"Piggybacking" and the open-source trademark issue

Rosetta Stone's trademark lawsuit against a competitor brought to mind the simmering issue of trademark violations in open source. One of the opportunities and challenges in open source is that presumably anyone can be an expert in Project X, Y, or Z. Because of this dispersed expertise, the opportunity to run afoul of trademark violations is rife.

JBoss was constantly wading through this issue, so much so that a website was devoted to it. Bill Burke, then JBoss' chief architect, listed trademarks as one of the most important considerations for any open-source business. Why? Because being the source of code arguably matters more than source code in an open-source business.

Trademarks are all about source. Who is the source of a given product or service?

It's therefore not surprising that Bill Dudney got swatted for offering JBoss training and services. Why? Because he was using JBoss' trademarks and goodwill to sell his own product, and his advertising could have been construed to be advertising services offered by JBoss, the company.

Back to Rosetta Stone. It is complaining about "piggybacking" by a rival, Rocket Languages. What is piggybacking, and how does it relate to open source?

...[S]ome of Google's biggest advertisers are growing angry over "piggybacking," a practice in which smaller advertisers use the trademarked words of big brands in the text of search ads to divert traffic from the sites of bigger advertisers to their own sites.

See the problem? For those offering "MySQL training," as an example, unless they're authorized by MySQL to offer the training, they're likely violating MySQL's trademarks. MySQL can choose to turn a blind eye...and risk diluting its trademarks. Trademarks are only useful if enforced.

I'm not talking about rampant violations of copyright and trademarks (like this Russian rip-off of Alfresco's website). No, I'm talking about the allegedly mainstream offering of services for commercial open-source projects under the banner/trademark of the project. It's wrong to use someone's trademarks in this way, but I see it all the time for a wide array of open-source companies, including my own.

I think the primary problem here is ignorance. Most people don't want to disobey the law. They simply don't understand it. The same holds true outside of open source with Google AdWords, which is the subject of Rosetta Stone's lawsuit.

I'm an attorney, but not a practicing one. It would be great to have Mark Radcliffe, Heather Meeker, Jason Haislmaier, or another practicing IP attorney put together a post on rough groundrules for ensuring trademark compliance in open source.

[The_Decider]

With that logic no one can offer training or add ons to anything without permission and likely hefty fees. That will damage usage making the trademark moot.

This just shows the idiocy of business people and the need to totally revamp IP laws.

[JCPayne]

Toss the whole IP nonsense out the window...... Life should be simpler.

[Sortova]

As someone who has been making a living for six years with a free and open source product, trademarks are dreadfully important. The brand is everything. At OpenNMS we have worked hard to create the sense of value and quality about the product, and it would be easy for someone to ruin that.

Case in point. Shortly after I started maintaining OpenNMS, a company in Orange County, California, started offering "OpenNMS for Mac". They had set up their own website, independent mailing lists, and even released their own "alpha" version.

There were several problems with this. One was that OpenNMS required a Java 1.4 JDK, but one was not available for Jaguar (the version of OS X that was available at the time). The 1.4 JDK was required because the 1.3 JDK simply couldn't handle the work we were throwing at it, so this company had removed all of the 1.4 dependent code, basically forking OpenNMS, and in doing so had made the product much less stable. (Note: we support OpenNMS on OS X via the fink project).

Now what would have happened if, say, MacWorld downloaded their version, tried it out, and then published that it sucked? That would have negatively affected our brand. I tried to explain this to the company and even invited them to get involved actively with the project, but they were determined to associate their company with our brand and would have nothing to do with the project as a whole. I even pointed out that they could call it "BubbaNMS" or whatever, just not OpenNMS, but still they insisted on using not only the name but the logo.

So, for the first time since I started the company, I wrote a check to a lawyer in order to send a Cease and Desist letter and they finally stopped.

OpenNMS is 100% free software, but the brand is not. As a services company The OpenNMS Group has to protect the reputation of the product as being a replacement for OpenView and Tivoli, and the reputation of the company for delivering expert services. While we encourage people to use OpenNMS and even offer services on the product, we have to draw the line if they imply any kind of "official" partnership status, since those partnerships are one of the ways we generate revenue and so we can also exercise some level of quality control.

While everyone knows that CentOS is a fork of Red Hat Linux, and Oracle forks CentOS, it really wouldn't be either fair or accurate to say that they were the same. Red Hat deserves some credit for the brand they have created.

At OpenNMS we live by two main rules: OpenNMS will always be 100% free, and OpenNMS will never suck. We work extremely hard to live up to those rules, and we ought to be able to take credit for it.

[jlehr]

Matt,

It's ironic that in your blog posting taking companies to task for what are often innocent, unintentional violations of copyright, you yourself have may have engaged in defamation. That strikes me as a more serious matter -- even if unintentional -- especially since the offending words target an individual rather than a company.

Contrary to your statement, Bill Dudney never misused JBoss trademarks and goodwill, nor did he offer, or even so much as attempt to offer, JBoss training and services. Bill simply wrote a blog entry in March 2007 lamenting the aggressive legal tactics Red Hat's legal team was employing against companies such as Genuitec (the developer of MyEclipse) at the time. Why the false accusations?

Also, please note that Bill is one of three certified Alfresco trainers (including myself) who provide Alfresco training on behalf of my company, About Objects, Inc. -- the sole certified Alfresco Training Partner for North America. If Bill had actually engaged in any of the conduct your post claims I suppose it would be fair game for your blog, but that's simply not the case here. And as an Alfresco partner, it would be nice if you could at least refrain from libeling our instructors.