[ih] A name from the past

[John Day]

> On Apr 26, 2022, at 21:26, Noel Chiappa via Internet-history <internet-history at elists.isoc.org> wrote:
> 
>> an article "Data Communications at the NPL (1965-1975)" in the 'Annals
>> of the History of Computing' (as it was then; Vol 9, No 3/4, if anyone
>> wants to look at it)
> 
> This has some other interesting tidbits.
> 
> On pg. 231, it observes: "This document [Scantlebury, Bartlett, "A Protocol
> for Use in the NPL Data Communications Network", April, 1967] contains what
> appears to be the first occurrence in print of the term 'protocol' in a
> data communications context". Interesting!

Perhaps. When were the SWIFT and early airline networks done.  Baran didn’t use the term?
> 
> On pg. 240, it describes the NPL network's 'isarithmic congestion control',
> defined as a "computer could .. inject a data packet into the network only
> [in place of another]", which sounds remarkably like the basic principle of
> Van's congestion control design!

How do you figure?  I thought Van’s congestion control was based on AIMD?

Isarithmic congestion control is essentially static allocation of buffers which was known (by their own admission) to be very inefficient.

The earliest congestion control work on connectionless networks was done at the University of Waterloo in the mid-1970s with extensive simulations and modeling.

The NPL Net was packet-switched by virtual-circuit and I think it was a star network, but don’t quote me on that yet. ;-)

> 
> Finally, in the "Acknowledgements" section, we find listed a "B. E.
> Carpenter", who I rather suspect is our Brian Carpenter! The paper came out
> in 1988, and Brian had done his and Doran's book on "Turing's ACE Report" in
> 1986, and it involved close interaction with NPL; its Acknowledgements
> section thanks Martin Campbell-Kelly, the author of the NPL history paper.

John
> 
> 	Noel
> -- 
> Internet-history mailing list
> Internet-history at elists.isoc.org
> https://elists.isoc.org/mailman/listinfo/internet-history