Earlier this week I wrote about a useful study of the economics of copyright, pointing out that we need more such analyses in order to adopt a more rational, evidence-based approach to drafting laws in this area. Of course, precisely the same can be said of patents, and software patents in particular, so it's always good to come across work such as this newly-published doctoral dissertation [.pdf]: “The effects of software patent policy on the motivation and innovation of free and open source developers.”
Here's part of its summary:
The theoretical framework analyses in detail how the sixteen motivational factors collected from the FOSS motivation literature may be affected by the presence of software patents. As a results from the theoretical analysis, I expect software patents to increase the levels of extrinsic motivation and decrease the level of intrinsic motivation.
This is the main result:
In the first empirical study, the effects of software patent presence on selected, representative motivational factors is investigated. To do that a concept of software patent presence is introduced that not only includes jurisdictional components (legal availability and patent incidents), but also a new metric that captures the varying patent pressure across software domains. Empirically, none of the two camps of proponents and opponents of software patents find support: software patents do not negatively affect intrinsic motivation as predicted by opponents, but they also do not positively affect extrinsic motivational factors as proponents predict.
In the second empirical study, the analysis is extended. Here, the effects of software patent presence and motivational setup on the individual innovation behavior of FOSS developers are investigated. To do that a new metric for innovation behavior is introduced that ranks individual code contributions according to their level of innovativeness by distinguishing algorithm-based from reuse-based contributions. The argument is that algorithm-based contributions are more innovative than reuse-based contributions. Beside this ordinal innovation scale, ‘reverse-engineering’ as a specific type of code contribution that is important to the FOSS world, is also analyzed.
One key result is that intrinsic motivation triggers more innovative code contribution, while extrinsic motivation correlates with less innovative code contributions. The presence of software patents (using all three metrics mentioned earlier) does not seem to play a significant role, except in the case of reverse-engineering. This type of code contribution correlates with a stronger presence of software patents.
Those are all useful data points, but I do wonder whether there might be another issue about the real, rather than perceived threat that software patents represent to free software.
After all, do many "ordinary" hackers really spend much time thinking or worrying about software patents? Certainly, the community talks around them quite a lot, and there will doubtless be a sense of their constant presence in the world of software. But since the accepted wisdom is to *avoid* trying to find out whether you are infringing on software patents as you code – because if you do check, and you then infringe, damages are often greater for “wilful infringement” - surely there is a bias against understanding the full threat they represent.
Nor is avoidance of "wilful infrintement" the only factor. People's perception of the threat posed by software patents is inevitably coloured by what's happened so far. There have still been relatively few cases that have come to court involving free software, so some people may take the view that software patents aren't really a problem for them. But that ignores all the cases that *didn't* come to court, because they were settled behind closed doors, and the ones that are on the way but haven't yet broken.
What this might mean is that although hackers' views and motivations are relatively unaffected by the existence of software patents, they might in fact find themselves hugely affected if major software companies or patent trolls start trying to assert their software patent portfolios – something that many fear might happebn. True, this is only speculation, but at the very least, it might provide an interesting topic for further research....