FOSS Community Can Combat Bad Patents

An anonymous reader lets us know about a new initiative designed to help shield the open source software community from threats posed by patent trolls. The initiative, called Linux Defenders (the website is slated to go live tomorrow, Dec. 9), is sponsored by a consortium of technology companies including IBM. "The most novel feature of the new program... will be its call to independent open source software developers all over the world to start submitting their new software inventions to Linux Defenders... so that the group's attorneys and engineers can, for no charge, help shape, structure, and document the invention in the form of a 'defensive publication.' Linux Defenders will then also see to it that the publication, duly attributing authorship of the invention to the developer who submitted it, is filed on the IP.com Web site, a database used by the US Patent and Trademark Office and other patent examiners throughout the world when they are trying to determine whether a proposed patent is truly novel..."

Comments

Does this count as prior art?

[Raul654]

When a patent is under examination, others can submit examples of prior art. However, if they do and the patent is subsequently granted, the patent cannot be challenged in court on the basis of that prior art. That's why nobody does it -- they assume the examiners are idiots and prefer to take their shot in court instead. So - does this website count the same way? If so, it might not be such a great idea...

Re:Does this count as prior art?

[theoddball]

If the article is accurate, then no, this is not about examination.

The point of this initiative is to get information about innovations that exist in the wild and document it ("publish" it for statutory purposes) -- and make it easy for examiners to locate it if they want to.

This is about creating prior art that will serve as a bar under 35 USC 102(a). Examiners may find it, or they may not, but if nothing else, the publication will be sitting there for someone to assert in litigation / reexamination -- and quite possibly able to invalidate a bad patent.

Re:Does this count as prior art?

[kenshin357]

Where are you getting these restrictions on prior art from?

There may be a rebuttable presumption of validity if you want to apply the same piece of prior art in the same way as it was used earlier on in the prosecution history, but there is nothing preventing you from using it court. Think about it: it would be fundamentally unfair to not allow someone to use something as evidence in court simply because it was brought up in an earlier ex parte proceeding.

As for the "nobody does it"--do you know the different in cost difference between a third party submission versus an infringement suit in court? We are talking thousands versus millions of dollars here.

There are certainly plenty of reasons why someone would prefer to take an issue to the courts, but sitting on a clear, obvious piece of prior art is not one of them. Of course, patent trolls are the exception to this rule.

Re:Does this count as prior art?

[Elektroschock]

There is only one efficient initiative and it is legislative reforms.

I would recommend to donate to the FFII.

Re:Does this count as prior art?

[Anonymous Coward]

No, in litigation, the court is supposed to review all relevant prior art, including that which the Examiner already reviewed. The court will learn why the examiner thought the reference didn't make the invention unpatentable, but have you ever read something written by a patent examiner? Maybe it's just my field of electronics, but if their argument is anything more than incomprehensible gibberish, it's almost always a ridiculously stupid and irrelevant argument. The court certainly has the power to rule the other way.

Re:Does this count as prior art?

[jvkjvk]

I don't know about that.

How "efficient" has legislative reform been so far? How close are we to invalidating business method patents, or software algo patents?

Efficiency means that in the real world something gets done with the near minimal energy required to produce some desired output range. Since patent laws have not been reformed, despite the tremendous energy that has been put forth already to do so, I question the efficiency of the approach.

I do not question that it is a "better" method than nibbling from the bottom in terms of overall wholesale effect, should it ever be accomplished. But, in terms of actual accomplishment it seems to be much less efficient.

Re:Does this count as prior art?

[Elektroschock]

Efficiency means that you invest and have real impact and direct your investments to the archilles heel.

"Since patent laws have not been reformed, despite the tremendous energy that has been put forth already to do so, I question the efficiency of the approach."

This is true for the US, basically because the analysis of the patent critics is premature. The patent system is technocratic and shielded against intervention. If you get your community trapped in the most trivial defense nets, no progress can be achieved.

Think of the patent community. They have about seven important conferences per year. Be there and influence the debate.

Educate the software community how to crack the system, and do not attempt to convince developers with arguments made up for them because they are not the target.

Trade agreements, a thing to watch.

Legislation, educate your legislators and keep reform talks.

Why does it have be just Linux?

[actionbastard]

Why can't this be a defender of all FOSS? To be just a defender of just FOSS OS limits its ability to represent anyone who wishes to create free software.

Re:Why does it have be just Linux?

[Anonymous Coward]

The same reason a single hospital cannot cure everyone, limited resources.

Sounds Doubtful

[zach297]

This sounds to good to be true. I could be of base on this but is it possible that these Linux "Defenders" may be trying to influence open-source software to their own benefit.

Re:Sounds Doubtful

[Tatsh]

I agree. I like the idea, but this sounds dangerous, and this is NOT because I see nothing related to FSF or other already existing groups. Why should anyone just trust these people?

