Slashdot Mirror
Perens on Patents
lewiz writes "An interesting interview with Bruce Perens over at the BBC. He's up discussing the role of HP, IBM, et al and the move towards Linux. However, his main point is about software patents and how they are much more of a problem than SCO: 'We're looking at a future where only the very largest companies will be able to implement software, and it will technically be illegal for other people to do so.'"
These companies will be the only ones developing software if they actually choose to enforce the patents. IBM has more patents every year than any other company(like the last 5 years running i think), but I haven't seen a high-profile case where IBM went after Joe's Software Shack for IP infringement(doesn't mean it hasn't happened).
Yes, patents are evil, but mostly only when they are enforced, otherwise it's more of a "Hey, look what we thought of".
Several people have filed patent claims on work I did, in one case 5 years after the idea had made its way into Apache.
And do't get me started on shopping carts...
Looking for an Information Security student project suggestion?
Try http://dotcrimeManifesto.com/
Speaking of interviews with Bruce, wasn't there supposed to be a /. interview with him? I remember submitting questions but not seeing the answers.
I watched C-beams glitter in the dark near the Tannhauser gate.
A decade ago (when it would have been easier for me than now), I was hesitant to go into music. The reason? I felt like there were only so many notes, so many rythms...and that every song of merit that could be written already had been.
Fortunately, I was wrong.
There will still be new ways of doing things. New languages, new platforms, new audiences, new ways at solving the same old problems.
Had Linux not come along, we'd be in MS world right now - UNIX owes it's life to Linux at this point (Linux kept it relevant). Point being is that it did. Everything that is "obvious" as a solution nowadays was radical, or even considered impossible, not that long ago. So what will be the solutions of the future? If I knew, and then told you, you'd probably either laugh, or think me insane.
The circle of life - the world is funny that way - Why? No one knows. Its magic. Yummy.
I believe it is up to other inventors to bring up their prior art when disputing a patent.
My sig is blank, I typed this by hand.
Many people seem to fall into this conceptual trap. Infringing on a patent (knowingly or unknowingly) is not illegal, but infringing on a patent without the consent of the patent-holder makes you liable. A patent isn't a law, but it provides the owner certain legal standing. There's a difference. If the patent holder doesn't tell you to stop using their method, then you're perfectly free to do so, and have no liability in doing so.
If using methods patented by others were illegal, then every company would have to stop, or be punished by the government. Microsoft couldn't develop something with a method published by IBM, and IBM likewise couldn't develop something "pioneered" by Microsoft. The interlocking illegality would seize up the development in big companies just as much as anyone else.
Many big companies hold huge patent portfolios for defensive purposes. They never complain about others using the methods they've patented, but they have a bargaining chip (or weapon) to use if someone else tries to collect on another method.
Many other companies like to hold patents without developing them, and to submit as many patent applications as possible, so they can try to collect when some rich but not threatening company stumbles across the same obvious methods. It's this phenomenon which creates the danger against which Bruce Perens is warning.
It's possible to keep patents, and to use them as the early founders of Patent Law intended: to promote the sciences by protecting their discoveries for a limited time.
[
OK.... Now I understand how bad patents can fubar software development for open source (and for closed source too) but there is something that nobody on Slashdot ever considers: Why not go out and get the patents done in a way that is open???
Despite what many people here think, patenting software does not make it closed source, in fact since a patent requires the disclosure of the best known means of implementation it can actually facilitate open code. Just because something is patented does not mean that it cannot be used in open source... it depends on who holds the patent and what licensing terms are.
If the Open Source community truly is innovating why not just patent the concepts and then place the patents in a licensing escrow: if you use the patent with a GPL license (or maybe LGPL/BSD/whatever open license you like) then the
patent is royalty free.... if you want to use it in a closed source program you could then charge royalties. After all, if closed source is about enforcing IP then they should put their money where their mouths are and pay, and this could even go to fund open source development!
I'm tired of seeing whining and helplessness on Slashdot when all you need to do is get up and proactively use the system in your favor. To all of you who will respond 'Only big evil companies can get patents' that is a bunch of nonsense, everyday people get them all the time and if enough interest was generated the FSF or another body could act as a clearinghouse to make it even easier to get patents put into the open domain by
software developers. It's about time we did something constructive about patents instead of just wailing about them.
AntiFA: An abbreviation for Anti First Amendment.
afaik, the patent office does not look for prior art, they look for prior patents. It would be nearly impossible for them to scour the world for general prior art in each patent they grant, but they can easily look over other patents.
