Software Patents
I’m in the midst of laying the groundwork for a small offshore IT outsourcing operation and it’s become evident that it will face some unique challenges. The most obvious of them can be seen in the mainstream press attention that this business has been getting lately, much of which focusses on the backlash by out-of-work tech employees. Brian Behlendorf, however, suggests in a short article on Salon.com why those who would undertake such a business are not necessarily greedy bloodsuckers with no regard for the common good.
The other challenge involves our unwillingness to engage in projects that involve software patents. I’ve just reviewed some of the documents on the website of the League for Programming Freedom, and I’ve never been more convinced of the failure of software patents to achieve the goals for which patents were originally devised: encouraging innovation and fueling progress. In fact their principal effect currently seems to stifle invention with major patent holders intimidating developers who may have independently arrived at a given solution to a programming problem with the threat of litigation.
As Larry Lessig points out in an article written four years ago for the defunct Industry Standard, “On average it takes $1.2 million to challenge the validity of a patent, which means it is often cheaper simply to pay the royalties than to establish that the patent isn’t deserved.” Pursuing frivolous patents becomes a strategy for coercing potential competitors to licence your “innovation” rather than entangling themselves in expensive litigation.
If you’ve read some of my previous entries on this site, you will have noticed that I’m generally skeptical of many intellectual property claims. I am however, willing to concede that in a broad range of cases patents have helped achieve these goals. But I’m with the LPF in their efforts to abolish software patents. To summarize some of the main points in the League’s introduction to the issue.
To qualify for a patent, an invention must be judged not to be “obvious.” With the emergence of software patents subsequent to a 1981 Supreme Court decision which broadened the scope of what was patentable, programmers have discovered that computing techniques that were regarded as obvious by the programming community for many years have been patented, and that it is suddenly illegal to continue to incorporate these techniques. Before that case, it would have never occurred to a programmer to attempt to patent these techniques or algorithms. Part of the problem lies in the fact that examiners hired by the U.S. Patent Office are underpaid and frequently unqualified to judge what exceeds the this criteria of obviousness.
The Office continues to apply principles that are irrelevant to the field of software development. Because of the “tractable” nature of computer programming, hundreds of problems can be solved with dozens of “inventions” by a single programmer in a single day. The incentive patents provide is simply inapplicable here and, as I’ve mentioned, software patents have the opposite of the intended effect of encouraging innovation.
It may be useful to note that denying the legitimacy of software patents does not deny a software developer’s right to copyright and sell a program–it simply denies him the right to patent the programming methods used to create it. This issue is thus completely separate from the open-source vs. proprietary software debate.