As many of you may know, there are a number of initiatives around regarding prior art that all tackle the problem of software patents from different angles.  Whether its Open Invention Network’s  Linux Defenders, post issue P2P, or our own infant Prior Art Share project, each relies upon an underlying principle of cooperation.  The fact is that the ultimate defense – the way to eliminate a patent – is via prior art. It’s no doubt harder, but permanent, like sunlight to vampires.

Non-infringement arguments work, but only for the specific implementation. Ofcourse when you’re the defendant, you’ll gladly take either, but the real challenge is finding good prior art and developing it into admissible evidence within the time constraints of an actual patent case with a tight trial schedule. It can both invalidate the claims and/or narrow infringement arguments. Even if you can’t invalidate, prior art can establish safe zones — you can’t infringe by practicing what was “known” prior to the invention.

Notwithstanding the various projects, imagine a world where an attack on one is an attack on all, and developers across multiple FOSS communities responded to a call to action, in a coordinated and organized fashion, to find relevant non-patented prior art in response to the assertion of a patent against a FOSS project. Something like a NATO pact, but workable and without all the politics. The global hunt for prior art would happen not after the 3rd or “N” settlement, but in the first instance. In such a setting, a potential plaintiff would have to carefully evaluate the risk of asserting its patent because if found invalid, the asset would be worthless, and the licensing/royalty game would be over.  We did this once before years ago in the Wang v. Netscape patent case, and it worked. In response, developers provided a massive amount of prior art we would have never found on our own.

This theory is again in action, see Red Hat’s blog on the subject. If you want to contribute your knowledge on prior art related to the Tom Tom case (programs, documents, publications, prior to the date of the patent, that disclose the elements of the claims) they’re collecting prior art references here. Obviously, there are other long term techniques like defensive publications, advance tagging of software programs so prior art is found more easily, eliminating software patents via legislation, but in the short term, cooperation may be the most effective technique.  For those interested, the network is already in place, and if you’d like to get involved let me know.

This year the NAACP turned 100 and the challenges it faces are different both externally and internally. Like all organizations, they have to deal with the same core issue -how to stay relevant and differentiated. How does the NAACP address the problems of its constituents today?  How do they stay meaningful in people’s lives.  How does the organization attract and engage new contributors in its human rights mission? Who are its constituents?

Last fall the NAACP selected Ben Jealous as its new President and CEO.   What a good move. Ben is young, incredibly smart, trustworthy, steeped in the contributions of the NAACP, and yet at the same time, keenly aware of the organizations new role in the future. As Ben once said, [in the Brown v Board of Education era], “the fight used to be about going to the same schools, today people want to go to good schools.” I think he gets the relevance thing and the new challenges, but I suspect a 100 year old organization steers more like a tanker than a jet ski.

During the last U.S. Presidential elections, the NAACP demonstrated relevance through its voter registration efforts that used both old school get-out the vote techniques as well as viral, electronic organization methods. There are even some NAACP twitter channels which no doubt will be used to help engagement and generate heat on current events. This is a good start on the next 100.

It was even more exciting to learn that the NAACP is interested in exploring ways to use web 2.0, and open source tools and organization principles to enhance the impact of the organization. I suspect there’s a wealth of technical experience and an untapped willingness to assist out there which is available provided people connect with the purpose, and are given easy ways to engage. It will however require a new and broader concept of membership.

There are folks who’ll spend hours figuring out how to apply software to the problems, use social networks to spread information, make great contributions, and who may even remain anonymous, but they won’t go to chapter meetings or be involved in traditional ways. Morphing the organization to capture this kind of engagement will be essential to its future success, and I’m optimistic the current leadership will adapt.  For sure it will be uncomfortable and awkward at first, but the very willingness to solicit feedback and new ideas is a really good start. Merely communicating about the needs of the institution opens up the possibility for greater engagement by larger and more diverse audiences, all for the good.

