| |||||||||||||
Prior art won't solve the software patent problem (NewsForge)Prior art won't solve the software patent problem (NewsForge)Posted Sep 18, 2006 15:47 UTC (Mon) by BrucePerens (guest, #2510)In reply to: Prior art won't solve the software patent problem (NewsForge) by smoogen Parent article: Prior art won't solve the software patent problem (NewsForge) RMS need not present an alternative for prior art, since prior art won't solve the problem and is indeed being put up as a sham to convince European regulators that it's OK to go ahead with making software patents enforcible there. Look at the companies on OSDL's board: the largest software patent owners there are. Legislative change is the only solution. If that means we have to stand having some people call us hippies, well, sticks and stones... Bruce
(Log in to post comments)
Actually, he was offering an alternative ... Posted Sep 18, 2006 16:01 UTC (Mon) by JoeBuck (subscriber, #2330) [Link] RMS's alternative is for there not to be any software patents. This might not be politically feasible in the US, but it certainly is in Europe.
Actually, he was offering an alternative ... Posted Sep 18, 2006 17:51 UTC (Mon) by josh_stern (guest, #4868) [Link] This comment seems correct to me, but more of a case would need to be made to convince me that RMS's strategy is correct here.Five implicit premises are on or near the table: 1. that the U.S. patent office will in the future always be on the side of those trying to get new patents; 2. that the existence of things like OSDL database will help persuade European Union legislators to favor patents there; and 3. that people putting energy into things like the OSDL database might otherwise devote their energy to anti-patent lobbying. 4. that the existence of such a database as a resource will help to make the same patents that would otherwise be granted stronger, rather than saving everyone a bunch of bother and nonsense by leaving them overly broad (and granted). 5. that no free software will get written, which otherwise would not have, because the authors or their employers are more comfortable with the relevant patent situation. While "1" is true, the fewer bad patents granted, the better. If "1" were no longer true, then much of the backfire argument goes away as well. Making "1" no longer true by lobbying for change in the incentive structure of the patent office might be a more approachable goal than getting rid of software patents all together. I've got no opinion about the truth of "2" but it seems like something that needs more argument; it's not obviously true. The truth of "3" and "4" seems pretty questionable, and a little imagination should allow one to see that "5" is false. In summary, it seems to me that the OSDL database could obviously do some overall good except if we believe the "backfire" argument. But the backfire argument rests on questionable supporting premises that RMS does not bother to defend in the linked editorial.
Actually, he was offering an alternative ... Posted Sep 18, 2006 19:42 UTC (Mon) by BrucePerens (guest, #2510) [Link] I don't think #1 is valid. The U.S. patent office might be an enthusiastic implementor of software patenting these days, but for many years refused to grant them. That refusal was reversed by the supreme court in response to a case brought by IBM. Our big friend, IBM, right? Well, IBM created this problem for us.Having been told what to do by the judicial branch, the patent office does not have much choice. Some of the officers there, however, have told me they'd rather things were different. I can't see a way to fix this except with legislation. Europe is where the battle is currently being fought. It can be fought in the U.S., but a key part of that fight will be decoupling of software from other sorts of patents like pharmeceuticals. The drug companies want their patents bad enough to spend billions on the issue, and getting them out of the battle is our best strategy. Bruce
Actually, he was offering an alternative ... Posted Sep 18, 2006 20:29 UTC (Mon) by josh_stern (guest, #4868) [Link]
When I wrote "Making "1" no longer true by lobbying for change in the incentive structure of the patent office might be a more approachable goal than getting rid of software patents all together", what I had in mind was less than getting rid of software patents - more like making it so that the patent office performance was evaluated at least equally if they turn down a patent as grant it and that they go about their business in like conservative doctors with the motto "first do no harm".
Actually, he was offering an alternative ... Posted Sep 18, 2006 21:21 UTC (Mon) by BrucePerens (guest, #2510) [Link] Very good. But even that will probably take legislation.Bruce
Actually, he was offering an alternative ... Posted Sep 18, 2006 21:31 UTC (Mon) by josh_stern (guest, #4868) [Link] Legislation - one possible outcome of lobbying - should be understood as a surer solution than an executive branch order, though the chief executive doesn't seem to agree. :(
Actually, he was offering an alternative ... Posted Sep 18, 2006 23:39 UTC (Mon) by drag (subscriber, #31333) [Link] Contact your faithfull public servent from the 'House of Representatives' is the solution for us. That's specificly what these guys are for. Their purpose in life is to take up causes like this if there is enough of a grassroots movement for them.
