Published on 31 March 2011
This post image

Every couple of years or so yet another patent troll impacts my work. Whether it's Data Treasury, LML, or other "companies" I can't mention, these folks patent the obvious early enough (often while an idea is half-baked) and sit on it. They then wait for others to build and/or buy products remotely similar to the vague patent, and then sue in East Texas courts to shake down licensing fees.

It's a ridiculously bad situation that could actually get worse as the Supreme Court considers a case that could hold a programmer liable for writing code while being vaguely aware of neighboring patents. Since so many obvious and trivial ideas in software have already been patented, the only safe harbor for some may be to stop programming. For many of us, that's like abandoning breathing.

Hence the constant cries for software patent reform. Congress is currently taking up the issue, but, as in past attempts, the strong lobbies that are pushing it in the wrong direction may kill it. And since legislative and judicial wheels turn at a glacier's pace compared to the rate of technology advance, don't hold your breath.

A better way to light a candle in this darkness is to simply defensively publish new ideas before a troll can patent them. This has the extra benefit of sharing those ideas that really shouldn't be regarded as competitive advantage. The downside, of course, is that you may often have to publish something you'd really rather keep as a trade secret. But that's the price you pay to avoid the risk of trolls.

There's clearly a place for those (rare) worthy and innovative software patents that offer a unique competitive advantage. But you won't find those with patent trolls nor in East Texas courts.

I've often said (and firmly believe) that an active, effective software developer produces at least one patentable idea each week. It's just what we do, but that's also a commentary on the relatively low standard for US software patents. One primary reason we don't have billions of software patents at the USPTO is that filing for patent protection is an expensive and inefficient process. So a company's patent portfolio is usually more a measure of its rapacity or the size of its legal staff than the degree of innovation.

I'm hopeful that the current woeful US software patent situation will one day be corrected, and the end result will be better protection, innovation, and collaboration: even beyond what's currently found in FOSS work and licensing. Until then, I'm afraid we'll have to endure some stagnation while the trolls have their way.