PatentBS series #1: Abolishing software patents would harm Free Software

We read so much nonsense about patents these days. Calimaq started a #CopyrightMadness series and a blog so, sometimes, I feel like somebody should start another series: PatentBS. It is self explanatory.

Well, ok let’s start with this one from a Wired Opinion by Manny W. Schecter “Chief Patent Counsel, Associate General Counsel, and Managing IP Attorney at IBM – the top annual U.S. patentee for the past 19 years.” I won’t discuss the argument about how patents supposedly promote innovation, this rebuttal opinion already does it. So let’s concentrate on this piece of nonsense:

Eliminating patents for software will not enhance innovation or benefit our economy. Software is also the most easily appropriated type of intellectual property. Ever since U.S. courts made it clear that copyright is unavailable to protect their ideas, developers have sought to protect inventions embodied in their software via patents. Denying patent protection for software will cause these developers to look for other ways to protect their IP investment — resulting in code that is less open, less accessible, and less interoperable.

Such balkanization would discourage many of the collaborative ingredients of the very software ecosystem that has had enormous economic and technological impact.

Copyright can’t “protect” (i.e. monopolise) ideas implemented in software. So that’s why patents are used. We’ll pass on the rhetoric (ideas/inventions as if they’re the same thing).

The argument is: if we don’t grant patents to protect software developers, they’ll seek other means to protect themselves, and that will result in the production of less Free Software (aka open source software). That does not make any sense at all. Phew.

Patents and Free Software are antagonistic. They can’t work together, unless the patent owner grant a royalty-free license to anyone using the software covered by the patents, rendering the patent basically useless. So what would happen if we abolish patents? As the argument goes: we would have less Free Software because software developers would seek other ways to protect “their IP investment.” Well it’s funny, because it’s exactly what happens, and it’s a good thing.

Software is actually covered by copyright, and developers use copyright licenses to foster their development all the time! And this is exactly what’s used in Free Software, for instance with a copyleft license such as the GNU GPL. And that helps produce more Free Software, not less.

3 réponses à “PatentBS series #1: Abolishing software patents would harm Free Software

  1. FC dit :

    While we agree on the fact that this is utter BS, aren’t you somewhat misconstructing the argument?

    I think the gist is that he is using « open » in a way that does not refer to Free software, nor even « open-source ». The « less open » makes me believe so – it’s hard to define levels of openness, especially in this context. Either you give your source (whatever the license), or you don’t for fear of copycats. Seems pretty binary to me.

    So it’s possible, in a protracted way of thinking, that he argues that without patents, developers will have to obfuscate their code more to make it more complicated to understand decompiled code, restrict exposed APIs (less interoperability), use more DRM-type techniques (less accessibility).

    Again, that would be complete BS, too, but not quite the one you’re responding to.

  2. Hugo Roy dit :

    > aren’t you somewhat misconstructing the argument?

    Actually, the problem with most of this text is that there are no real arguments. It’s mostly rhetoric, shifting words and their meanings, obfuscating sense rather than making it.

    So I was trying to debunk the argument. In some way, that involves (mis)constructing it (the mis is for depending on whether you agree with my construction or not :-)

    It’s not clear whether “open” here refers to Free Software indeed. He’s not clearly stating “Open Source” but just open, accessible and interoperable. (And I guess it’s true that interoperability can also be achieved with patented technologies — it’s just that it usually involves paying FRAND royalties per-copies of the software that implement them, thus excluding Free Software.)

    But whatever meanings you put behind what the words “open code” convey in his argument, I think anyone reasonable will get to the conclusion this argument is patentBS.

    > Either you give your source (whatever the license), or you don’t for fear of copycats. Seems pretty binary to me.

    Yes and no. It’s true that checking whether a program is distributed with a license approved by the FSF, the OSI or the Debian FTP Masters (they 99.9% of the time reach the same verdict) is a pretty binary way to decide whether a program is Free Software or not. But I think there are various levels of “openness.“ For instance, there’s a word that comes here and then in his argument: protective. Copyleft licenses offer legal protection against free-riders. But you can also find projects using non-copyleft licenses (e.g. Apache v2) which are basically not open in their way of development (Google’s management of Android isn’t typically open to the community on github ;-) whereas some projects are open at core, like Mozilla (at least they claim so, and I tend to believe them). To me, real openness is a lot more complicated than Andy Rubin’s definition of “open”

    the definition of open: « mkdir android ; cd android ; repo init -u git://android.git.kernel.org/platform/manifest.git ; repo sync ; make »

    https://twitter.com/Arubin/status/27808662429

    In my view, it makes more sense to talk about the openness of a project rather than to talk about openness of code. Openness of code doesn’t really mean anything (either you can read it, or you cannot). But “open code” usually refers to Free Software, software anyone is free to use, share, study and improve. So that’s why I made this choice.

  3. [...] But in 2013, it seems that the fight for abolition might happen right there. [...]