Well, that is the big catch-22. Most of the stuff is (seemingly) so common or simple that nobody bothers to patent it. Thus, once somebody does come along with a patent that is butt simple, there is nothing in the existing patent records for it.
The second problem is the newly allowed "business processes". These also have no patent record because they are new. Congress should get off its ass and disallow biz process patents. The patent office says that the courts have allowed them and there are no laws disallowing them. Thus, they will continue to accept them until a law says otherwise.
Table-ized A.I.
We need to be asking our friends like IBM what they will do to help us. Our customers and users need to ask, as well. Many of them are IBM (and HP, etc.) customers too.
Bruce
Bruce Perens.
Lots of software companies have lots of patents. Sometimes they're silly, but it sets up sort of a Mutually Assured Destruction. Microsoft isn't going to try to bludgeon IBM with a patent suit, because they know IBM has just as many silly patents to bludgeon them back with.
It's because of this setup that we normally only see big IP cases come from companies that don't actually produce anything (and thus have little to fear from a counter-attack) - like SCO.
Let's not stir that bag of worms...
I've always said patents are for things and copyrights are for thoughts. Being as software is not a thing, it is data that you can't pick up and hold in your hand and look at and smell and feel (unless printed, but that would be ink and paper).
You can pick up a book and smell it and feel it but it smells and feels the same is the book that was sitting next to it on the shelf and the book that sat next to that book. The story contained within the book, of course is different than the one in the book next to it.
You can treat software the same way. Unless you have something physical, it shouldn't be patent-able.
One is not allowed to patent mathematical formulas. It is part of the original patent law. So, what is the difference between computer code and formulas? Not much, but generally it is the declarative nature of math formulas that appears to separate them in the mind of judges. So, perhaps if one writes key parts of programs in declarative languages, like Prolog, then they would be covered as a math formula. Even if it does not work, it would make a fascinating case to see lawyers and judges haggle over the difference between Prolog and math.
Table-ized A.I.
A patent effectively grants the holder a monopoly on exploiting an idea, or anything based on that idea, for a considerable period of time. Here in the UK it's 20 years, which is pretty long time by the standards of most commercial agreements. Given that, I figure getting a patent should be a time consuming and expensive process, if for no other reason to prevent people applying for - and potentially getting - trivial or ridiculous patents.
If the idea/invention is worth protecting by a patent - if it's genuinely revolutionary, or will take years of further development to bring to market - then by all means spend the time and money necessary to patent it. But anything that acts as a barrier to patenting an idea that some has whilst cleaning their teeth in the morning ("one click ordering" for instance) has got to be a good thing.
This is unfortunately how it ends up working:
bigcompany patents something obvious
smallcompany says "hey this is my patent!"
bigcompany goes "prove it"
bigcompany and smallcompany go into a legal battle over it and bigcompany drains smallcompany through endless legal wranglings.
I once advised a friend who jokingly said to me one of his IM "inventions" was patentable, and that he should go get a patent on it. I told him it shouldn't be a joke, and he should indeed patent it.
2 years later a large company came up with that exact feature, patented to themselves.
The only way to win in this patent system is to use it, currently. If you have an idea you're using, a unique one, patent it. now. then release it under license to anyone in return that they do the same with one of their patents, or patents in products that use your patent, or something.
Gnu Patent License, anyone?
In theory, the patent office does look for all prior art. In reality, they primarily look at prior patents.
Most original ideas that corporations stumble onto are never published. They are essentially trade secrets as far as the company is concerned. Besides if they publish them, others may sue them for patent or copyright infringement. Thus, they often keep quiet. However, now they are starting to patent everything in the book to at least protect themselves. It is snowballing.
I did some work for a large telecommunications company that successfully patented some automatic auditing algorithms, something that I thought was not that special, just some IF statements over statistical measurements to detect suspicious trends. It is the sort of thing that would remain internal before all this patent and counter-patent mess.
Table-ized A.I.
Before a patent is accepted, it's put up for public display, for anyone to dispute. What we really need is a user-powered web site (Another OSDN partner?) that watches patents as they are put up for review, and looks for prior art.
:) I'm sure there are a lot of out-of-work software developers who would also be interested in being hired for a job like that. Show that telecommuting can work. :)
Software patents would preferably be monitored (I can't imagine a user-based organization large enough to monitor all the different patents), but I guess other topical patents would be acceptable too.
