[Air-L] The Date and the Version Thang

[Yuri Takhteyev]

Numeric software versioning most certainly goes back much further than
Unix. Grace Hopper's pre-Flow-Matic compilers were called A-0, A-1,
A-2, and A-3. A-0 was done in 1951, nearly two decades before Unix.
(This was still on Univac, and the word "software" hadn't been coined
yet.) The versions of Fortran were similarly labeled "Fortran I",
"Fortran II", etc.

http://bit.ly/a0comp

The first X.Y style version that I know of is Lisp 1.5, from early
1960s. It was called "1.5" because it was different from "Lisp 1" but
wasn't yet the "Lisp 2" that was already already discussed. So, "1.5"
was supposed be half-way between 1 and 2. After 1.5 there was a 1.6.
(There are some papers on this from the 1970s in ACM Digital Library,
but not on the open web, so no links.)

I am not sure whether the later X.Y style versioning was influenced by
Lisp, but that would be plausible, since Lisp 1.5 was quite popular.
But perhaps there was software with X.Y-style version numbers even
before the 1960s.

  - yuri

On Wed, Jul 28, 2010 at 8:12 PM, Denise N. Rall <denrall at yahoo.com> wrote:
> Matt - (apologies to jeremy who got this first) -
>
> I think Steven Levy discusses Macintosh methods for versioning in his classic:
> Levy, S. (1994). Insanely great: The lives and times of Macintosh, the computer that changed everything. New York, Viking. or perhaps it is in the Hackers book (same author).
>
> I also think there is a mention of versions/revisions in the other PC history classic:
> Frieberger, P. and M. Swaine (1999). Fire in the valley: The making of the  personal computer. 2nd ed. Berkeley, CA, Osborne/McGraw-Hill.
>
> But overall as I understand it, the revision naming conventions that the Mac uses came directly from unix. Look at Matt Ratto's work on Linux, might be something there. Also there's the Software studies people:
>
> Software Studies: A Lexicon Edited by Matthew Fuller (2008) MIT Press.