On Fri, Apr 08, 2005 at 03:57:26AM +0100, Henning Makholm wrote: > Scripsit Michael Poole <[EMAIL PROTECTED]> > > > This has the strong smell of ranking some DFSG criteria above others > > in importance. If you want this kind of distinction, I think a less > > discriminatory way would be to flag (internally or on a central web > > site somewhere) each package in non-free according to which parts of > > DFSG it fails. > > I think it would be more manageable to flag freedoms that the package > still does provide, for example > > modified-noncommercial-redistribution > unmodified-noncommercial-redistribution > unmodified-commercial-redistribution > all-freedoms-in-the-gfdl > dfsg-freedom-of-all-runnable-programs > dfsg-freedom-of-all-main-cpu-runnable-programs
Euh, what are those two last ones ? > or preferrably some shorter names :-) Yes. We should declare such a field, and provide a description of those flags, and then we can go and examine all of those packages in non-free and submit patches to the maintainer to include this line or NMU them. As aj said there is no in-principle opposition to this, anyone can do this job, i believe, but it is best to clarify the terms used first. > That is, list reasons why somebody might want to *accept* the package > on his machine (or his redistribution) rather than list reasons why > somebody might wanto to *exclude* it. That way an overlooked tag would > lead to failure on the side of caution, and new tags could be added to > the system without retroactively reclassifying all packages in > non-free. Seems cool. CCing to debian-legal in order to obtain advice on the different terms and classification. Friendly, Sven Luther -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]