Re:Sounds Doubtful

[buchner.johannes]

So you have no problem releasing your software to the whole world, but this group is not allowed to see the idea you manifested in the code?

Re:Sounds Doubtful

[zach297]

"so that the group's attorneys and engineers can, for no charge, help shape, structure, and document the invention"
Its the help shape part that is concerning. Free software must be just that, it can not be shaped by anything but the community.

I could be mistaken here

[Aussenseiter]

But isn't the goal of patent trolling to win settlements from suing large, established corporations? Why would they go after the open source community, which by its nature tends to not be swimming in green papery things?

Re:I could be mistaken here

[Miseph]

Because this isn't intended to fight that form of patent trolling. This is intended to combat the type of patent trolling where large established companies file frivolous patent infringement lawsuits against small rivals in order to stifle competition and ensure profitability.

Re:I could be mistaken here

[GMonkeyLouie]

So I guess it's a way of saying that you're not confident you could pay to win a legal battle against a large established company who goes patent trolling?

Nice. Now being inventive is effectively taxed by the presence of bigger fish. Way to go, patent law.

Re:I could be mistaken here

[Anonymous Coward]

Micro$oft

Re:I could be mistaken here

[MaskedSlacker]

Um, welcome to decades ago? This shit isn't exactly new.

Re:I could be mistaken here

[Tatsh]

Microsoft is behind this one. Yup. They think they can act. // End sarcasm

However, I will not allow that idea to escape me.

Re:I could be mistaken here

[Anonymous Coward]

Linux kernel violates 235 of Micro$oft patents. It's a fact.

http://www.google.com/search?q=linux+violates+patents

Re:Hopefully this works ...

[Anonymous Coward]

Hey, I patented posting anonymously. You owe me $500 in damages!

Yeah but

[Orion Blastar]

can the FOSS Community Combat Bad Parents?

FOSS Community Can Combat Bad Patents?

What is a Bad Patent anyway? Slip the clerk a $100 bill to change the date stamp to the 1980's and claim everyone else stole from your patent?

I would rather think it would be something like trying to patent a FOSS name like "Linux" as a "trademark" and not a patent.

The cost of filing a patent is expensive, so when a FOSS invents something, someone else patents their work as their own, now that would be a bad patent. Can some FOSS group help FOSS afford patent fees to avoid others filing a patent for their work?

Re:Yeah but

[Tubal-Cain]

What is a Bad Patent anyway?

A patent that should not have been issued in the first place.

For example, patents are supposed to be for unique ideas. Prior art (someone else built something similar first) proves that the idea isn't so unique. But the patent might be granted anyways, for some reason.

Re:Yeah but

[Anonymous Coward]

> I would rather think it would be something like trying to patent a FOSS name like "Linux" as a "trademark" and not a patent.

Errrr ... the name "Linux" is a trademark. The trademark belongs to an individual called Linux Torvalds.

http://www.linuxmark.org/

If you were to fork the codebase, change it without going through kernel.org , and you then decided to distribute it, then according to copyright law you can only do so with permissions given to you by the copyright holders.

The copyright holders have set out their terms for granting such permission in a license document called GPL v2. The terms for what you were trying to do include a condition such that when you redistribute it you must offer the source code to anyone who asks you for it.

Trademark law says that you may not call your product "Linux" without Linus' permission. You may call it something else entirely and distribute it without Linus' permission, but if you want to call it Linux then you must have Linus' permission.

Re:Yeah but

[Elektroschock]

A bad software patent is a patent on software.

The way to solve the problem is reform of substantive patent law. In other words, donate to the FFII

Trademarks are different from patents.

Something similar

[Mr. Sketch]

How is this different from the USPTO Peer-to-Patent initiative?

Re:Something similar

[Elektroschock]

The peer2patent initiative is run by the USPTO and sponsored by IBM.

Re:Something similar

[JLF65]

Peer-to-peer is reactive - you help fight one specific patent under review. This new initiative is PROACTIVE - you help provide info that could keep many patents from being issued in the first place.

Permalink

[kabloom]

Please use the permenant link to the article http://legalpad.blogs.fortune.cnn.com/2008/12/08/a-no-fly-zone-to-protect-linux-from-patent-trolls/ not the link to the front page of the blog.

hell no

[larry bagina]

IBM is one of the worst offenders when it comes to filing stupid patents. Yes, they've donated patent for Open Source usage, but they're just as snakey as amazon (remember their bounty program, where they collected a-click patent prior art from the community and then used it to strengthen their claim that 1-click was innovative?)

Fuck IBM.

open patents solutions.

[AeiwiMaster]

This is funny, only a few days ago I was talking about with some friends about the patent problem.

We got the idea that the solution was to make
public patents license which does for patent-rights as the gpl does for copyrights.

