Defensive Software Patents for Open Source Projects? 51
jqh1 asks: "I'm the "inventor" of spamgourmet, a fairly un-obvious approach to spam prevention through auto-create/expire email addresses. I'm fully committed
to keeping the software free for all to use. Recently, I've seen other 'patent pending' disposable email address services that are not immediately forthcoming about their own processes. I don't have the time or resources to monitor and challenge patent applications, but I'm afraid that if I ignore them, I'll regret it. Should I seek patent protection for this open source software? If so, is there a way to do it without losing all semblance of an open source project?"
Licence (Score:3, Interesting)
your licence that you waive your patent rights.
IANAL, nor am I an Englishman, but I've already
thought about this.
Patents costs lots of money to keep. (Score:4, Interesting)
On the other hand, you might just as well go all the way and write a license that includes instant termination upon litigation. This is usually frowned upon in open source circles, but it is commonly used in industry to defend against suits and promote cross-licensing.
Re:Patents costs lots of money to keep. (Score:2)
Mh, to place it at a notary is what we learned in
German IT-law class in two hours about (european)
copyright law. But it's not a patent, thus the
level of protection is less.
Unluckily the teacher just left school, so I cannot
ask her for a discussion.
Good luck!
Re:Patents costs lots of money to keep. (Score:2)
> there's not much use in patenting.
The point is to establish irrefutable prior art, and place it where the examiners are likely to see it.
> Consider simply registering your method with a
> notary or similar. Then when you get challenged
> for patent infringement you can use your
> notarized testimony as prior art.
That would be completely useless. We are discussing patents, not copyrights. Prior art must be public.
How to make prior art public (Score:2)
Prior art must be public.
To make software prior art public:
You now have Pretty Good evidence that you were in possession of those bits at that time. You can back it up with a U.S. copyright registration ($30 at copyright.gov). Proof that you had actually published the package at a given time is left as an exercise for the reader (reply to this if you know how).
Re:Patents costs lots of money to keep. (Score:1)
> > notary or similar. Then when you get
> > challenged for patent infringement you can
> > use your notarized testimony as prior art.
> That would be completely useless. We are
> discussing patents, not copyrights. Prior art
>must be public.
You are not fully correct. It would be wise to have a notary certify the date of publication.
Get a patent (Score:4, Interesting)
I am not an IP lawyer, nor practicing law here, so absolutely do consult with a specialist. In the state, many lawyers participate in free or nominal fee 30-minute consultations. One hitch it that to actually obtain a patent may cost several hundred dollars (quid) for which you might solicit user donations. Set aside a little for enforcement; the GPL people have lawyers, too.
And, no, getting a patent is not crossing over to the dark side.
It's been done before (Score:2)
This doesn't count as non-enforcement - He has specifically granted the license to certain people for use in certain programs. (i.e. if you want to write a GPL program, feel free to use the patent. If you want to use it in a closed-source program, you'd better negotiate a licensing agreement to use the patent.)
Re:It's been done before (Score:2)
I have no idea of the feasibility, nor of foreign law.
I would guess, however, that the technology may be ephemeral enough that the patent is not worthwhile. Also, unlike Amazon, the poster is not in a fiercely competitive for-profit situation, and can probably outcompete on price alone over also-rans. Thus you avoid the cost and aggravation of patents and stay out of the software patenting game altogether.
I'm HOPING that anti-spam techniques in general will be rendered obsolete in the next few years by effective legislation. A dream, I know. Right now I'm getting about 70% spam, which the Apple spam filter pretty efficiently kills off. As the spam grows, so does the wait for measly few messages I care about. Sigh. (Yes, I could sign up for a filtering service, but the idea of spending money to rid myself of junk I shouldn't have to deal with at all is irritating.)
