On Sat, Feb 09, 2008 at 12:21:17PM +0100, Ralf Ertzinger wrote: > On Fri, 8 Feb 2008 16:09:04 -0500, Alan Cox wrote > > > A better test would be whether it is possible to produce new free > > data sets for them. > > ...and somebody has already done so. Because in theory the above is > always true. Ok perhaps we should replace possible with practicable The thing is we have a lot of tools which have dependancies on or are primarily used with proprietary data sets. Most of those nobody sees as a problem. I can use gcc to build proprietary software or free software I can use the quake engine to play proprietary or free game sets etc.. My email client spends almost all its time reading copyrighted works by third parties (email messages) that are not freely licensed. The case where the software is not truely free isn't the case where there is no existing data set, but the case where it is not resonably practicable to build one. Alan -- fedora-devel-list mailing list fedora-devel-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/fedora-devel-list
