bill-auger <[email protected]> wrote .. > some of these criteria will no doubt require clarification
But hopefully not to the point of trying to document and cover all possible cases. This is probably not possible. Even if it were if evaluations were treated as a simple, automated, rote process where someone worked from some sort of fixed recipe then things would probably be missed. Meaning that having some vagueness is actually a good thing rather than trying to spell out all possible things and then missing things or not hitting the intended point. This leaves things open to have a discussion on the mailing list, potentially with different answers in different cases depending on the specifics of the circumstances. > FSF does not consider data (raw information) or art to be > "practical"; I'm not sure what you mean by "data (raw information)" but even if something were to fall into that Non-functional Data part of the definition, it must at least provide permission to copy and redistribute, both for commercial and non-commercial purposes. So it's also not acceptable for it to be entirely proprietary either. > it is not at all clear what this criteria covers Things that are neither software nor documentation. Think fonts, PostScript Printer Description files, data sets used by speech recognition software to learn how to recognize words, hyphenation patterns in dictionaries, etc. It's probably not possible to make an exhaustive list of everything, which circles back to the point I made at the beginning that it's best to evaluate things in light of the specifics of a given case instead of trying to enumerate things.
