> Raul> It's also important to recognize that the DFSG does not even > Raul> address the problem of preventing buggy software. I feel that > Raul> stupid modifications of standards documents are in some way > Raul> analogous to this.
Manoj Srivastava <[EMAIL PROTECTED]> wrote: > Buggy software, when fixed, affects only Debian systems. But that's > OK, we only distribute stuff for Debian systems.Even in that case, we > are bound to feedback our changes upstream. Nope. When the debian systems implement servers non-debian systems may use them. > Standards are meant to be a a common ground that everyone, (not > just Debian) recognizes, and tries to follow. When we make changes > to the standard, the others do not follow; essentially we have > become non-conformant, and have injecxted a rogue document intot the > universe. As long as there's no fraud involved, I don't see the problem. > A standard that no one follws is not a standard. > > I think the best way to proceed is become non-standard, and give the > reasons for doing so and attempt to get the standard modified. Of > find some other standard to adopt. Sure, a part of the process of issueing standards is issueing draft standards. That's certainly not the whole process. > Standards don't eveolve via unofficial modifications (at least, I > am not aware of a single instance of that happening). I have seen > standards bodies looking at unconforming applications, and updating > the standard. Well, there's defacto standards, but I believe the fundamental point you're trying to make is that a standard which is not being followed isn't a standard. Which is fine. I think we do need to make a place for standards which we can redistribute but can't modify. I don't think that place is main. Perhaps that place is contrib, or something like contrib. -- Raul