It's a mistake to go through the judical branch of the government ala the EFF or how the this 'Prior Art' thing is going. Sure judicial activism is popular these days and it seems a nice way to enforce your opinion on others when you don't have the democratic backing of your ideas (because either if they are unpopular or most people just don't seem to care or are ignorant), but it's a mistake to go that way when you realy want to solve problems. From the court's point of view everything is 100% legal and constitutional and on the up and up. They simply can not do anything about it.
You may be able to nail a couple cases with prior art, but it's going to be about the equivelent of doing a insanity plea to get out of murder; It's used in less then 1% of murder cases and when it is it's successfull less then 1% of the time. (I beleive thats right)
Of course even if the concept (making software patents invalid via legislation) is relatively simple, doesn't mean its going to be easy...
Actually, he was offering an alternative ... Posted Sep 18, 2006 23:49 UTC (Mon) by PaulMcKenney (subscriber, #9624) [Link] > Of course even if the concept (making software patents> invalid via legislation) is relatively simple, doesn't > mean its going to be easy...
And if we fail to provide prior art to defendants, the money flowing to patent-based attackers will be used to make it even harder to prevail via legislation. Sad, but true.
Paul E. McKenney (speaking for myself, not necessarily for my employer)
Actually, he was offering an alternative ... Posted Sep 19, 2006 0:54 UTC (Tue) by drag (subscriber, #31333) [Link] My point is that prior art simply isn't going to be effective. That's all.
Is there any statistics on how successfull a 'prior art' defense in a lawsuit is?
That is something I am VERY curious about. I've heard that is very difficult and that almost invariably the judge is going to side against you if there is any gray area.
I think that it's to the point were devoting a lot of time and effort to it is pretty much pointless. Energies may be better spent in educating the public and making the case known that software patents are a threat to the growth of small and medium businesses and is to the point were it's stunting technological growth.
Actually, he was offering an alternative ... Posted Sep 19, 2006 1:45 UTC (Tue) by PaulMcKenney (subscriber, #9624) [Link] In my own experience, if you can quickly put together a solid case, you can often convince the attacker to back off entirely. After all, they want easy money, not a hard-fought court battle. Even if they have already filed suit, the presence of solid prior art can help convince the judge to refuse to grant a preliminary injunction, which, if granted, would shut the defendant down completely. Such an injunction puts the defendant at a serious disadvantage in settlement negotiations, and can in fact force the defendant to quickly settle.
Without such an injunction, the defendant is in a much better position -- there is time to try re-examination procedures on the plaintiff's patents, search for more prior art, come up with convincing non-infringement arguments, etc.
And, no, judges do not always favor the patent holder. Though I don't have the statistics handy, I would guess that you can easily get your hands on them.
All that said, I do readily agree that we should also educate the public on the dangers of software patents. Just not to the exclusion of other efforts, including prior art.
Paul E. McKenney (speaking for myself, not for my employer)
Prior art db necessary, as _part_ of our response to SWPats Posted Sep 19, 2006 20:20 UTC (Tue) by kenm (guest, #40607) [Link] I attended the recent OSDL meeting on this topic, as an interested bystander. I attended, not because I like software patents, but because I am Scared To Death of them. Attendance was free, the subject matter is monumentally important. Why weren't you ALL there? I use GPL and other OSS software. A lot. I create software for a living. No conflict there. Yet I could not hope to avoid conflict with the current SWPAT system, because under the current system, things that I consider obvious, trivial, and essential will be patented. During the OSDL meeting, I had a chance to discuss this with a patent attorney. In the current situation, the mere aggregation of two well-known, non-patentable things, is patentable. Think about that. The patent office will grant a patent, merely because no existing patent covers it. Yes, that's crazy, and that's the real world for you. So a prior art database is essential; the difference is between A) documenting prior art, so it can be used to fight "bad" patents, and B) having no available database of prior art, and being unable to fight. The prior art database will need to be chock full of the painfully obvious. It will need to cover things you and I take for granted. So, if this prior art database backfires, it is only because we didn't put enough into it. Yes, we need to "fix" the law, but it won't happen overnight. Given the capitalist incentive to use patents as a weapon, it may never happen. I'm glad large companies like IBM are promoting and defending OSS and the GPL. I'm glad that IBM realizes they have as much to lose as I do. And I'm glad that RMS is out there, fighting for my rights.
Prior art won't solve the software patent problem (NewsForge) Posted Sep 18, 2006 17:22 UTC (Mon) by emkey (guest, #144) [Link] Sticks and stones may not break our bones, but poor public perception certainly can.
