[Ivan Kohler] > We will be guided by the needs of our users and the free software > community. We will place their interests first in our priorities. > Currently GFDL is a license acknowledged as free by the great mass of > the members of the free software community and as a result it is used > for the documentation of great part of the currently available free > programs. The problem with this is that it assumes developers make informed decisions to use a particular documentation license, and not just take whatever the FSF throws at them. At least once, and I think more than that, someone has said "WTF? How can Debian say something from the _FSF_ is non-free?" Then I explain a bit about _why_ the GFDL is considered problematic, and I get the reply, "Oh. Yeah. So what license should I use, then?" At which point I have to explain that the Creative Commons family of licenses are better, but may also have their own little bugs. I end up suggesting GPL or MIT-like things. This illustrates two things. One, people tend to assume, without really checking, that anything issuing forth from the mouth of RMS is a free license - but often they can be convinced otherwise with a bit of explanation. And two, there aren't a lot of high-profile alternative licenses targetted at natural language text (i.e., documentation), so the GFDL gets attention partly because it's in a thin field of options. All of which is to say, the mere fact that lots of people hack on free software and also use the GFDL doesn't necessarily mean a lot. > If Debian decided that GFDL is not free, this would mean that > Debian attempted to impose on the free software community > alternative meaning of free software, effectively violating its > Social Contract with the free software community. Social Contract point 4 is repeatedly employed to imply that anything and everything that might possibly inconvenience some users somewhere is wrong. This case is no more valid than others. Serving users and free software does not mean doing everything they think they want, or abiding by everything they think they believe. If indeed the GFDL has some problematic conditions, which is what this vote is about, issuing a statement saying so is a way to raise awareness of a problem many users may not be aware of. That most definitely serves both users and free software. > thinking we really need to work this out with the FSF and the > community towards a better GFDL v2, not issue divisive proclimations. That has been tried. For over three years. We know invariant sections are here to stay, but some of the other problems with the FDL seem to be entirely unintentional on the part of the FSF - mere bugs in wording, cases they seem not to have considered. If they really were acting in good faith, don't you think those could have been corrected some time in the past three years? Couldn't we have seen at least a _draft_ of a new, better GFDL by now? At some point you have to accept that you're being strung along by endless delays and empty promises. (It should also be noted that the proposed GR is careful to note the version number of the problematic GFDL text (version 1.2), and that later versions may correct some or all of its flaws.) > How are you voting? Why? Convince me. :) I'm not a Debian developer, just an interested bystander, possibly a future developer. I'd vote for the original GR, then Dato's amendment, then further discussion. > Bonus question: Is GPLv3 draft DFSG-free? If not, why? :) Some other day. (: It has a few problems, but they are very minor and it seems to me they will probably be worked out before the document becomes final.
Description: Digital signature