Subj : Re: GNU Public Licences Revisited (again) To : comp.programming From : Scott Moore Date : Tue Aug 23 2005 02:26 pm Randy Howard wrote: >>And by >>putting everything under the GPL, we ensure that the bad software makers >>will never be able to come back, since they can't pull the same >>"steal the code, tweak it, and release it as closed source" tricks they >>did back in the '80s. > > > Sorry, but I don't see this. There was very little 'open > source' software to steal from back then. Apart from the TCP > stack, what other examples can you point to? > This has become the urban legend of the GPL movement, but I believe it largely traces back to the AT&T vs. MIT case concerning X-Windows. MIT produced X-Windows, and let the source go free, unrestricted (even the GPL). This is about 1984 (according to Wikipedia). AT&T filed a patent over the ability to save the contents of a window, then restore that when it would be uncovered. This is the "backing store" patent. The patent was filed even though MIT clearly demonstrated the technique in their released code, and AT&T clearly got the idea from their code. AT&T then made MIT REMOVE the code to avoid voilating the patent. See: http://lpf.ai.mit.edu/Links/prep.ai.mit.edu/pike.demo Which was filed by Mr. Stallman himself, so I won't be accused of bias (?). The patent was clearly wrong, however, it was a patent, not a copyright issue, and plenty of bad software patents were issued back then, because the USPTO was (and is, IMHO), run by clueless morons. This would be unlikely to happen today, because the USPTO is less clueless, and MIT could have proven that the code was theirs by better record keeping (for example, by signing and dating a copy of the source code and keeping it for future records). In fact, just publishing on the web consitiutes legal proof, since there are web sites (wayback) that record snapshots of older web versions. This can and has been the basis for legal decisions. This has been used by the GNU organization as "proof" that openware software can be "taken private" by commercial entities, but it is not true. The other "proof" of (abscondtion ?) GNU uses is that a open source program might be "improved" by a commercial entity, and users might grow to like that better than the original, open source program, perhaps because they advertised it better (or other, non functional reason). Then, the users are "locked into a proprietary standard". However, this ignores the fact that the original, open source version is well available, and users are free to choose whichever version they like. If the GNU organization cannot get users to use their version instead of a proprietary version, that is hardly a freedom matter. It is more a matter of user preference. .