Maybe a role and source of funds for the EFF/GNU? (Score:3, Interesting)
In the corporate world, patents are usually held and owned by the company for whom the actually inventor works, not the inventor themselves. For many freelance open source developers working without the umbrella of a corporate master, why not make that company the EFF or GNU for example? The admin of the patent, and any licensing fees (only chargeable for commercial products of course), goes to the EFF/GNU, maybe with a percentage to the original innovator.
If nothing else, the sheer irony of seeing the likes of Microsoft indirectly funding the development of Linux because some spotty geek happened to invent something first would make it all worth while. ;)
Re:Maybe a role and source of funds for the EFF/GN (Score:1)
I think you misunderstand the GNU project. This idea goes against the basic ideal of the GNU project that everyone should have the same rights to use GNU software. If you descriminate against any group, including commercial entities, then the GNU project really won't have anything to do with you. GNU has nothing to do with promoting non-commercial software, and you will never see the GNU project support a license that descriminates against commercial uses.
Re:Maybe a role and source of funds for the EFF/GN (Score:1)
Nice deliberate lie.
Re:Maybe a role and source of funds for the EFF/GN (Score:1)
Re:Maybe a role and source of funds for the EFF/GN (Score:2)
The fact that a company bases it's business around the sale of software is a choice made by *that* company. That has nothing to do with discrimination.
The company is still free to sell the software (yes, they have to include source code with it), or sell support and services around it, or do a multitude of other things. In fact, the GPL explicitely grants permission to sell GPLed software.
All the GPL does is prevent a company from using something someone else wrote in a way that the original author doesn't approve of (under the terms of the GPL, which they've licensed their code under). If it wasn't under the GPL, you wouldn't (under copyright law) have a right to do much of anything with it.
The GPL is very uniform and non-discriminatory. Everyone has exactly the same rights of use under it, regardless of whether it is an individual, a corporation, or being used for non-profit, for profit, or for personal use.
You say, "a company MAY NOT distribute my software without the source". . . um, hello? If it's your software, then it's entirely up to you to determine the license it's released under. You don't have to GPL it. If you do, then yes, some other company has to release the source code if they distribute the software. . . that's the agreement they're accepting when they decide to distribute software that someone else has written.
There was no lie.
I'm not sure whether your comments were deliberately misleading, in an attempt to discredit the GPL, or simply caused by ignorance and lack of understanding about the definition of 'discriminate'.
Re:Maybe a role and source of funds for the EFF/GN (Score:1)
OK that's a funny sentence. Open Source (No Money) != Drooling Lawyers.
Publice (Score:3, Interesting)
Re:Publice (Score:1)
I heard, from said person, that once a patent has been granted, its very hard to ungrant, even with prior art. For example If I bribed a few people and patented the wheel, it would be hard work to claim that I shouldnt have been given the patent originally. Is this true?
IANA(P)L, but (Score:2, Informative)
SFAIK He's right (Score:1)
But I have to keep them non-public.
They relate to inteligent and self managing systems, which arn't quite pratical for home use at the moment (due to current PC architectures/performance &co.)
Microsoft may already hold some patents in that area with things like longhorn.
Ask Stallman what to do (Score:1)
Shared/Open patents (Score:3, Interesting)
Here's the deal,
Some OSS peeps form a Open Patent group,
The patents can be used freely for free(GPL) software or freely by a company if they submit patents to the group.
Everyone else has to pay to use the patents.
GPL is important here because a company could just ripoff some BSD software that implemented a patented algo.
Truly defensive patent (Score:2, Interesting)
Any serious use of this license should be accompanied by a written admission that the patent owner is behaving moronically.
This license does not follow copies of the binary if they are provided without full sources as stipulated by this item, even if such copies are permitted by the code license. All entities using or distributing such a release are in violation of this license
The intent here is to prevent the NOSE from using their patents offensively against open source projects. The time is limited so that the FIPF can re-assess whether the NOSE has begun to use patents against the open source community. This also allows a NOSE to negotiate a patent swap for some of their technology without giving up all patents, while the FIPF retains the right to re-evaluate whether the NOSE has begun behaving badly.
