[-- Attachment #1: Type: text/plain, Size: 878 bytes --] This is tangentially related to Unix, and came up randomly at work yesterday. In Kernighan's Unix memoir, on page 9, he touches briefly on the typography of "Unix": "(Multics was originally spelled MULTICS, but the lower-case version is less visually jarring; as with UNIX versus Unix and some other all-caps words, I’ll use the nicer-looking form even though it’s not historically accurate.)" Here, he is talking about interning at MIT in 1966. bwk would certainly know better than me, but I can find no historical reference to this "MULTICS" spelling; is anyone familiar with that? The earliest reference I can find (the 1965 paper from the FJCC: https://dl.acm.org/doi/pdf/10.1145/1463891.1463912) uses the more "Multics" styling, but it may have been typeset later. Alternatively, could someone send me Brian's email address? - Dan C. [-- Attachment #2: Type: text/html, Size: 1603 bytes --]
[-- Attachment #1: Type: text/plain, Size: 1105 bytes --] The Book (https://mitpress.mit.edu/books/multics-system - I've always assumed this is the canonical non-mimeographed reference) uses "Multics". /Frew On 2022-05-11 10:13, Dan Cross wrote: > > This is tangentially related to Unix, and came up randomly at work > yesterday. > > In Kernighan's Unix memoir, on page 9, he touches briefly on the > typography of "Unix": > > "(Multics was originally spelled MULTICS, but the lower-case version > is less visually jarring; as with UNIX versus Unix and some other > all-caps words, I’ll use the nicer-looking form even though it’s not > historically accurate.)" > > Here, he is talking about interning at MIT in 1966. bwk would > certainly know better than me, but I can find no historical reference > to this "MULTICS" spelling; is anyone familiar with that? The earliest > reference I can find (the 1965 paper from the FJCC: > https://dl.acm.org/doi/pdf/10.1145/1463891.1463912) uses the more > "Multics" styling, but it may have been typeset later. > > Alternatively, could someone send me Brian's email address? > > - Dan C. > [-- Attachment #2: Type: text/html, Size: 2689 bytes --]
> From: Dan Cross > In Kernighan's Unix memoir, on page 9, he touches briefly on the > typography of "Unix": > "(Multics was originally spelled MULTICS ..." > Here, he is talking about interning at MIT in 1966. bwk would certainly > know better than me, but I can find no historical reference to this > "MULTICS" spelling; is anyone familiar with that? I looked at my early Multics stuff, and it's "Multics" almost everywhere: - "GE-645 System Manual", GE, 1968 - "The Multics Virtual Memory", GE, 1970 - "Introduction to Multics", MIT MAC TR-123, 1973 However, in my "A New Remote-Access Man-Machine System", on the title papge it says "Reprints of the MULTICS system presented at the" [FJCC, 1965]. No clue as to who printed it, or when - and all the FJCC papers themselves use "Multics". I have yet to ask Jerry Saltzer, but I suspect that if it ever was 'MULTICS', it was at a _very_ early stage, and was formally changed even before the FJCC papers (which were themselves very early). BTW, ISTR hearing that it was 'Unix' originally, and the 'UNIX' spelling was adopted at the insistence of Bell lawyers. So I went looking for an early (i.e. PDP-7 era) scanned document, to see what it was then, and all I could find was: https://www.tuhs.org/Archive/Distributions/Research/McIlroy_v0/UnixEditionZero.pdf which seems to be from just after the PDP-7 -> PDP-11/20 transition, and it uses 'UNIX'. Would the Bell lawyers have already been involved at that stage? Noel
The main FJCC 1964 papar, by Vyssotsky, Corbato, and Graham, spelled Multics with an initial cap. By contrast, Ken transcribed the aural pun as UNIX. The lawyers did their best to keep it that way after most of us had decided it looks better as a proper noun. As I recall, there was an acronymic reading of Multics, but it wasn't taken seriously enough to drag the word into all caps. Nobody proposed an acronymic reading of UNIX. So both words defy the convention of rendering acronyms in upper-case. Doug
[-- Attachment #1: Type: text/plain, Size: 1276 bytes --] Thank you, Doug. I wrote to Brian, who responded very quickly, suggesting that he was likely mistaken. He's going to make a note in the errata for his memoir. Tom Van Vleck also wrote saying that he was unaware of there ever being an acronymic rendering, and that he recalled an early meeting in which Jerry Saltzer was quite adamant that Multics was a proper noun, not an acronym, and therefore mixed-case. He did say that occasionally people joining the project would mistakenly write 'MULTICS' until corrected; apparently some of the GE folks in Phoenix were in the habit of doing this, perhaps due to prior familiarity with GECOS. - Dan C. On Wed, May 11, 2022 at 7:49 PM Douglas McIlroy < douglas.mcilroy@dartmouth.edu> wrote: > The main FJCC 1964 papar, by Vyssotsky, Corbato, and Graham, spelled > Multics with an initial cap. By contrast, Ken transcribed the aural > pun as UNIX. The lawyers did their best to keep it that way after most > of us had decided it looks better as a proper noun. > As I recall, there was an acronymic reading of Multics, but it wasn't > taken seriously enough to drag the word into all caps. Nobody proposed > an acronymic reading of UNIX. So both words defy the convention of > rendering acronyms in upper-case. > > Doug > [-- Attachment #2: Type: text/html, Size: 1702 bytes --]