RFC 152

SRI ARTIFICIAL INTELLIGENCE STATUS REPORT ✨

okay so this is RFC 152 🎀, and it's a response to RFC #116. It's written by M. Wilber for SRAI (NIC #6756), dated 10 May 71, filed under category G.3 💅. It obsoletes None and updates None, so nothing's being replaced or patched here. It's a fresh status update, bestie 🫶

The Stanford Research Institute's Artificial Intelligence Group (SRAI if we're doing four-letter abbreviations 🍓) is expecting to hop onto the ARPA net as a research center once they convert from their current PDP-10 to a TENEX this summer 🦋. Their connection goes through the IMP already sitting at SRI for the Network Information Center, plus a PDP-15 that runs their PDP-10 as a controller for peripherals ✨. Their hardware interface to the IMP is still in the bidding process, and they plan to reuse as much of the TENEX network software as they can. (Probably the only thing they'd need to change is the lowest level, most strongly hardware-oriented part of the NCP 😭.)

The most optimistic guess for being functionally connected is mid-July 1971 🌸. Right now their system support group is busy with various hardware and design changes, so their contact with the Network Working Group has been pretty minimal and passive 🙈. It's totally possible their participation gets stronger and more active once they cross their other bridges 🌈.

They can project their network participation into the first few months after connecting. They can support several outside users at once on a system that's nominally up during the business day and often up at other times 💖. Lapses in continuity are usually from scheduled maintenance or hardware failures, and they tend to happen every hour or several days with remarkable consistency 💀. Their main service to other network folks is access to parts of their own research software 👑. The big ones are QA3.6, a first-order resolution theorem prover; STRIPS, an extra-logical problem solver; and possibly QA4, a language oriented toward problem-solving strategies 🧚.

The services they'd want from the network come in two kinds. They could conceivably use other people's AI programs on a trial basis, and they might use the network to make occasional contact with other people in it 🫶.

        [ This RFC was put into machine readable form for entry ]
        [ into the online RFC archives by BBN Corp. under the   ]
        [ direction of Alex McKenzie.                   12/96   ]