCLA that begins:
>> >> 
>> >>  "In 1969, UCLA Professor Leonard Kleinrock directed the transmission
>> >>  of the first message between two networked computers..."
>> >> 
>> >> I found myself wondering a bit about that characterization:
>> >> 
>> >> 1. Didn't BBN do some inter-host packet exchanges, when testing the
>> >>  IMPs, before shipping them to UCLA and SRI?  Wouldn't that have
>> >>  counted as the actual first?
>> >> 2. There were other packet research projects, at the time, but I don't
>> >>  remember the details of timing of other 'WAN' and 'LAN' project. 
>> >>  1969 was early enough that it's entirely possible the others were
>> >>  later, but I'd be interested in hearing the details.
>> >> 
>> >> I suspect the refinement of the UCLA statement would be:
>> >> 
>> >>  "In 1969, UCLA Professor Leonard Kleinrock directed the transmission
>> >>  of the first message between two networked computers
>> >> 
>> >> -- 
>> >> Dave Crocker
>> >> 
>> >> Brandenburg InternetWorking
>> >> bbiw.net <http://bbiw.net/>
>> >> bluesky: @dcrocker.bsky.social
>> >> mast: @dcrocker at mastodon.social
>> >> -- 
>> >> Internet-history mailing list
>> >> Internet-history at elists.isoc.org <mailto:Internet-history at elists.isoc.org>
>> >> https://elists.isoc.org/mailman/listinfo/internet-history
>> >> -
>> >> Unsubscribe: https://app.smartsheet.com/b/form/9b6ef0621638436ab0a9b23cb0668b0b?The%20list%20to%20be%20unsubscribed%20from=Internet-history
>> > 
>> > -- 
>> > Internet-history mailing list
>> > Internet-history at elists.isoc.org <mailto:Internet-history at elists.isoc.org>
>> > https://elists.isoc.org/mailman/listinfo/internet-history
>> > -
>> > Unsubscribe: https://app.smartsheet.com/b/form/9b6ef0621638436ab0a9b23cb0668b0b?The%20list%20to%20be%20unsubscribed%20from=Internet-history
>> 
>> -- 
>> Internet-history mailing list
>> Internet-history at elists.isoc.org <mailto:Internet-history at elists.isoc.org>
>> https://elists.isoc.org/mailman/listinfo/internet-history
>> -
>> Unsubscribe: https://app.smartsheet.com/b/form/9b6ef0621638436ab0a9b23cb0668b0b?The%20list%20to%20be%20unsubscribed%20from=Internet-history
> 
> 
> 
> --
>