Sorry, it's getting confusing as it seems this list is forking a lot -- some of the replies are going to endsoftwarepatents.org and others aren't. For simplicity (and since I have no idea who is on that other list or how big it is), I'll just keep my reply within MFSIG. On Sun, Feb 12, 2012 at 2:42 PM, Ben Sturmfels <ben@stumbles.id.au> wrote:
The first three paragraphs are quite flawed though. I realise I'm going well off topic, but I'd like to address Matt's points about free software:
Endowing a benefit to free software by excluding it from patentability in no way "transforms the BSD into the GPL". Sure if you use BSD-licensed code in a proprietary program you wouldn't receive the proposed patent exclusion benefit. That's no different to today's situation though. This suggestion really is just inciting GPL/BSD flaming.
I'm not trying to incite flaming. The reason I said that it transforms it is that at least in today's environment, all free software developers (GPL and BSD-like) know to avoid patents wherever possible. I'm not going to write a BSD-licensed video player that uses H.264 when I can use WebM. Under my (hypothetical) BSD philosophy, I'm happy for proprietary software manufacturers to use my (hypothetical) video player. Now let's make a law that exempts free software from patent litigation. Now free software developers are going to stop worrying about patents, and I may very well implement a video player using H.264 (being the more popular standard), without fear of patent litigation. However, what I may have forgotten is that my downstream proprietary software manufacturers (who I am philosophically happy to use my software) will be unable to use my software without patent risk. So while the *copyright* status of my software is still permissive (BSD), the *patent* status is effectively at a GPL level of restriction (you can only use this in a free software project). When you say "That's no different to today's situation though," I think it is different, because free software projects will be more inclined to implement patent-encumbered standards. What I am really getting at here is that I think "software freedom" is about more than "free software". Free software is great, but software freedom is deeper: it is about my right to use computers for any purpose I desire. That includes the right to produce proprietary software. Patents are an attack on software freedom, and it would be short-sighted, I think, to consider them to be merely an attack on free software. I am nervous about this proposal because it promises a world in which the free software community is no longer worried about patents, but in which my freedom (to write proprietary software using patented technology) is still restricted. That, in some ways, *strengthens* the patent system, because patent holders know that they can convince free software projects (such as web browsers) to implement patented technology as standards, and force any proprietary project that wishes to follow those standards to pay up. Free software makes no distinction between software distributed at zero
price and software sold for a fee. This means the questions posed about cost aren't relevant.
Fair enough. Discussions of copyright are irrelevant to discussion on patents. Besides
that, endowing a copyright benefit to free software would be endowing a benefit to society as a whole. There's no hypocrisy there.
-- going off topic -- I was making an analogy between RMS asking for special copyright provisions for free software and this discussion asking for special patent provisions for free software. What disappointed me about asking for a copyright benefit to free software is that it would break the thing that makes the GPL so special in the first place: it does nothing other than grant *additional rights* to that of copyright. To quote the GPLv3: You are not required to accept this License in order to receive or run a copy of the Program. ... However, nothing other than this License grants you permission to propagate or modify any covered work. These actions infringe copyright if you do not accept this License. Therefore, by modifying or propagating a covered work, you indicate your acceptance of this License to do so. There are people who find the GPL's terms unacceptable. They say "the GPL is bad because it locks me into this set of restrictions." To those people, I say, "ah, but you are a proponent of proprietary software -- you cannot possibly have a problem with the GPL because it grants you a superset of the rights that regular copyright grants you. You don't have to accept the GPL if you don't want to -- you can simply treat it as an ordinary copyrighted piece of software if you wish." Having a shorter copyright term, but keeping the long term for free software would mean that the above clause is no longer true. In this world, I would have certain freedoms with proprietary software that I would not have with GPL software. For example, I would be allowed to reverse engineer a 5-year-old proprietary program and incorporate it into my proprietary software, while the GPL software would still be under copyright and I could not do that. So the GPL would no longer be merely granting additional rights to the user; it would also be taking away some rights (the lapse into the public domain after a short period). -- end off topic section -- On Sun, Feb 12, 2012 at 5:54 PM, Adam Bolte <abolte@systemsaviour.com>wrote:
With free software patents out of the picture, most probably one could expect that attacking the remaining software patent issues would then be easier. I could imagine software companies eventually turning around to complain that free software is too competitive and they need to be exempt from patents, and we would have proven that changing the patent system is possible, so maybe we would get even more support. Then there's the additional reasons I mentioned before.
I wouldn't necessarily think that it would be easier to attack the remaining patent issues. What that would accomplish, in my view, would be to divide the free software and proprietary software communities. Once the free software community has their patent problems solved, do you think they will continue to campaign against patents? No ... patents would be seen as "solved" in our community, leaving the proprietary software community to fend for themselves. You may argue they have a lot more money to combat the issue, but they also don't have the passionate community support, and they aren't necessarily against patents either (I don't know if Apple or Microsoft seriously want to see the end of software patents). I feel like taking action specifically against free software patents would be like throwing everybody else under the bus to further our own cause. As I said earlier, I don't see patents as a free software issue. I see it as a global software issue. If I could put it in terms of the recent SOPA debate, what would have happened if it was just the free software community that got angry about SOPA, and staged large protests, and finally convinced the government to make a provision that said free software sites could not be blocked under SOPA, but all the other sites still could. We might consider that a victory, but wasn't it more important to protect free speech than simply the sites run by members of our community? Now, we have an even worse scenario, where we have SOPA, but nobody in the free software community is going to fight against it because they got their way already. I don't think it's appropriate for the free software community to "claim" the patent war in their own name. If we're going to go with a "divide and conquer" approach, I would be much more interested in securing small (but global) repeals in the patent system. As the questioner suggested: a) requiring reasonable proof that the patent was copied (not penalised for independent invention) and b) having a "fair use" provision for interoperability with standards. Those two things would be far more reasonable than killing patents entirely, and would satisfy Adam's desire to "have proven that changing the patent system is possible."