PARENT PARENT POST IS NOT TROLL (Score:1)
If a few companies hold key patents then they get togeter and agree licensing terms (often free)
Intellectual Property 101 (Score:4, Informative)
First off, we need to clarify what patents DO and DON'T DO for you.
The only thing a patent really does for you, is gives you the right to exclude others from using an invention. The way in which you can exclude them is by (civil) lawsuit (or threat of a lawsuit). There is NO reason to obtain a patent UNLESS you want to enforce it by excluding others from using an invention.
If all you want to do is ensure your/anybody's right to use an invention, you want to make sure future patents aren't granted to others - or failing that - that if they are, you have adequate proof that the invention was part of the public knowledge prior to the patent application which gives you the right to use it.
To do this, you DON'T need a patent yourself. All you need to do is "pollute the waters" by publishing prior art. In a perfect world, if a future application with similar technology crosses a patent examiner's desk - he'll find your prior art and reject the patent. Since prior art isn't always found, a patent may issue, but that prior art can be used in your defense when you are accused of infringement by others.
Publishing information for the express purpose of adding to the pool of 'prior art' is a process used by lots of companies, and is commonly known as 'technical disclosure' or 'defensive publication'.
At the risk of sounding like a corporate shill (bye bye karma) - you can read more information about how to do this [ip.com] on the website for the company I work for. (I wrote most of it, so I hate repeating myself here).
Whether you use the services (Prior Art Database [ip.com]) there or not doesn't matter to me - I don't get commission
It really sounds like technical disclosure is what you want. You may still get a cease and desist letter from somebody in the future, but you'll have evidence to back up your ability to use the invention yourself. That is all that having a patent would do for you as well. But unlike a patent, technical disclosures are much cheaper to produce, and should provide the protection you're looking for.
Re:Intellectual Property 101 (Score:2)
> publishing prior art. In a perfect world, if a
> future application with similar technology
> crosses a patent examiner's desk - he'll find
> your prior art and reject the patent.
The evidence indicates that the examiners no longer look for prior art in the literature. They probably do search for prior patents, though. Thus a patent is the best "technical disclosure".
> Since prior art isn't always found, a patent may
> issue, but that prior art can be used in your
> defense when you are accused of infringement by
> others.
How am I to come up with the tens or hundreds of thousands of dollars required to pay your firm to present that defense for me?
> But unlike a patent, technical disclosures are
> much cheaper to produce, and should provide the
> protection you're looking for.
Patents are much cheaper than lawsuits. If he patents his invention before anyone else does he is much less likely to be sued.
I think that a Free Software patent-pool of software patents dedicated to the public domain would be an _excellent_ idea. Someone has to fund it, though.
Re:Intellectual Property 101 (Score:2)
You're right in that they'll most likely find a prior patent first. That doesn't negate the fact that other forms of publishing are valid. Companies like IBM or Motorola wouldn't go to the trouble of publishing their TDBs if this were the case.
How am I to come up with the tens or hundreds of thousands of dollars required to pay your firm to present that defense for me?
If you publish in a public venue, you shouldn't need to have defense other than your legal counsel. Our pubs, for example, can be searched online (with datestamps) and the contents are also published in a paper-based journal that is sent to an increasing number of libraries (including the USPTO). You would use this in the same way you'd present evidence of an article in a magazine or trade journal.
Patents are much cheaper than lawsuits. If he patents his invention before anyone else does he is much less likely to be sued.
Patents aren't CHEAP, however. If you can prevent the patent in the first place (cheaply) all the better. There's also a good chance that waving prior art in front of somebody threatening you would shut them up. They'd rather leave you alone than risk bringing attention to the fact their patent can be busted. At least this way they can squeeze licensing fees out of others who don't know about the prior art.
think that a Free Software patent-pool of software patents dedicated to the public domain would be an _excellent_ idea. Someone has to fund it, though.