If OSDN is interested in making that a job, they can email me.
Keep in mind that "prior art" does not include stuff written as soon as the patent is seen. It has to be prior to when the patent was filed. (IANAL, but that's what seems likely to me.)
tasks(723) drafts(105) languages(484) examples(29106)
patenting business processes is really bad.
Some crazy friend I know says that patents are good because they provide a reason to disclose new ideas. Then other people can learn from the new idea and create bigger/better ideas.
Here's an example: a construction company that patents building residential houses that have a "business area" within the home. The invention is the incorporation of another type of room within the structure to serve business uses (whatever that means).
So this one small construction company gets the business-model patent. Until the patent expires, this one business has a monopoly on building houses with a certain room. It doesn't matter that they can only construct one home at a time and are geographically isolated - no one else can build a house with a similar feature.
The USPTO is giving out market monopolies by awarding business patents. That crazy friend I mentioned earlier also told me that the US government doesn't like monopolies. Like I said, that friend is crazy.
This is not my sig.
We're looking at a future where only the very largest companies will be able to implement software, and it will technically be illegal for other people to do so.
Damned good rhetoric, Bruce, but it is too sweeping a generalization, and cannot support a call for change. Experience through 200 years of industry demonstrates that patents aren't a problem to innovation merely because they create exclusive rights in some types of software development, except when they are improvidently granted.
The problem, to me, isn't that patents are granted in software, and it isn't even problematic that bad patents are granted in software. To me, the problem -- and I believe it is a serious problem -- is that the legal system does not provide adequate quality assurance to neutralize the bad effects of that software.
It is not that the Congress hasn't tried. Relatively low-cost procedures for taking bad patents out of circulation, such as inter partes reexamination were created, but alas, with modifications that made them too expensive or too toothless to have the broad-sweeping effect that was desired. Ironically, it was large entities, such as IBM who were promoting these low-cost Q/A procedures, while the small "independent inventor" lobby fiercely opposed them.
I believe this is the area where the most change is still possible, and the biggest bang for the buck to cure the problem exists. It would be better for luminaries like Bruce to push for repairs of post-issue Q/A than to promote what is, essentially, anti-patent FUD.
We can make a difference, but we are not going to see huge changes.
...Thanks for correcting that 17 year thing :-) The new cheap method for challenging a patent is already a law. It is called inter-parter re-examination. In the sense that you can ask the USPTO to re-examine the patent, where you can submit prior art to them, and the patent filer can be the opposite party. This proceeding is in the patent office and it is more or less between the patent applicant and the USPTO, where the challenger supplies the prior art. Unlike litigation where discovery and trial consumes most money, this is a cheap and effective way to challenge a patent. Of course there are safegaurds to protect against frivolous challenges.
My point was that that patents are allowed in almost all technical areas so why not for software? There is a long-term benefit in compliation of software patent literature just as it is for other technical fields. Broadness of initial patents is just a passing phase and open-source fanatics are damaging the overall purpose of software patents which is to develop a repository of knowledge which would otherwise be locked up in the vaults of giant corporations.
Set up an anti-patent database, findable on the web through obvious domains (unpatentable.org, anti-patent.org, etc). Accept all submissions of ideas, completely keyworded and timestamped, just as searchable as the USPTO's patent database. It doesn't matter too much if the same idea is submitted twice; advise people to search and avoid that, if you want to save disk space. When you find out an idea in the database is patented, leave it in the database but add references to the patents. Maintain security of the database and accuracy of timestamps, because without credibility the database is worthless.
A centralized site like this makes it easy for anybody who wants to use these ideas to say, "Look - you can't sue me for using your patent; this idea was published to the world two months before you filed for patent." (IANAL)
As a side effect, patent examiners could use the database to find prior art, but they don't have to use this system for it to help.
Don't expect the USPTO to do the job you want them to do. You know they're broken, and they'll drag us through thousands more bad patents before they change, if they ever do. Here: the new Director of the USPTO as of 2004-01-12 touts his help getting the DMCA passed: Jon Dudas. What more do you need to know?
Fixing the broken USPTO will take time, organization, money, and expertise. Publishing anti-patents outside of any government might be the simplest way. Maybe the EFF could organize the effort.
People outside the US have different laws and jurisdictions to worry about. But this database might help in those jurisdictions too (e.g. Microsoft patenting obvious XML usage in Europe).