Grokking Peer to Patent

The following are comments from Pamela Jones, founder and editor of the award-winning website, Groklaw, and one of the contributing authors of the book, Open Sources: Voices from the Open Source Revolution.  Jones is a leading voice in the Free and Open Source Software (FOSS) community.  This email exchange took place during April 2006.

PJ: I saw the discussion you posted between you and Larry Sanger. I sensed what seemed to me to be animosity toward the community's participation in this project. As for his worries about a lack of legal expertise and the suggestion that folks post their resumes, in effect, such as how many patents they have, they reflect a lack of understanding of the FOSS community. It's patent lawyers and proponents that got us into this mess in the first place, is it not? How will they fix it, then? Let's get down to core principles here. You can't fix a problem by doing more of what you did to get in the pickle you're in.

BSN: For what it's worth, nothing in the current design will involve rating people based on their "objective expertise," as he puts it. By now, we all should know that our mothers, with or without 4 PhDs, are the smartest people we know ;-) Expertise is not conferred by a degree.

There can be various strategies. While there are those who are working on getting Congress (or the EU) to ban software patents or change the rules, I remain convinced that we need to work, in parallel, to reform the institutions themselves and the way they work. The patent examination system is constructed around false and outdated assumptions about expertise that must also be addressed.

PJ: Well, technical expertise is not in evidence at the USPTO. That's the problem. Who can qualify for USPTO examinership needs to be addressed. If you allow software patents, what could the rationale be to exclude whose who have software technical experience? An entire body of prior art is not being considered, and Sanger's remarks seemed to me to be about making sure it stayed that way, although I don't think he knew he was saying that. You just shift the battlefield, not that it's not worth doing even that, but with all the same people and attitudes. Anyway, they don't know the FOSS prior art. If they did, the USPTO wouldn't have issued all the stupid patents they've already issued, like the FAT patent. When it comes to software, the community knows where the bodies are buried. I found that out doing Groklaw and Grokline. Just let them loose, and they'll find whatever is needed and available to overturn every stupid patent in that area. My question is: do the usual players really want that?

BSN: You make two comments here which I hope to address in turn. The goal of this project is not to create a repository of all art but to create a system to comment on and protest specific patents and turn up the art to attack those patents specifically. There are strong incentives for the patentee to participate in this system, namely, that a patent that survives peer review will be a stronger, better, more qualitative patent with increased potential for licensing revenue. There are many who will want that and also plenty of usual-suspects who don't want any change at all. But only by doing a pilot that empirically demonstrates value from peer review will we ever be able to change the law to make peer review and greater openness and transparency mandatory. The Patent Office will work with us to create incentives for participation, including jumping applications that consent to peer review to the front of the line, waiving fees and other incentives.

PJ: Let me express what I think some in the community will wish said: if the USPTO's goal is to issue patents, and the community is used to help patent trolls figure out how to write a better patent, please count them out. The problem is, the USPTO seems to measure success by how many patents they issue. I have yet to see a word from the USPTO that they'd like to issue *fewer* but better patents. How is that problem addressed by your project? I am sure peer review can work, but I don't think it will unless someone notices and takes seriously that the goal post needs to be moved to get the community involved.

A lack of legal knowledge is a fixable problem when it comes to what constitutes prior art. The community can and will learn if we can solve issue number one, above. A lack of technical expertise can be fixed too, primarily by allowing the FOSS community to do what it does best, which is tell us where the prior art is. They know. Most of them, however, are opposed to software patents at all (why shouldn't they be, when Microsoft keeps threatening to destroy Linux with patents? -- anyway, individual developers can't afford to play the Rich Boys' Patents Game and they know it), and if that becomes a negative disqualifying factor, then you lose all that expertise, which is, after all, the missing piece that could prevent stupid patents from issuing in the first place.

That is the goal, no?

Groklaw is here. I'm happy to publish any teaching articles lawyers would like to write to educate folks, including what the answer is to number one, above. Without that education, the community won't have the ability to really help. They won't know what constitutes prior art, for example. Actually, a lawyer is working on an article about that now for Groklaw. But there are many other things that need to be explained.

Second, what exactly should the community be doing? There is a blank there. No one exactly knows what they are supposed to do, even if they wanted to help, which a few are willing to do, but most won't unless they see there is an answer to the problem they perceive of their work being used by proprietary entities to write more effective patents to hurt FOSS with. Better communication is important. But there needs to be some organization.

BN: On May 12, the USPTO will announce plans to launch the project and put an implementation committee in place. Over the summer, they will be working on designing the system, based on the guidance given from the project workshops, and will aim to have it up and running by January. We need to get people involved and interested in the project and to get their feedback on the design. I will give you the draft design to circulate for input and feedback from the community. It will also go up on this site and on the new listserv.

Second, we need to brainstorm how to get people to participate. I agree that there is no concrete way to participate yet but that will change will the launch of the project and the official announcement by the USPTO that it will accept peer review of applications.