On an all too frequent basis, we receive reports of websites selling the Mozilla Firefox browser, using the Mozilla trademarks to promote other products and services, or using modified versions of the Mozilla trademarks. The problem is that these activities are deceptive, harm users, cause consumer confusion, and jeopardize the identity and meaning of the Mozilla brands – not to mention being illegal. The cases seem to fall into three different categories that I’ll nominally call the good, the bad, and the ugly.  When we receive reports or identify problematic activities, we “exercise due diligence, care and prudence”  all of which means we analyze the reports and treat each case differently based on the intent and severity of the matter.

The Good. There’s a category of cases that involve good intentions but improper use of the trademarks. Typically, these folks really support the project, the brand, and the mission, and in their efforts to engage others and share their excitement about the products, they may have used the trademarks in a way that’s improper. In truth, we’re lucky to have this problem because it indicates people care about what we’re doing. This is almost always easily corrected with a phone call or short note. These matters are not troubling in my view, because they’re indicative of a user’s desire to embrace the brands in ways that are relevant and meaningful to them. Maintaining trademark protection (good trademark hygiene) and having supporters embrace the brand should not be mutually exclusive nor inconsistent. Trademark law as it exists today is technically more constraining, but I hope to see it evolve to recognize these complementary concepts. For further perspectives on trademark law, see a recent paper Tiki Dare and I wrote on this topic in the International In-house Counsel Journal.

The Bad. This category involves people who are intentionally trading on the brand for their own benefit. At the core, these cases are based on people or entities misrepresenting themselves as Mozilla. The problem may manifest itself as domain name hijacking, using the marks to promote other products, or manipulating search terms to acquire web traffic and users. Some calculations estimate that 2 to 7 million potential Firefox users per year are diverted to these fraudulent sites. This is especially offensive because these actors are trading on the value of the Firefox brand built by the community and ripping off users in the process.

The Ugly. These cases involve a clear intent to deceive, manipulate and steal from users in a highly organized and syndicated fashion. They’re a form of fraud, and frequently include other software products as well, and they seem to make a business out of charging for FOSS code and shareware. Often the identities of these infringers are intentionally hidden under layers of corporate obfuscation across multiple countries. When we can ascertain their identities, we notify them and first try to resolve it amicably. In response, we generally get the proverbial stiff arm. At times I wonder why we even bother with this step because if you’ve gone through the effort to set up an elaborate scheme to hide your identity and rip people off, why would you just stop if we say “please” and ask nicely. As an added bonus, these sites may also continue to charge the user’s credit card even after the user realizes the deception and cancels the subscription.

Many of you have sent us links about sites you suspect infringe. Thank you. At any given time there are 50 – 70 matters under investigation. Also, we now have a central place for everyone to report such sites. The more information you provide us when you file the report, the easier it is to evaluate and respond appropriately.

When we come across the bad and the ugly, and we can’t reach an accord with the person(s) responsible, we sometimes have to use legal remedies.  This may include legal action or administrative procedures where and when appropriate. For example, if a cease and desist letter does not work, we have instituted UDRP proceedings at WIPO (World Intellectual Property Organization). In some jurisdictions, we have filed and obtained preliminary injunctions to compel the infringer to stop. Surprisingly, sometimes a court order is not even enough. So far, we have been successful in the actions we’ve initiated. Recently we recovered a bunch of domains from a domain hijacker and in other actions, we obtained preliminary injunctions against web site operators engaged in fraudulent practices. In almost all of these cases, a community member reported the problem or it may even have been first reported in the media.

These actions are not only expensive and time consuming, but they divert us from our primary purpose. Unfortunately, it’s an area where I foresee continued growth and continued efforts to defend the meaning of the brands. Having the support and help of our community makes our work easier and more worthwhile.

