Arpa address 3477

slight nit ... arpa had funded arpanet and development of packet-switching protocol. this was relatively homogeneous environment requiring fairly expensive interface "IMPs" to be part of the network.

the infrastructure cut-over to internetworking protocol was on 1-1-83 ... with gateways, internetworking ... etc.

I've frequently butterted that the internal network was larger than the whole arpanet-internet from just about the beginning until approx. summer 85 ... because the internal network had a kind of gatewaying capability from nearly the beginning (i.e. the 1-1-83 cut-over to internetworking had approx. 250 nodes while the internal network was nearing 1000 nodes).

NSF and other organizations were started to fund educational networking connectivity ... first csnet ... minor reference

then NSF released the NSFNET backbone RFP ... which should be considered the "operational" precursor to the internet (as opposed to the technology-protocol). as i've mentioned before ... my wife and I had an operational high-speed backbone at the time

but we were told we weren't allowed to bid on the NSFNET backbone RFP ... although we eventually got a technical audit by NSF ... that said something about what we had operational was at least five years ahead of all the NSFNET bid submissions. minor ref:

it was also starting in this time-frame that arpa and various other parts of the federal gov. were starting to mandate that internetworking protocol, internet, etc, was to be eliminated and every thing converted to ISO OSI protocol ... along with the various GOSIP documents ... recent reference about OSI mandate and GOSIP

the NSFNET backbone had fair use guidelines about not being used for commercial purposes. the transition to today's internet somewhat involved providing provisions for commercial traffic (much of the backbone involved commercial companies loosely under NSFNET contract to provide networking for educational and non-commercial purposes).

--