[ih] TCP adoption in 1984

[Greg Skinner]

On May 2, 2026, at 8:03 PM, Jack Haverty via Internet-history <internet-history at elists.isoc.org> wrote:
> 
> If "reached via TCP" means over the Internet, in the early 1980s the only wide-area connectivity was through the ARPANET, or perhaps SATNET to Europe.
> 
> LANs were appearing, but the key event might be when the first Gateway appeared, connecting the ARPANET to some Ethernet LAN.  That would make the TCPs on that LAN "reachable via TCP".   I suspect that TCP implementations on workstations first started by communicating just across their LAN.
> 
> I don't recall when the first Ethernet gateway appeared, or who did it.  May have been MIT, or Berkeley, or even UCL?  Or perhaps one of Dave Mills' Fuzzballs.  I don't recall that BBN did it.
> 
> Anyone remember when the first Ethernet IP gateway appeared, or who did it and where?
> 
> /Jack Haverty
> 

Not sure if this was the first, but this could be one of the earliest ones.  In the Network World 27 March 2006 interview of Bill Yeager, he states that he started work in 1981 on IP over 3Mb Ethernet for the router linking  Stanford’s EE department, CS department, and medical center. [1] (For those who are new to this list, how this router became part of Cisco has been discussed before.  Check the archives for more info.)

--gregbo

[1] https://web.archive.org/web/20141115160333/https://www.networkworld.com/article/2309917/lan-wan/lan-wan-router-man.html