Matt Asay: Completely Wrong on IBM Patents

by Jason

Friday, April 09, 2010 at 08:34 AM EDT

Wow. I don't often agree with the Canonical COO, Matt Asay, on many things and even when I do in general, I often find myself in disagreement with details.

In Mr. Asay's recent piece "IBM patent claims show open source has arrived", that's not a problem -- because I disagree with everything, filled as it is with inaccuracies and sloppy thinking.

Here we go

Mr. Asay starts rather badly, calling IBM's letter to TurboHercules a "cease-and-desist" letter to the "OpenHercules" open-source project.

Problem 1: The open-source project name is "Hercules", not "OpenHercules".

Problem 2: The letter is not a "cease-and-desist" letter. A "C&D" is an order or request to halt an activity, or else face legal action. IBM's letter does no such thing.

So we see from the start that Mr. Asay's understanding of the situation is perhaps less nuanced than one might hope.

Building on a foundation of sand

Because I doubt Mr. Asay actually understands the situation, I am not surprised that his analysis is poor.

For example, consider Mr. Asay's attempt to frame IBM's actions a being "willing to defend its mainframe business against any and all threats, open source or otherwise."

The problem here is that IBM explicitly pledged not to assert 500 named patents against Open Source and that 2 of those patents where among those listed in the IBM letter.

You don't get to explictly promise not to assert patents against Open Source and then turn around and assert patents against Open Source. It doesn't matter if Open Source is a "threat" or not. You made a promise.

Of course, promises are made to be broken, especially if the cost-benefit analysis says it is more profitable to break the promise than keep it. This is just one of so many reasons that people take Microsoft's various promises, pacts, covenants and double pinky-swears with a grain of salt.

Watch out

Be careful now -- if you want to excuse IBM by pointing out that they can chose to enforce patents outside of the 500 named (let's assume they listed the 2 by mistake and will retract them), then you must in turn acknowledge that projects like Mono and Moonlight which range far far beyond the standardized core are in explict danger as well.

I do not think IBM is legally out-of-bounds here, no more than I think Microsoft would legally be out-of-bounds to shut down vast portions of Mono and Moonlight now or in the future. (Perhaps after Novell is bought out and agreements are no longer renewed?)

I'm just surprised Mr. Asay finds this an encouraging development for Open Source.

Revealing a flawed understanding of Open Source

Mr. Asay reveals -- not for the first time -- his corporatized view of "Open Source":

This doesn't make IBM an enemy to open source, because that sort of statement doesn't make any sense anymore, now that open source is just how software gets written by just about everyone, at least, at some point in the software supply chain.

This is view of Open-Source-as-a-design-methodology absolutely misses the entire point of Open Source.

By reducing Open Source into "that's just how software gets written" and ignoring everything else, Mr. Asay shows his understanding of Open Source is either superficial or fatally biased.

And I'm not talking about "Free Software" here -- look to the OSI's own definition, paying special attention to "No Discrimination Against Fields of Endeavor".

Really off the Reservation

Now, even though we've seen that Mr. Asay neither grasps the specific situation of IBM and Hercules, nor the general situation of Open Source, the next bit is still a bit shocking:

IBM has always been an opportunist when it comes to open source, just as every open-source company, project, and developer is. Try ripping off Red Hat's trademarks and see how long it takes before a cease-and-desist letter lands on your door. Or try stealing GPL code from the project of your choice, without contributing code modifications back as per the license, and see how that makes the developer feel.

This is an express train of ignorance that shot pass Stupid Junction in Wrong City and ends up derailing just shy of Did He Really Say That Central?

Problem 1: Every open-source developer does not try to stop competing developers with patents. Every open-source developer doesn't even file for patents. Every open-source project does not try to stop competing developers with patents.

This goes to one of my main complaints with Mr. Asay and others of his pro-corporate bent: stop trying to pretend like every person on earth is a souless entity focused on profit above all else.

Some are, and some aren't. Even assuming one stays within the bounds of the law, there's a whole lot of grazing land between What Is Legal To Do and What Is Ethical To Do.

Stop pretending like everyone who participates in Open Source would sell out every other participant if they thought it would benefit them. Put plainly, it's damn disrespectful.

Problem 2: Trademarks and "stealing GPL code" are not the same as enforcing patents. I suspect Mr. Asay knows this and knows he is (at best) using bad examples of equivalency here, so I must wonder why he is doing it?

One of the main differences is that it is highly unlikely that you will innocently "rip off Red Hat's trademarks" or "steal GPL code". However it is highly likely that you will innocently infringe a patent if your software product is non-trivial.

Another is that "ripping off" and "steal" implies malice in appropriating other people's property. Even if Hercules is infringing an IBM patent, it does not follow that it was done maliciously. Why use language then, to characterize it so?

Train keep a rollin'

You think I'm reading too much into this? Consider the very next paragraphy from Mr. Asay:

If, in fact, TurboHercules is violating IBM's patents, shame on it. Just because OpenHercules is open source isn't a license to steal, any more than IBM should have the right to pilfer code from it or any other project without complying with the license grants afforded by such projects.

This is ignorant in the extreme, and not just because Mr. Asay can't even be bothered to Google the project under discussion to get its name right.

As we've already seen, "violating" patents can be near unavoidable and most certainly does not carry the same mens rea that "stealing" or "pilfering code" does.

In this short paragraph, Mr. Asay reveals a fundamentally flawed understanding of both software development and patents and copyright.

What we are talking about

Stacking bewildering point atop bewildering point, Mr. Asay at last comes to his truly bizarre conclusion:

This isn't cause for concern. It's cause for celebration. It means open source truly has arrived.

Really? Not a cause for concern? If I were working on an Open Source project that might compete with an IBM offering, I would be just a touch concerned.

I certainly would not be celebrating - what sort of idiot party do you suggest? The "Glad I'll Have More Free Time When MegaCorp Shuts Me Down" Shindig?

Overall situation

Leaving behind (mercifully) Mr. Asay's questionable analysis, I am witholding my own interpretation of the overall event.

I strongly oppose software patents, especially when wielded offensively, and double-y especially when wielded offensively against Open Source projects.

However -- in this particular case -- it is not clear to me who all the actors (and motivations) in this story are, and to be frank it's not clear to me that IBM is actually going after the Open Source project in question. I need to see more of the document trail and I think associations need to be vetted a bit more.

This article originally appeared on The Source.com.