[ih] Historical fiction

[Dave Walden]

BBN IMP was the fifth IMP on the network, in early 1970.  Before that 
we did our IMP operations work and support of host work by phone and 
by flying to the first four sites.  One or more people from BBN went 
with each of the first four IMP deliveries to help the host people 
get connected.  We made other trips to the first four host sites.  I 
did at least one new software release by flying in turn to the U of 
Utah, SRI, UCSB, and UCLA dropping off paper tapes as I went.  I have 
the IMP program listing on the telephone stand in my apartment so I 
could take phone calls from the host sites outside BBN east coast 
work hours (day or night).  (I already mention Bob Kahn and I going 
to UCLA once the first four nodes were connected to do tests which 
revealed a problem (Bob anticipated) with packet reassembly lockup.) 
Once we had the first cross-country line and the IMP at BBN, we could 
do new releases over the net, monitor the IMPs remotely, etc.

I would say that the main "home bases" were ARPA which was calling 
the top level shots, the IMP development team and Network Monitoring 
Center at BBN, the Network Information Center at SRI, the Network 
Measurement Center at UCLA, the highly distributed Network Working 
Group and interface development people at all the host sites, and the 
Air Force contracts office which acquired the wideband communications 
circuits from ATT.  I believe Network Analysis Corporation did 
repeated topological designs for ARPA leading up to acquisition on 
additional wideband circuits.

The first four IMPs were delivered and a host connected to each in 
approximately the last four months of 1969.  Then an IMP was 
delivered approximately every month for quite a while (with BBN which 
already had IMPs on site being the fifth site connected).

People were working on the net, writing RFCs to each other, creating 
or revising various protocol specs, and when appropriate updating IMP 
and/or host software.  Look the first few hundred RFCs for notices 
anticipating or reporting on various network activities.

Dave


At 07:15 PM 5/10/2012, Sytel wrote:
>I see. This is interesting, and definitely new; from the accounts 
>I'd read (admittedly, mostly from Kleinrock's perspective) I was 
>under the impression that the NMC was the "headquarters" for the 
>early net; indeed, I might have read somewhere that it wasn't even 
>connected to BBN's offices until some time after... I'm aware that 
>most of the original team went their separate ways after the ECCC in 
>1972, but up until then I'd been sort of picturing Kleinrock's team 
>as "running" the network from UCLA, with a lot of contact with BBN, of course.
>I'm sorry if there's some false assumptions in there... as 
>mentioned, this is something that's often hard to find out about in 
>the more widely available histories.
>What would have been the main "home bases" in the 1969-1972 period, 
>then, and what would be happening at each one? Were they actually 
>connected to the network, or were they more working on things that 
>would be used at nodes that were connected?
>
>----- Original Message ----- From: "Bernie Cosell" <bernie at fantasyfarm.com>
>To: <internet-history at postel.org>
>Sent: Thursday, May 10, 2012 3:21 PM
>Subject: Re: [ih] Historical fiction
>
>
>>On 10 May 2012 at 11:43, Sytel wrote:
>>
>>>I guess what I'm interested in is "zooming in" a little more, if you
>>>will. November, 1969, a typical afternoon in room 3420
>>>(http://www.flickr.com/photos/3420boelterhall/5609051340/) -- what might
>>>people be working on, reading, doing with the computers? What's a test
>>>that might be running on the network, and how is it still not quite
>>>working right? If something strange happens, who asks who about it? What
>>>exciting plans are in the pipeline, what deadlines are looming?
>>
>>I think that given the reality of all of the folk around the country
>>working on the ARPAnet, between the actual network development and the
>>host system software and the application development, I don't think that
>>Boelter Hall was all that central to what was going on either in the
>>large or in the small, so your work will be really quite "fictional" if
>>you're focusing on UCLA...
>>
>>  /Bernie\
>>
>>--
>>Bernie Cosell                     Fantasy Farm Fibers
>>mailto:bernie at fantasyfarm.com     Pearisburg, VA
>>    -->  Too many people, too few sheep  <--
>>
>>


--
home address: 12 Linden Rd., E. Sandwich, MA 02537
home ph=508-888-7655; cell ph = 503-757-3137
email address:  dave at walden-family.com; website: 
<http://www.walden-family.com/bbn/>http://www.walden-family.com/bbn/