The other night I had the chance to see Mos Def at Yoshii’s in San Francisco, an awesome venue with a great tradition of delivering seminal jazz artists.  Sadly, Mos didn’t add to that tradition. I was pretty disappointed mostly because it seemed like he mailed it in. Perhaps he was tired from a long road trip, thought he had to do the jazz thing, or he’s been doing too many movies.  Most of the show was him rambling about this or that, and occasionally a sorta song or jazzy beat that seemed more like a practice session.  The problem is that Mos is so much better than what he showed and he has an ample body of work to prove it.  The crowd came to hear that work, but alas, this was a miss. On the up side however, Yoshii’s shines even more as a venue for a broad range of artists and all types of events.

In October, we asked for feedback on creating a deeper partnership with Yandex, one in which we would make Yandex search features the default  in our Firefox Russian language builds.  Over the past few months, we have listened to feedback, talked with our localizers, studied the trends of our Firefox Yandex builds, and reviewed the Yandex user experience.  All this activity led us to the conclusion that our Russian users really wanted direct access to the Yandex search services in official Firefox RU builds.  As a result, we’re planning on setting Yandex as the default search provider for the Firefox 3.1 Russian locale builds (these changes should go into the current beta for testing). This means that, upon download and launch, the Firefox Start Page for RU locale builds will use Yandex for search queries, and the search bar will default to Yandex.  Implementation details are in bug #471561 (https://bugzilla.mozilla.org/show_bug.cgi?id=471561).

Thanks for the help.

Just a quick update on a long awaited US Federal Circuit decision on patent law issued last week. The case is called In Re Bilski (PDF), which addresses business method patents, of particular importance in the software, medical, and financial domains.  In my view, it’s positive, because it constrains (v expands) what subject matter is patentable.  At issue was whether mental processes that aren’t tied to something real could be patentable. The Federal Circuit – in an en banc decision – said no, “the claims are not directed to patent-eligible subject matter.” Claim 1 of the patent at issue is shown below:

“A method for managing the consumption risk costs of a commodity sold by a commodity provider at a fixed price comprising the steps of: (a) initiating a series of transactions between said commodity provider and consumers of said commodity wherein said consumers purchase said commodity at a fixed rate based upon historical averages, said fixed rate corresponding to a risk position of said consumer; (b) identifying market participants for said commodity having a counter-risk position to said consumers; and (c) initiating a series of transactions between said commodity provider and said market participants at a second fixed rate such that said series of market participant transactions balances the risk position of said series of consumer transactions.”

The Federal Circuit is the US appellate court responsible for patent matters in the US and sits between regional federal district courts and the Supreme Court.  The case will likely be appealed to the Supreme Court, so more to come if the Supreme Court accepts the appeal, but for now, this is a positive step by the Federal Circuit. The Federal Circuit did not go so far as to say that business method patents as a whole are invalid or address software patents as a distinct category.

Below are some additional reports and analysis that may be of interest:

http://www.groklaw.net/article.php?story=20081103134949355

http://www.bingham.com/Media.aspx?MediaID=7761

http://www.sonnenschein.com/practice_areas/ipt2/pub_detail.aspx?id=48164&type=E-Alerts

http://blogs.wsj.com/law/2008/10/30/court-reverses-position-on-business-methods-patents-in-bilski-case/?mod=googlenews_wsj

Interestingly, two of the dissenting opinions both argued that the decision stifled innovation, but for completely different reasons. Perspective is everything. Judge Mayer’s dissent argued that business method patents should be declared ineligible for patent protection because affording patent protection to business methods lacks constitutional and statutory support and hinders innovation.  Conversely, Judge Newman  argued that not protecting process inventions ignores the congressional mandate to promote useful arts and sciences. Dissents aren’t law, but they provide interesting insight into how jurists may respond to arguments in future cases.

On the downside, I don’t foresee that the decision will reduce the impact of software patents as a whole going forward. It certainly won’t make it worse, which it easily could have.  But to comply with Bilski, patent applicants need only draft their claims slightly differently (as many already do) to include processors or some general purpose computer language in their claims.  Thus, applicants will still seek and likely obtain patents on methods of manipulating and processing information, as long as they are tethered to a real machine or they arguably transform something real.