While I am the idiot who is, as you have suggested in earlier correspondence is necessary, "willing to answer every question and every email" to get this up and running and find out whether it work, this needs to be “our” project, a community initiative built on the solid base of people.  I am a confirmed believer in the ability to redesign our institutions using technology to deepen democratic practice, promote collaborative decisionmaking and improve the legitimacy of our political institutions. Designed right, we have an opportunity, working together to test whether we can make a chink in the seemingly impenetrable patent process and its wrongheaded notions

PJ: The main thing, other than the above, is this: how does the community find out about patents that they might care to try to work against?

I might point you to this page: http://www.mozilla.org/legal/wangsuit.html

When Wang tried to assert patent claims against Netscape, there was a call to the community to help find prior art.

They did. http://www.mozilla.org/legal/wang-dismissed.html

One reason they were able to find prior art is because of the way someone broke down all the individual claims of the patent at issue, as you can see on the second URL, above, so folks knew what to search for. The patent was explained and some suggested prior art was mentioned, the exact date prior art needed to be found prior to, and where to hunt for it. It's an example of community involvement that worked.

The FOSS community believes one reason it worked is because they knew of prior art Wang didn't know about. But I am using the example to say that to be effective, someone needs to break down the patents to say: here's what would be needed in the way of prior art to beat this patent.

I don't think RSS feeds of patents themselves will work, without that kind of explanation.

BSN: Here are some of the contemplated innovations designed to make participation easy, relevant, manageable and, hopefully, fun.  Really good RSS-feeds based, not only on the patent classification system but on community-generated folksonomy. Easy mechanisms to share and circulate notices of published patents received. Key to this are visualization tools to show the number of patents within each class and subclass to avoid under- or oversubscribing leading to a participant being overwhelmed. Sophisticated searching, including by assignee (e.g. Microsoft) and by other categories, much of which is available now but not combined with visuazliation. In addition to the subscription mechanisms, the interfaces for uploading prior art will be optimized to target participation. As you mention, prior art submitted will be tied to particular claims so that it is manageable, targeted and relevant.

PJ: So, that's all I can think of. I'd like to help any way I can. I seriously believe in what you're trying to do.

The part I'd like to add is that the FOSS community has expressed to me concerns about any prior art initiative. I get a lot of email and the worry is that by pulling up and organizing all the FOSS prior art, all that will be accomplished will be that the Microsofts of this world (and the patent trolls) will then really know how to write a killer patent, that they'll use FOSS prior art to formulate better, more effective patents to kill off/wound Linux and FOSS. Richard Stallman belongs to this camp, last we communicated. If they believed stupid patents were really going to stop being issued, they would help, but they don't think that the USPTO cares about obviousness at all. To get your project and OSDL's really to work without the FOSS community...well, I don't see it working too well. It's the largest group of folks with a high motivation to invalidate stupid patents. But they won't do it if they think they are enabling the enemies of FOSS.

I don't share the view that it's better to hide things inside the community vest, although I had to think a long time before I could decide where to plant my flag. We won't know until we see it play out for sure, but I assume, frankly, that Microsoft has both the will and sufficient means to find any FOSS prior art without any input from the community. So, from my standpoint, having in place some threatening muscle to *discourage* patent infringement lawsuits is a good thing. That's my analysis. So the above description is not my personal view.

I note that in the proprietary space, they have patent cross licensing and a Cold War style of prevention of too much litigation. FOSS was not part of that Rich Boys' Patent Club, and mostly they can't afford to play. I see the various initiatives as a way to protect FOSS development by forcing the proprietary wolves to at least think twice. Avoiding patent litigation in the first place is better than winning at trial.

BSN: I agree with your assessment. Piloting peer review on software-related patents runs the risk that the FOSS community senses itself to be divided. That fear, I would argue, is misplaced. There are those who oppose involvement with the Patent Office and with software patents, which they want to abolish altogether, and those who favor incremental reform, if it comes along with meaningful commitments from the Patent Office and the opportunity to defeat non-meritorious patents. The “Peer to Patent” project does not aim to gather all prior art into a repository which can be exploited, as you suggest, but to get at prior art relevant to narrowing or defeating specific patent claims. There is an urgent need to test peer review empirically and find out if doing it right can lead to the rejection or narrowing of bad patents, greater informational transparency, the creation of communities of innovative practice and, finally, the passage of only the best and most deserving patent applications. This is a design for a practical new institution to evaluate innovation. My hope is that the FOSS community will want to participate in this targeted initiative, which will directly aim to knock out bad claims. Also it is my belief that, to do this right, requires testing on a wider variety of patents, not just software, but also in communities that have a strong culture of voluntarism. We need to get as many people involved as possible to find out if this works.

Comments

[P.L.Hayes]

BSN: "There is an urgent need to test peer review empirically".

%-}

If this experiment achieves nothing other than that it exposes patent system administrators and policy makers to the ideas of rational and empirical science and consequently inspires them to learn the basic principles of patent system economics, study the literature and perhaps one day even apply those principles, it will have been be a very worthwhile project.