JMRI Defense: Fighting to Keep an Open-Source Project Alive

Donate to JMRI This page is a brief historical summary of the entire JMRI-Katzer legal dispute. The entire history, event-by-event, is available as a single large page or as series of individual pages. For information on the February 17, 2010 settlement agreement that hopefully brought this to a successful end, please go here. For some of the press and community reaction, please see the Articles by Others page. If you'd like to read all the legal filings, we've got them all available on a separate page.


JMRI is an open-source project for developers to write software for model railroads. It has achieved wide-spread acceptance, with more than three hundred people writing parts, and users on every continent except Antarctica (and we're working on that one!)

As an open-source project, JMRI software is distributed free of charge. Our licensing terms allow people to redistribute the software, provided they abide by certain simple restrictions. It's not been hard for anyone to understand and live up to those - except one person.

Starting in 2004, Matt Katzer and his company KAMIND used various coercive tactics, some of which we believe are illegal, in an attempt to put a stop to JMRI's work or to extract money from JMRI.

Katzer, through his attorney Kevin Russell, obtained a patent on model railroad technology that other people had developed years before. Using a "continuation" application, they applied for a patent that covered JMRI after JMRI had openly published its code. Because Katzer and Russell didn't describe the prior inventor's work ("provide prior art") to the Patent Office, the patent was promptly issued.

After the patent issued, Katzer and Russell first demanded $19, and later $29 in royalties per downloaded copy of JMRI. We believed that the patent was invalid because of the prior art, and if they had provided the prior art in their possession to the patent examiner, that it would not have been granted in the first place. We told them that (see the page containing this correspondence) Never-the-less, Russell sent Bob Jacobsen, one of the JMRI developers, bills for more than $200,000 on a roughly monthly basis from August 2005 to January 2006.

When that harassment wasn't enough to stop the project, Katzer and Russell contacted Bob's employer and accussed them of patent infringement in an attempt to intimidate Bob.

In March 2006, we couldn't allow this to go on, and filed a "Declaratory Judgment" action in Federal Court to get a legal determination on our rights.

While we sought to hold Katzer and Russell responsible for certain of their actions, such as contacting Bob's employer about the patents, the Court ruled that Katzer and Russell cannot be held responsible for them. Those were removed from the complaint.

On the other hand, we found compelling evidence that Katzer took JMRI decoder definition files and included them in his products, thereby thereby claiming them as his own, in violation of the JMRI license. We've therefore added copyright claims to the case by filing an amended complaint and asked for a preliminary injunction to stop Katzer's misuse of our work. In a motion to dismiss, Katzer argued that we can't use our copyrights to hold him responsible for his violations of the JMRI open-source license. We replied to show that open-source licenses are valid and enforceable. The District Court ruled that Katzer was right, and that we were not likely to succeed on the merits of our copyright claim. We disagreed emphatically with this, for multiple reasons, and appealed the ruling. Our first brief, presented our side of the issue. A strong consortium of free and open source advocates, including Creative Commons Corp, The Linux Foundation, The Open Source Initiative, The Software Freedom Law Center, Yet Another Society (Perl Foundation), and The Wikimedia Foundation, filed a "friend of the Court" brief supporting out position.

On August 13, 2008, the Appeals Court ruled entirely in our favor, vacated the District Court's decision, and remanded the case for further proceedings.

In the Fall of 2008, there was another round of motions. After a hearing on December 19th, the Court issued a mixed ruling. We still didn't get an injunction, but the Court did require Katzer to finally (after years!) answer our complaint. That answer came with a Katzer's counterclaim for $6 million, for using phrases from QSI manuals to make the decoder definitions Katzer copied! Katzer had bought the rights to a QSI manual from QSI for $500 so he could do this. Clearly, this is just another of Katzer's depradations that we had to fight.

In addtion to all that, Katzer also improperly registered one of our trademarks, DecoderPro, as the domain name "". The World Intellectual Property Organization (WIPO) has ordered it returned to us, and wrote an entire section of the order on "Katzer's bad faith". We think it's important to have the Court order Katzer not to repeat this type of misbehavior against JMRI or anybody else, but the Court originally told us to remove the corresponding cybersquatting section. After a series of motions, the Court has allowed us to include it in our Second Amended Complaint.

In early December 2009, there was a hearing on Summary Judgment motions that will decide many aspects of the case.

On December 10th, 2009, the Court ruled on both sides' motions for summary judgement. The ruling was a very positive result. Some highlights, in the order they appear in the decision:

