The Multics Book

Tuesday, September 06, 2022

From Amazon, I’ve acquired a copy The Multics System: An Examination of Its Structure by Elliott I. Organick. It arrived today. Published in 1972, this particular copy was withdrawn from the Fenwick Library at George Mason University (I guess even university libraries have to prune occasionally), but it’s in extremely good condition. It even has a sheet of paper inserted at p. 228 to note an erratum.

I have, briefly, run and logged into Multics. You can do this now, albeit on a simulator (the last real Multics site went dark in 2000). I didn’t do much other than poke around, though I was amused to see that Emacs worked just dandy on it—Emacs is everywhere. Still, Tom Van Vleck, who runs the Multicians site dedicated to collecting information and stories associated with Multics, reviewed the book in 2002, cautioning that it “describes Multics as it was planned in the late 1960s, before it had been released to users…. Multics became a commercial product…and continued to evolve for 20 years….” Very well, I am forewarned.

I vaguely recall being told that one of the criticisms of Multics at the time was that it was big and bloated; and indeed, in the forward (which is by Fernando J. Corbató) I read that “Today the Multics system is by most standards a large one involving about 1500 modules each averaging 200 typewritten lines; the full system compiles into about a million lines of code.” Granted, fifty years have elapsed since the book was published, but for context, by one report the Linux kernel had a bit shy of 28 million lines in 2020. I don’t imagine it’s shrunk in the last two years.

Anyway, really looking forward to diving into this treatise on a historic project, and if I have any further observations, I’ll put them here on the blog.