We had a joint venture with the Foresight Group a while back for such a thing. It was paid by grants from the Foresight Group and was dedicated to free publishing of technology related to open source software and nanotechnology.
NOBODY used it. It was discontinued.
Re:Intellectual Property 101 (Score:2)
> need to have defense other than your legal
> counsel.
What legal counsel? Do you think I have an IP lawyer on retainer just in case someone sues me over my Free Software? I repeat, How am I to come up with the tens or hundreds of thousands of dollars required to pay an attorney to present that defense for me?
> There's also a good chance that waving prior art
> in front of somebody threatening you would shut
> them up. They'd rather leave you alone than risk
> bringing attention to the fact their patent can
> be busted.
There's a better chance that they will call my bluff and force me to withdraw my software due to lack of money to defend myself.
> At least this way they can squeeze licensing
> fees out of others who don't know about the
> prior art.
They aren't likely to be too successful at that if they allow me to continue to distribute Free Software that implements the invention.
> We had a joint venture with the Foresight Group
> a while back for such a thing. It was paid by
> grants from the Foresight Group and was
> dedicated to free publishing of technology
> related to open source software and
> nanotechnology. NOBODY used it.
I'm suggesting sonething entirely different: a project that would identify inventions in existing Free Software and preemptively file patent applications on them.
BTW has any court ever ruled that release of Free Software source constitutes disclosure?
Re:Intellectual Property 101 (Score:2)
Unless you know the ins-and-outs of patent law yourself, you'd need to hire an attorney to do your research and filing of your patent application (if you wanted a patent). So I don't see how this is different.
Get a patent == HAVE to hire an attorney
Tech disclosure == MAY have to hire an attorney if somebody sues.
I'm suggesting sonething entirely different: a project that would identify inventions in existing Free Software and preemptively file patent applications on them.
Couldn't be done. If they've been in the PD for more than a year the patent shouldn't grant anyway. So whoever you are trying to prevent from using the innovation would be able to locate prior art to bust your patent.
Also, this "project" wouldn't be the inventor, so who would be filing for the patent? The project can't do it for itself, and if it did it for the inventor (and paid for it) the inventor could turn around and say 'screw you, thank you for paying for my patent application, but now everybody pay up.'
--
Technical disclosure is an extensively used process by MANY companies. If you don't want to use it for yourself, that's great. But don't think it's because it is some new, hairbrained idea in which the 'bugs' haven't been worked out yet.
The other advantage of disclosure over patenting is that the language you can use in a disclosure can be much broader than in a patent application, since disclosures aren't bound by the same levels of 'enablement' that must be present in a patent.
Re:Intellectual Property 101 (Score:2)
> more than a year the patent shouldn't grant
> anyway.
Of course it "shouldn't". So what? The fact is that software patents for which prior art exists are granted regularly.
> So whoever you are trying to prevent from using
> the innovation would be able to locate prior art
> to bust your patent.
You misunderstand. The whole point is to dedicate the patent to the public domain in order to make sure that no one _is_ able to prevent anyone from using the innovation. Thus busting the patent is meaningless.
> But don't think it's because it is some new,
> hairbrained idea in which the 'bugs' haven't
> been worked out yet.
I've known about technical disclosure for a long time, possibly for longer than you have. However, the recent behavior of the patent office makes me distrust it.
> The project can't do it for itself, and if it
> did it for the inventor (and paid for it) the
> inventor could turn around and say 'screw you,
> thank you for paying for my patent application, > but now everybody pay up.'
The inventor would, of course, assign the patent to the project, just as other inventors regularly assign their patents to companies.
> The other advantage of disclosure over patenting
> is that the language you can use in a disclosure
> can be much broader than in a patent
> application, since disclosures aren't bound by
> the same levels of 'enablement' that must be
> present in a patent.
