People seem to find the idea that free software existed before Richard Stallman birthed the GNU philosophy and the Free Software Foundation (out of the primordial ylem, one presumes) surprising. Here's an article I recently posted to /. about it:
That's just it: Stallman didn't create the "free software" philosophy as much as react badly to someone's use of one of his programs and take a radical divergence from the rest of the folks in the whateveryoucallit-software community who were mostly just writing good code and hoping people would use it.
Here's some publications and programs that you can look into:
Software Tools: Kernighan and Plaugher (the original, more than the Pascal version). The Software Tools group eventually had a regular tape release containing Fortran and Ratfor versions of many of the important UNIX utilities. From a 1983 conference announcement:
The Software Tools Users Group is focused on a set of license-free, UNIX-like utilities and system calls, written in Ratfor and Pascal. When run in conjunction with almost any local operating system, the Tools package presents a virtual operating system interface consisting of a virtual machine (system calls or "primitives"), utility programs, and a command language, thus achieving inter-system uniformity over a variety of operating systems. Originating from Kernighan and Plauger's book Software Tools, the enhanced package now includes programs for text formatting, mail systems, enhanced Ratfor preprocessors, a source code control system, command line interpreter similar to the UNIX Shell, and many other utilities.
Doctor Dobbs' Journal of Computer Calisthenics and Orthodontia, especially the first few years before they turned into a Microsoft-only rag. It's ironic, you know, that the GPL first reached the general public via Dr. Dobbs' Journal... and now people have forgotten how much Dr Dobbs did for the whateveryoucallit-software community and think Stallman invented it. Just about as ironic as the idea that Eric Raymond's responsible for the opposition to the GPL, I guess.
Byte magazine, the early years, when they had code listings. For that matter, just about any of the hobby computer magazines of the era.
The Forth Interest Group's FIG-Forth releases.
Ron Caine's Small C Compiler, and its successors. First published in Dr Dobbs'.
The early Usenet sources newsgroups, which had distributed thousands of programs before anyone had ever heard of the GPL.
Most universities had regular source code releases, with no significant restrictions. The Berkeley tapes were actually more restricted than most because they included AT&T-derived code so you needed an AT&T UNIX license to get them. BSD code was otherwise almost completely unencumbered, and parts of it that didn't contain AT&T code did get redistributed separately... a process that eventually culminated in the Net/2 release and 4.4-lite.
Another thing to consider is that US copyright law was in rather a turmoil at the time. Before the US joined the Berne convention, any software release was into the public domain unless it was explicitly copyrighted. Some of the programs you'll find if you look will have odd restrictions or non-functional ones (for example, lots of people would release stuff with incomplete copyright marks), but many didn't bother... they wanted people to take their work and make something of it. You had "educational use only", "no commercial use", "postcard-ware", and all kinds of other variants.
The GPL was first seen as just another of these, and the reaction to clause 2b was pretty much negative... the opposition to the viral nature of the GPL started very early and it was the result of an existing community reacting to a kind of coup, not commercial interests fighting a new force in the world.
The fact that people are surprised by the very idea that this community could have existed without Stallman, well, that should give you an indication as to how effective that coup has been, and should help explain why the people who built it tend to react badly to the man.