Note, I'm all for preventing software patents where they don't exists and getting rid of them where they do but it isn't clear to me that those two options are feasible.
Building up a library of prior art and patents is a poor backup plan but it is far better then no backup plan at all.
Prior art won't solve the software patent problem (NewsForge) Posted Sep 18, 2006 20:02 UTC (Mon) by BrucePerens (guest, #2510) [Link] Sticks and stones may not break our bones, but poor public perception certainly can.But not nearly as badly as assuming we'll fail before we start. Sure, some people will try to label us hippies, but that just means that we have opposition. It's taken for granted that you will have to fight that sort of stuff whenever any kind of attempt to reform law is taken on. Bruce
Prior art won't solve the software patent problem (NewsForge) Posted Sep 18, 2006 20:28 UTC (Mon) by emkey (guest, #144) [Link] I don't assume failure, I fear it. Though I certainly see your point.
I also understand the concerns of those who fear that attacking this problem on to many fronts will dilute the effort and lead to failure.
However, I tend to think there is a happy medium ground. I also tend to dislike stances that claim there is one true and perfect approach to solving a particular problem. In my experience this is never true. As a general rule there are multiple paths to any particular destination. Each of them will have somewhat different tradeoffs of course and that should certainly weigh into the path each of us chooses to follow.
I also feel that in most cases it is counterproductive to criticise other people for choosing a different path in situations where everyone is heading for the same destination.
Prior art won't solve the software patent problem (NewsForge) Posted Sep 18, 2006 21:29 UTC (Mon) by BrucePerens (guest, #2510) [Link] I would not feel the need to be critical of OSDL's efforts except that I feel they will do harm to Open Source.OSDL's program will be used as a sham to convince governments that the problem software patenting presents for Open Source is being dealt with, and that regulators can go ahead and impose enforcible software patents in Europe. And that may just end up killing the golden goose. Bruce
Prior art won't solve the software patent problem (NewsForge) Posted Sep 18, 2006 21:41 UTC (Mon) by emkey (guest, #144) [Link] Understood. Though it isn't hard to argue that the OSDL solution is no solution at all, meerly a resource that makes it somewhat easier for the creators of open source software to defend themselves. In other words it is a poor second to real legislative action that would actually deal with the underlying issues.
So rather then attacking it I'd be inclined to say, "Look at the lengths we are having to go to in order to somewhat blunt the very fundamental and serious problems that software patents are causing."
Prior art won't solve the software patent problem (NewsForge) Posted Sep 19, 2006 7:39 UTC (Tue) by Wol (guest, #4433) [Link] One HUGE problem I see with OSDL's prior art database is the Patent Office's definition of "inventive and non-obvious".
Given that that definition seems to be "nobody has previously mentioned it in the literature" (which could well be because it's so obvious that no-one has bothered), that still leaves program version X.1 wide open to attack for minor upgrades even if X.0 is in the prior art database!
Actually, should we actually have a case like this, it might provided some hefty anti-patent ammunition ...
Cheers,
Prior art won't solve the software patent problem (NewsForge) Posted Sep 19, 2006 16:25 UTC (Tue) by smoogen (subscriber, #97) [Link] Bruce, you state that the OSDL is run by the companies with the least interest in changing the patent system. I would argue that most elected representatives in European Union and United States are also 'run' by these companies too via donations, fancy trips, and all the other things that civilized nations gloss over bribery as. And so a simple lobbying campaign is not going to make changes unless the patent system becomes as important as other 'hot-item' poll numbers that representatives have to worry about every couple of years when they get re-elected.
And things aren't as good in Europe as they would appear to be. While the EU as a whole has not gotten into a software patent system, my understanding is that many member states have signed commerce treaties with the US that have similar rules as the recent Australian pact did. This means that a problem with a member states companies program that infringes on a patent would be covered by a US court and be binding in that member state... These backdoors are with most of the new members of the EU and various other countries that are in line for membership..
So basically the US patent system is getting used in more and more places even if it isn't shown/known. And it will stay in control as long as people think that it could be some persons lottery ticket to make several million dollars.
The only way I see of changing the attitudes of millions of voters is a) getting a large amount of prior art, b) a long list of bad/weak/poor patents that show how easy it is to defraud the country, c) a follow the money expose that shows that the companies have been basically 'paying' for the votes that continued the defrauding and which representatives have been on the til the most. And even then, I do not see software patents going away but their being adjusted in a way that better suits that segment of the economy (shorter length of monopoly, higher standards for awarding them, etc).
|
|||||||||||||