Since before Firefox 1, search has been built into the browser because it makes the user experience better.  It *also* provides economic benefits, of course, but that’s not the reason that it was included.

In most places in the world, Google is our default search engine –the one that shows up in the search box in a default installation– and our community of users and localizers generally want that to be the case. When we think about which services to build into the browser, and partnerships as a whole, the first goal is to make sure it advances the principles of the Mozilla mission (more fully described in the Mozilla Manifesto). To me this means it has to be both consistent with the values of the project and be additive, whether to the user experience or to promoting the goals of the mission.   This is the first filter and gating to moving forward whether its a formal or informal relationship.  After that, we think about distribution – getting the Mozilla goodness to more human beings.  Once those make sense, we can look for ways the partnership can contribute to the sustainability of the project in a manner that’s consistent with our values.

In Russia, there’s an interesting situation, though: our localizers have been suggesting Yandex as the default search provider for some time. See bug 348096.  The key user benefit offered by Yandex for Russian language users is that they have focused on linguistic characteristics of Cyrillic and on Russian content to create relevant search results. The Russian user base obviously likes them a lot as well.  LiveInternet reports Yandex as having 55% of the Russian search market. Yandex also shares the first and second place for the most trafficked web property in Russia.

In June, they built a “Yandex-flavored” version of Firefox that they distribute themselves, and it’s had a big impact. Since then, users of the Yandex version of Firefox have grown to 300K+.  To date, this represents approximately 1/3 of the total Firefox Russian language user base.

This presents a unique opportunity. It’s one of the rare settings in which an indigenous search engine tuned to a particular language has turned out to be wildly popular, and where the local Mozilla community has consistently made strong, data-based recommendations for a solution unique to the Russian language.  The popularity of Yandex suggests that a number of Russian speakers may be eager to see Yandex as the default.  (Other search engines will of course be provided, as has always been the case with Firefox.)

Given the input of the localization teams, demonstrated user preference, and the existing base of Firefox users, we’re evaluating making Yandex the default search provider for Russian language builds. We’re very interested in broader community feedback to ensure we’ve considered all of the issues.

We’re editing the language and working in the comments as appropriate.  Although there were a number of questions, not all addressed in this post, one comment asked how this proposal differs from the Fedora implementation. In its fundamental concept, it doesn’t really differ. We used some different words, referenced the MPL specifically, and took out as much extraneous material we could (material originated by Mozilla not Fedora). It was also suggested that we work in a larger open discussion with all the stakeholders who are working towards integrating web services into the linux desktop… This is an excellent suggestion which we should do.

Below is the sample notice page. The reference to the MPL should link to the document.

Below is the web services language which will open from the notice page above.

We’ve received a lot of feedback on the licensing proposal. The commentary overwhelmingly indicated the proposed approach wasn’t good enough (that would be an understatement). We looked at it again, incorporated suggestions from the community at large and from some of the Linux distributors. The new design addresses both presentation and content. We’re still tweaking the language a bit and working on the implementation, but directionally this is where we’re going.

Presentation: There’s no click-through, or license splashed in the users face on start-up (or at any point thereafter). We’ll either include some text on the first-run page or in an info box that links to a static page in the browser that contains a notice about your rights. We’re still working through which implementation works best – so this isn’t final.  Some examples below:

Content:  There is no EULA. There are no caps except where grammatically required.  There is a notice page that points to the MPL, provides summary information on the rights that come with it, includes a statement about trademarks, and a statement about optional web services (like safe-browsing) that are not covered under the MPL.  The notice includes a link to the terms related to the services. Screenshots and text will be coming in the next hour or so.

We’ll likely see the first implementation of this in the Ubuntu builds, and depending on shipping schedules we see how this rolls out across other distributions/platforms and in our packages.

Obviously, we’ve been working with Canonical to get this worked out; however, the Red Hat and Fedora folks have also provided invaluable guidance.