Researchers' Right To Open Source Research

bstadil writes: "There is an interesting debate over at SiliconValley.com about the right of researchers funded by Universities to make their IP Open Source. It's not at all simple. On one side Universities claiming their derive 5% of their Budget from IP licenses and it's vital for continued high level of 'Output,' on the other hand researcher who claim the public is billed twice by licensing the output."

Comments

It's not just employees...

[cperciva]

I'm a graduate student at Oxford University, and in the University's statutes they claim ownership of any code I write while I'm here. I am negotiating with them to try to get permission to release some of my work -- right now I'm working on network protocols, and a protocol isn't much use if nobody uses it -- and they haven't been entirely unreasonable, but after two months I still haven't got anything in writing.

It is one thing for a university to claim ownership of work produced by their employees; it is quite another for a university to claim ownership of work produced by people who are paying to be there.

Why can't they do both?

[alienmole]

I would think in many situations, universities could play both sides of the fence: make systems available as open source, but charge money to license code to companies that want to package it without source, in proprietary products (the SleepyCat approach that was discussed here recently).

This approach has a better chance of working for universities than it does for ordinary commercial enterprises, for at least two reasons:

1. The sort of software universities produce is more likely to be the kind of code that will be integrated into other systems, which lends itself to a dual licensing approach. Universities aren't selling shrinkwrapped software to consumers: they're selling more basic technology to companies that want to exploit it commercially. This could be perfectly suited to a dual licensing approach. Legitimate businesses, for the most part, are unlikely to try to base products on software that they don't have rights to.
2. Universities don't rely on software licensing for their entire livelihood, so if an open source strategy happens to result in somewhat lower revenues, they can handle it. However, open source may be one of the best and cheapest ways of "advertising" a university's software products, so these factors could balance out.

Besides, this is exactly the sort of issue on which we should look to universities to lead the way. Open source is an important form of cooperation, and its heritage is the very academic freedom and open sharing of information pioneered by universities. There are benefits to this cooperation that may not be completely in conflict with the profit motive; however, the truth of that claim can only be verified by those with sufficient vision to look beyond the next quarter's results. Universities are one of the few organizations which have both the vision and financial ability to do that. MIT's recent decision to make its course material freely available over the web is an example of this.

It should be the researcher's decision

[YoJ]

I think that research work done while taking money from the NSF and other public agencies should be freely available to the public. In general this means that the research is published in public refereed journals that anyone can buy and read. In some sense the "intellectual property" of the research is being given away to the public, since anyone can read it. In another sense, the researcher and the university "own" the idea since no-one else can claim credit for it. But products and patentable ideas get a bit murkier.

I think everyone agrees that it is immoral for someone to do research while accepting public money and then keep the research secret and proprietary (except in extraordinary circumstances). There is also something fishy about a company being granted exclusive rights for an idea that was developed using public funds. The universities would like to patent everything themselves, but in practice it is often the decision of the researcher whether an idea should or should not be patented.

If I am on a project and write code, I ask whoever is in charge if I may release the code to the public. If they say no, I would want a pretty convincing explanation of why not. I don't think public research should have any secrets.