  • The JMRI decoder definition files are copyrightable. (Page 5) All of Katzer's contentions that they were "merely duplicates" of NMRA information, "mere spreadsheets of data", etc, were rejected.
  • Katzer's misuse of the decoder definition files damaged JMRI, and we can hold him accountable for that. (Page 6) Katzer had argued that since we distributed our work "for free", there was no harm in misappropriating it. The Court said this was not a valid legal argument. That decision helps both JMRI and other open-source projects. (This decisions is the second time that Courts have slapped back Katzer's claims, and in the process made good law for FOSS)
  • Katzer cyber-squatted on "". (Pages 6-8) We still have to show what he intended when he did that, which is a question of fact to be decided at trial, but all the other pieces are in place to show that what he did was illegal. We don't expect much trouble proving his intent at trial, as he'd done this to other people before, and knew that DecoderPro was the name of our software.
  • Katzer did infringe JMRI's copyrights on the decoder definition files, in violation of the law. (Page 8) The Court ruled that his is "liable" for this, the civil equivalent of "guilty". All that's left for trial is to show the amount of damages, but it's already decided that they do exist (see above), and that he must pay them.
  • JMRI did not infringe Katzer's copyrights on the QSI manuals while creating decoder definitions. The Court found that
    1. we had a license from QSI to create the decoder definitions, despite Katzer's contention that we didn't, except he forgot to tell us (page 9),
    2. we used the material fairly (page 9)
    3. Katzer's counter-suit for $6 million was a "litigation tactic barred by the equitable doctrine of laches", in other words he made the claim up for the purpose of setting a trap.
    This was a decisive rejection of Katzer's arguments.
  • Katzer violated the Digital Millenium Copyright Act (DMCA) when he removed the copyrights from the JMRI decoder definitions he misappropriated. (page 10-11) We still have to show intent at trial, as that's a question of fact which Katzer has not yet admitted to, but we have little doubt we'll be able to do that. This part of the decision is also a victory for open-source in general, as it demonstrates that our copyright notices must be respected, not just removed to make it easier to hid misuse of open-source software.
Together, this set of decisions was a complete victory for JMRI at this stage of the proceedings.

After an extended series of settlement discussions and conferences, the case finally settled in February 2010.

The settlement agreement, which is now a legally-binding contract signed by both sides, accomplishes several things:

  1. A permanent injunction will be entered by the Court, expressly forbidding Katzer from further misuse of the software that JMRI has created, and forbidding him to register any domain names that should rightly belong to us.

    In the past, Katzer has told people that we've copyrighted manuals that others wrote, or trademarked terms that belong to others. Neither of these are true, and we've shown in Court that Katzer has no evidence for his statements. To prevent there being any uncertainly about this, the agreement explicitly states that we're not talking about anything from the NMRA or other manufacturers.

    This injunction is important. We consider his misuse of JMRI's domain name and the decoder definition files to have caused significant damage to JMRI, and this injunction makes it clear that the Court will not allow it to happen again.

  2. Katzer explicitly releases all JMRI users and developers from any liability for anything related to JMRI, up until 18 months from now. This means that he can't try to extract patent royalties, claim infringement, argue that somebody infringed a copyright, or anything else. He has no right to complain about anything we've done in the past or will do in the near future.

    The goal was to make sure that this dispute really is over, and won't rise from the grave like a zombie that has to be fought over and over again. Katzer has said that he just wants to move on and put legal disputes behind him, and this provision is aimed at making sure that's what happens.

    As a matter of fairness, Jacobsen extended the same release to Katzer.

  3. A procedure was put in place for handling any future disputes, after the 18 month period is up. This has some important features:
    1. It's "loser pays", so neither side can bring spurious complaints without risk.
    2. It's based on mediation and arbitration, so it's faster and can include people with more technical and model railroading background.
    3. It forbids going outside the agreed procedure, for example to contact an employer or relative in an attempt to apply pressure.

    If anybody really does do something wrong, this provides a way to deal with it, while making it riskier and more difficult for anybody to try to make spurious claims.

  4. Katzer will pay $100,000, spread over the next 18 months.

    Despite generous support from over a hundred people, the costs defending against Katzer over the last six years are significantly more than 100K$. (Things like court reporters for depositions, travel expenses, and expert witnesses have to be paid for, and it adds up very quickly.) Katzer's payment gets us part-way back to where we would have been if all this hadn't started in the first place.

    To be sure, he has had significant costs of his own, and he has to pay them too.

    Although this doesn't entirely cover our costs, we believe this is the best we could do at this point, and there was no certainty that going to trial in June would result in a damage award large enough to offset the large additional costs of a trial.

    Sometimes, the best you can do is the best you can do.

This is not air-tight, in that it doesn't extend to e.g. future patent complaints about future patents, after the next 18 months. We believe that we've still got a very strong position on those, as Katzer stated in Court that JMRI didn't infringe any of his existing patents, because Katzer's ongoing chain of patents actually document the same set of inventions, and because quite a few people have helped us accumulate several file drawers of strong prior art to defeat future patents.

It's very hard to craft an agreement that constrains all possible future behavior, so in the end you have to trust that the people making the agreement are going to live up to the intent of it. As Henry Stimson said, "The only way to make a man trustworthy is to trust him", and that's what we're doing here. We believe that this agreement has the right structure to motivate everybody to just work on their own stuff without getting into fights, and that it's fair. We're going to trust Katzer to approach it on that same basis.

We could not have reached this point without the help of the hundreds of model railroaders who have contributed money, time and knowledge to this effort, the open-source advocates and organizations who have worked on the legal effort, and particularly without the huge efforts by Victoria Hall, David McGowan and the other lawyers who have worked on this over the past six years. We'll have more to say about this later, but the message for right now is that we all owe them a huge debt of gratitude.

Despite reaching this point now, we still need people to Donate to JMRI contribute to help pay for legal expenses. JMRI gets no commercial revenue, but we're still 10's of thousands of dollars in debt because of all this legal maneuvering. Anything would be greatly appreciated.