We made a facebook group to devlop our idea
you may join it if you like.
http://www.facebook.com/home.php#/group.php?gid=39345211199

Re:open patents solutions.

[setagllib]

It's really funny you should say that and not recognize what the GPLv3 does in this area. It still hinges on a company actually distributing GPLv3 code, after which they cannot assert patent claims against the users they just gave the code.

It also helps null the Novell-Microsoft deal by making any GPLv3 software Novell or Microsoft distribute perfectly free to use by any other distribution as well. The partnership can't use the work of FOSS developers against them any more.

Re:open patents solutions.

[AeiwiMaster]

No, the PPL is for every type of patentable technology not only software.

It will be designed to do the same for inventions as the GPL did for software.

Re:open patents solutions.

[Anonymous Coward]

In the United States, you are not required to register with the Copyright office to secure the rights to your work, although it can help. You hold the rights upon completing the work. You can distribute work to which you own copyright while attaching conditions to its reproduction. The theory behind many open source licenses is that if a company decides to use software distributed under such a license, it can be required to stop.

To assert patent rights in the United States, you must have a patent issued by the United States Patent and Trademark Office. The ONLY legal mechanism to prevent private parties with whom you have no other contractual obligation from making, using, and selling your invention without your authorization is to obtain patent from the USPTO.

The best you can do in the public domain is disseminate defensive publications; that is, you can provide prior art so that companies cannot patent your invention. But if you do not also get a patent on the invention, those same companies are still free to use your invention without attribution.

Corporations can have binding contracts with one another that allow them to cross-license patents. Unless someone in the "open patents solution" is willing to foot the bill to file for and prosecute patents, and then provide a limited license under which they may be made, used, and sold, this idea cannot get traction.

Re:open patents solutions.

[setagllib]

Very well, since you clearly have legal expertise I won't dismiss your ideas, but take care not to confuse patent law with copyright law. Perhaps there's a good reason there's no such general patent license.

Re:open patents solutions.

[Anonymous Coward]

The problem with software patents is that they give far more privilege than traditional patents. The standard example used to explain patents is the better mousetrap. And over the years, there have been some 4,000 mousetrap patents issued (Google it). If the level of protection given to software patents were given to traditional patents, there would have been one patent covering the concept of mousetraps!

I actually tested the patent system in one of those companies that tries to attract inventors. I came up with a simple invention (swingset anchors to keep the legs on the ground while kids are swinging). I had to draw a schematic of how the invention would work, which included a parts list. This limited my invention to a very specific product.

To bring software patents in line with traditional patents, they should require a prototype which would limit the patent to a specific product. Of course, this would make software patents very expensive copyrights, but my reading of the history of patents is that is really all they are.

Later . . . Jim

a GPL for patent rights.

[AeiwiMaster]

Why limit the action to open source software ?
Why not make viral license for patents
like the GPL for software.

This question did me and my friends discuss
a few days ago and we made a facebook group for the perpose.

join us at
http://www.facebook.com/group.php?gid=39345211199

BountyQuest Redux?

[jambarama]

If bountyquest couldn't get enough high quality prior art submissions by offering 10k, what makes IBM think they'll get better submissions without offering anything? Bountyquest's failure to bust the 1-click patent was quite telling of its patent-busting power. Salon.com's postmortem explained "BountyQuest tried to overcome the inability to build momentum [from a few big patent busts] by cold-calling patent lawyers and trying to sell them on the idea of running a contest for one of their cases [b]ut few have proved willing to bite." "There just didn't appear to be a market for its service." O'Reilly (one of the sponsors of the project) said in his postmortem of the project, "the patent mess is a thorny thicket that doesn't lend itself well to penetration by amateurs."

Apart from paying less to hunters and charging less to clients, how is IBM addressing these problems?

Re:BountyQuest Redux?

[sir_eccles]

The problem with all these services and all discussions about patents on slashdot is that finding prior art is not as easy as it looks. Amongst other things it has to be published, it has to be published at the right time and here is the bit where most prior art falls down it has to be enabling.

To be enabling, the prior art has to teach the same feature/ method/ process/ or whatever as stated in the claims and in light of the description. It is no good looking at the title and making assumptions about how the application does it and then saying you did it years ago.

So while documenting how Linux does stuff is useful, it does not guarantee anything.

Bad parents?

[Z80xxc!]

Somehow I read that as "FOSS Community Can Combat Bat Parents". I was really excited, too. Oh well... been a bad past few days I guess.

