[ih] Nit-picking an origin story

[Lars Brinkhoff]

Jack Haverty wrote:
> Performance was also an issue as the ARPANET grew and traffic
> increased.  One of the limiting factors to performance was the routing
> algorithm.   Packets were always sent on the "shortest" path.   But
> that meant that the aggregate performance was also limited to
> 56kb/sec, which was the maximum line speed of any path.

I beleve the correct number is 50,000 bits/second.

ARPANET Information Brochure, from 1980.
"The complete network is formed by interconnecting the nodes through
wideband communication lines, normally 50,000 bits per second (50KBPS),
supplied by common carriers."
https://apps.dtic.mil/sti/pdfs/ADA096798.pdf#page=19

BBN Report 1928, from 1970.
"In the last quarter, we designed and implemented a test program to
obtain data on the performance of the fifty kilobit communication
circuits"
http://bärwolff.de/bbn-arpanet-reports-collection/BBN%20(1970)%20Interface%20Message%20Processors%20for%20the%20ARPA%20Computer%20Network%20(Report%201928,%20Quarterly%20Technical%20Report%204).pdf#page=10

This was due to the Bell/Western Electric 303C wideband modem using a
group service of 12 voice circuits.
https://bitsavers.org/communications/westernElectric/modems/303_Wideband_Data_Stations_Technical_Reference_Aug66.pdf#page=6