Brian Thomas Sniffen wrote: > Matthew Garrett <mgarrett@chiark.greenend.org.uk> writes: >>Andrew Suffield <asuffield@debian.org> wrote: >>>This idea is a variation on "You may not use this software for >>>military applications" and goes against DFSG#5/#6. They're both >>>intrinsically non-free, no matter how laudable you may consider them >>>to be. >> >>Why is discrimination against people who want to sue you significantly >>different to discrimination against people who want to distribute >>binaries without source? Neither prevents or restricts use, modification >>or distribution of modified works. > > Er, yes, discrimination against distribution without source is all > about protecting the ability to modify the code. > > But in any case, the difference in my mind is about causality -- where > did each party get the property right in question? In a copyright > case, it's clear that the modified work comes from the original work, > so it's OK for the original author to control the modified work > somewhat. In a patent case, the property right to the patent existed > before the "original" software was ever written. For the person who > wrote the software *after* the invention was patented to try to > blackmail the inventor is horrible[1]. No "property rights" exist at all in these cases. "Intellectual property" is a misleading and incorrect term that presupposes such things should be treated as property. >>>You cannot use a license to enforce your political position. >> >>Why is copyleft other than the use of copyright to enforce a political >>position (ie, that the source should always be available to people with >>binaries)? > > Copyleft is very narrowly targeted -- if I don't like your copyleft, I > can always reimplement the code. I can't reimplement around a patent > problem, because it's the method or technique that's patented. This sounds to me like an argument *for* limitation of patent rights, not an argument against patent "retaliation" clauses. > 1: Acknowledged, software patents are foolishly and incorrectly > granted all the time. So fix that system, don't introduce a new > brokenness in the DFSG. The software-patent system (or more generally the patent system, but let's not get into that argument since we only care about the software case here) is fundamentally broken. The only fix is to remove it entirely. Software patents, when granted, are *always* "foolishly and incorrectly granted". There are no legitimate software patents. - Josh Triplett
Attachment:
signature.asc
Description: OpenPGP digital signature