On Wed, Mar 05, 2003 at 10:36:54PM -0500, Glenn Maynard wrote: > Is there consensus that DFSG#10 really is a grandfather clause? I've seen > this interpretation offered a number of times, but I've never seen any > strong agreement to it. I find the interpretation hard to buy, personally, > given that it's presented as "examples"; I'd sooner read a conflict > between DFSG#10 and other clauses as a simple inconsistency that needs > repair, though that may be less convenient. I have mixed feelings about this. In the ideal, I share your perspective, but we need to be careful not to render ourselves irrelevant by declaring the 3-clause BSD license or GNU GPL v2 as non-DFSG-free. If we have problems with those licenses -- as some of us do -- I think the only practical recourse is to work to reform them, whether formally (say, by conferring with representatives from the FSF, as we've done recently with GNU GPL (2)(c)) or in their actual application to various pieces of software (as I did a while ago in the BTS with a bunch of uses of the 4-clause BSD license). Those two licenses are so incredibly ubiquitous that I don't think any other approach would be accepted by the Project. The Artistic license is another matter; its original version is atrociously written and, in my opinion, too ambiguous to be safe. I think the original Artistic license should be stricken from the list of DFSG-free licenses because of this, and because the practical consequences would be minimal: 1) most stuff that uses this license is dual-licensed under the GNU GPL v2; 2) the original proponents of the Artistic license are promulgating the Clarified Artistic license, which is vastly superior in its wording, and -- the last time I looked at it -- pretty clearly DFSG-free; 3) the promulgators of the Clarified Artistic license have at least some authority to relicense most of the works in Debian main that use the original Artistic license -- namely, Perl. I have often used the term "grandfather clause" myself when referring to DFSG #10, but perhaps that term is too slippery, or inadequately communicative of my real meaning, which I just expressed above. If someone wants to come up with a catchy term for the above doctrine, I'm all ears. :) If/when, we reform the DFSG, I think any reference to a specific license should be moved out of the normative part of the document, and perhaps eliminated altogether (though such references could certainly continue to exist in adjunct documents that "explain" the DFSG). -- G. Branden Robinson | Software engineering: that part of Debian GNU/Linux | computer science which is too email@example.com | difficult for the computer http://people.debian.org/~branden/ | scientist.
Description: PGP signature