[ih] Danny C.

[John Shoch]

Regarding Danny Cohen, unreliable voice, and the TCP split:
I wish we all had perfect memories of events that took place over 42 years
ago......
We can, though, try to reconstruct some of the history and the meetings we
attended:

--I'm sure Danny argued at many meetings for real-time voice, datagrams,
unreliable/low-latency transmission, etc.
--As I recall, through the end of 1977 TCP was one layer.
--There were also two sets of meetings going on:  an Internet group and a
TCP group.
--There was a meeting of the TCP group on Jan. 30-31, 1978, in S. Cal.,
where there was a lively discussion about layering, voice, fragmentation,
etc.  It appears that Postel drafted notes, dated Feb. 8, 1978 (although it
got numbered as IEN 67, which must have been later).
   https://www.rfc-editor.org/ien/ien67.pdf
>From that memo:
"Introduction and Objectives -- Cerf
The main objective is to get TCP-3 straightened out...."
"Arrangements -- Cohen.  ...  Uses this time to complain about TCP-3
becomming [sic] all things to all people.  *Also illustration of the
approach to voice service via an "unreliability" package on top of TCP*."
[emphasis added]
[As many of you know, Danny had a really insightful, satirical wit.]
--The following day there was a meeting of the Internet group, on Feb. 1,
1978.
Postel also wrote up notes from this meeting, dated 3 February 1978 (IEN
22).
   https://www.rfc-editor.org/ien/ien22.pdf
"Introduction and Objectives - Cerf
Vint primarily presented a summary of the results of the TCP meeting which
took place the preceeding [sic] two days.
The main result is a decision ... to provide for a set of parallel
protocols that all utilize a common "datagram" type internet protocol."

Danny C. and Jon P. -- two exceptional people, whom we miss....

John Shoch
[PS:  To put this history in context -- imagine we were all again in that
room at ISI in 1978, arguing about who said what 42 years before, in
1936.....  Fortunately, we now have more documents preserved on the ...
internet....]