Disclosure certainly would be preferable to patenting (for one thing, in the case of Free Software it could be done without the active cooperation of the inventor) IF the examiners could be trusted to find and honor the disclosure. Can you assure me that they would?
Patent vs Technical Disclosure (Score:2)
Without the patent, a company can commercialize the technology even if it is protected by the GPL. They just can't copy the code and call it theirs, but they can re-implement the technology in new code and sell it as theirs. After the patent expires, they are the same.
Re:Patent vs Technical Disclosure (Score:2)
Without the patent, a company can commercialize the technology even if it is protected by the GPL. They just can't copy the code and call it theirs, but they can re-implement the technology in new code and sell it as theirs.
You can't patent code. You CAN patent a business method (much to the dismay of MANY).
The method would remain unchanged, so they couldn't get a patent. You're right that they could write brand-spanking new code that does the exact same thing as the old code (to get around the GPL), but what does that get them? They can copyright the new code, but if it just does what the original code does, then anybody else could just use the original code and not the new proprietary code.
This would still mean that anybody would be free to implement the idea themselves freely.
As I originally stated. If your intent is to free the idea, a patent isn't necessary. If you want to restrict it (even just to some) then that's when a patent can be useful.
Re:Patent vs Technical Disclosure (Score:2)
Re:Patent vs Technical Disclosure (Score:2)
The only thing a patent really does for you... (Score:1)
Intended purposes? (Score:2)
Is there a fairly straightforward way to monitor patent applications so that the writer could bring prior art or the patent to the examiner's attention? I know many of the records are now online, and have flipped through them myself out of curiosity.
I'm new to this stuff and would welcome clarification.
Re:Intellectual Property 101 (Score:2, Interesting)
I'm reasonably familiar with IP law, and the high quality of the discussion here makes me wish I had laid out my concerns with more precision.
I think everyone has outlined the driving issues: certainly one of them is that I'm looking for the least-cost alternative that will enable the software to proliferate without interruption, and without FUD for me and for those who use it.
I'm willing to file the patent application myself (pro se), even though it may result in several iterations with the PTO due to my ignorance of the finer points of filing. Additionally, I'm really not at all concerned with the quality of the patent as regards my ability to prevent others from using the technology (because, of course, I want them to!).
Based on my own litigation experience (and fees) I think it's easy to say that, if getting the patent prevents the patent office from issuing a conflicting patent (and therefore keeps me out of court), it's bound to be cheaper. Picture yourself sitting in Texas (or the locale of your choice), and getting served for a lawsuit that is proceeding in New York or San Francisco (or the other locale of your choice). May as well get out your check book right away...
I'd be interested to know whether release on SourceForge would count as publication (and therefore defensive publication) - the software has been available there for quite awhile now. I suppose I'm the one to do the research on this one...
Thanks everyone!
Re:Intellectual Property 101 (Score:2)
> SourceForge would count as publication (and
> therefore defensive publication) - the software
> has been available there for quite awhile now.
I'd think it would convince a judge (though IANAL), but is there any chance at all that a patent examiner would see it there? They seem to do little research.
> I suppose I'm the one to do the research on this
> one...
Please post your results. I'd love to see an opinion as to whether or not release of Free Software constitutes effective prior art
OT: spambot (Score:1, Offtopic)
what if some spambot has a rule that has
s/\w+\.(\d+)\.(.\w+)@spamgourmet.com/spamena
would this mean the user is buggered?
Re: spambot (Score:1)
I have used this service for a long time (1998, I think?) and I have yet to have someone get past the "no maintenance/no-brainer mode."
Spamgourmet is the best anti-spam tool I have encountered.
Publishing beats patents (Score:2)
What actually constitutes publication is perhaps a more difficult question, and IANAL. But I would have thought that a SourceForge project would be easily enough for the courts to accept as publication.
DRM and patents (Score:2)
I'm fairly sure this was real and not part of my weird imagination.