by Darius Kazemi, October 9 2019
RFC-282 is titled “Graphics Meeting Report”. It's authored by first-time RFC author Mike Padlipsky of MIT Project MAC and dated December 8, 1971.
The technical content
This RFC is the meeting notes from the long-planned and long-delayed Network Graphics Group Meeting at Stanford Artificial Intelligence Lab in late November 1971.
The document is divded into : reports on experiments, presentations, and the results of a discussion about a “
first-pass” network graphics protocol. (Keep in mind that many have been proposed to date, but none accepted.) I won't mention everything here but I'll offer some highlights.
RAND and UCSB demonstrate remote usage of the UCSB On-Line System via graphics terminals at RAND. They even show the film that they promised in RFC-225.
Several experiments are reported as having been “
plagued with problems” ranging from difficulty coordinating people across time zones to software and hardware bugs. Jack Bouknight from University of Illinois expresses “
concern as to whether the Network as a whole is as yet sufficiently reliable to support graphics work”. The Multics host is apparently the most reliable, and the consensus at the end was that the reliability of the network overall “
is not crippling to graphics work, but can be quite hampering”.
Various prepared talks include papers submitted ahead of time as RFCs, like RFC-178 on attention handling, as well as a presentation on issues in interactive graphics, a proposal for using Telnet for simple graphics, and how protocol needs are different for different kinds of graphics devices.
A committee is formed to make a graphics protocol, modeled on the committee that made the (wildly successful, in my opinion) Initial Connection Protocol. There is agreement that a handbook should be prepared specifically for network graphics. General consensus is reached on some basic assumptions to be embedded in a network graphics protocol, including:
- use of a virtual screen
- origin (0,0) is in the center of the screen with coordinates being signed fractions
- ASCII for text
- absolute, relative, and local (relative to a defined sub-frame) drawing modes
The protocol will be concerned only with display to begin with. Later iterations will tackle interactive graphics. The functions discussed are kind of the usual suspects we've seen in earlier protocol proposals, very similar to what is found in say RFC-86.
They hope to convene a second meeting in mid-April 1972, at which point they would like to have a draft protocol.
This article published in 2000 by Mike Padlipsky is full of juicy, gossipy reminiscences of the Network Working Group, RFCs, and the NIC . It also addresses the various claims that so-and-so “invented” the internet or “invented” email. In particular he calls out the idea that Ray Tomlinson invented email, and claims that Tomlinson was not even present at the meeting where they all agreed to add mail to the FTP protocol.
In fact, according to Padlipsky, he invented (though apparently this is a controversial claim) anonymous FTP specifically to help people grab RFCs from the NIC:
I remember pretty clearly [..] being at SRI for some sort of meeting sometime around 1973 and being told that “the NIC” (or at least Jake [=Elizabeth Feinler – PHS]) was worried about this idea to put the RFCs on-line, because they'd have to establish all sorts of accounts so people could FTP them. “That's easy,” I said, “just use my NETML trick.” By which I meant, and went on to explain, that just as I'd had to propound a conventional universal “dummy” id and password [...] all the NIC needed to do was establish a single, known account everybody could use to slurp the RFC's from. “'guest' would be a perfectly fine id,” I went on, “and the password should be 'anonymous', since we'd gain some measure of security in that people'd have to know how to spell it and of course not everybody does.”
Padlipsky died in 2011.
How to follow this blog
I'm Darius Kazemi. I'm an independent technologist and artist. I do a lot of work on the decentralized web with ActivityPub, including a Node.js reference implementation, an RSS-to-ActivityPub converter, and a fork of Mastodon, called Hometown. You can support my work via my Patreon.