Rule #1 : vendors lending legal support

[NZheretic]

As I stated over two years ago

1) Any patent lawsuit against a user of a software component used by major vendors will automatically result in those vendors lending legal support to reduce the chance that their own customers will also end up being sued.
2) Any patent lawsuit costs the suing party at least several hundred thousand dollars.
3) Any patent put before the courts is at very great risk of being destroyed by prior art.
4) Any payout awarded from a single end user has to be in proportion to value of the patented technology. The value of a single instance will could only be measured in hundreds of dollars, not coming close to covering the costs of the lawsuit to the plaintiff.
5) Patent lawsuits take six years to over a decade to work it's way though appeals.
6) Developers will release new software using a method that circumvents the patent in question within two months. This will be quickly adopted and by the time the first patent case is resolved there will be no further customers for the patent holder to sue.
7) The outrage generated in taking out a case against any open source will result in Groklaw [groklaw.net] and other groups putting the suing party and their lawyers under the closest scrutiny. You will not believe the level of bad publicity, let alone the the amount of prior art, dirty business practices, and legal suspect practices and even violation of statutes that will be uncovered.

Lastly to quote Pulp Fiction, and then "we are going to get medieval on your ass."

Any IP case against users of open source puts the attacker at a far greater risk.

What now? Let me tell you what now. IBM will call a coupla hard, case-hittin' lawyers, who'll go to work on the trolls here with a combination of prior art and counter suits.

IBM: the Linux Knight

[reporter]

As others have noted, companies have only a finite amount of money, so they must be judicious in funding efforts to defend the open-source community from patent trolls. Doubtless, the most valuable open-source application is Linux. It is an operating system and is the heart of a computer. Linux is the #1 application on which most programmers develop open-source programs. Typically, open-source work is done first on Linux and, only later, is migrated to Windows.

So, Linux deserves most of the resources for defending against patent trolls.

Here, again, we see the steady hand of IBM. It has been the central company in bringing Linux into the mainstream.

Before IBM announced its decision to deploy Linux on IBM servers, Linux was just a niche application used by brilliant scientists and engineers. In 2000, IBM changed the computing landscape by officially supporting Linux and deploying it across its entire range of servers. If IBM backed Linux, then the business community would try it. IBM was willing to sign contracts guaranteeing the reliability of Linux.

Linux is now the only Unix-ish operating system in many financial and engineering firms.

IP.com ?

[mmu_man]

I thought it would be about Internet Protocol or something else really useful. Bloody domain squatters!

Re:IP.com ?

[mmu_man]

Don't they know about google and search engines anyway ?

Bad start

[Anonymous Coward]

Facebook Login
You must log in to see this page.

Maybe that wasn't the best place for open discussion on a topic designed to fight proprietary strangeholds.

IBM?

[Anonymous Coward]

Getting IBM to help you defend your patent is like getting a wolf help you defend your chickens....

Re:IBM: the Linux Knight

[Anonymous Coward]

Linux is now the only Unix-ish operating system in many financial and engineering firms.

I don't think "many" is a word for the financial and engineering firms.

I work for a large Avionics company that employs over 20,000 people. Currently Linux is used for some engineering desktops and some compute farm clusters. Alot of development relies on Solaris of all different versions. Most of the servers are either Solaris or HP-UX. All of the financial stuff, the most critical, is run on AIX.

Large corporations and firms rely on pay for support from vendors for their hardware and operating systems. Red Hat's support for example is atrocious at best. They constantly release kernels into the RHEL 5 tree that break many components causing loss of local data, and in some cases total periphial function lost.

I certianly don't have anything against Linux, just saying you may be overstretching with that last statement.

It's a trap!

[WiglyWorm]

IBM.
Patent Reform?

I find it more likely they're going to exam your work, patent it, then sue you for infringment.

Awww

[Anonymous Coward]

Looks like the FOSSies are good for something finally.

The irony of IBM being involved.

[geekmux]

Is it just me, or does anyone else find it rather ironic that IBM, holder of the how-to-split-your-dinner-bill patent, is co-sponsoring a project to protect us from bad patents?

Hey Pot, the Kettle called while you were out. He left a message mumbling something about WTF...

Re:IBM: the Linux Knight

[orasio]

Linux is not an operating system.
GNU is an operating system, and Linux is its most used kernel.
Most development targets GNU, specifically glibc and stuff.
Without Linux, we would have other kernels, and porting would not even be necessary in some cases.
Without GNU, everything would have to be rewritten, at least, for BSD.

Wrong solution

[Trails]

Well intentioned, but this solution is fundamentally wrong.

The only thing that can really fix the problem is a modification to legislation surrounding patents, as well as clearer guidelines to patent examiners and stronger/any penalties for trolls. The above proposed puts the onus on OSS to file tonnes of stuff that should be considered obvious, simply to avoid predatory patent lawyers. In other words, innovation is still hampered.

Re:Hopefully this works ...

[Hurricane78]

I patentey owing! You owe me $500 for every dollar you owe anyone!

Including me!

MUHAHAHAHAHAAA *slashes AC with tentacles*