Common Knowledge

Getting It Right: WikiPathways

Deciding on the right license for an online community can be a touchy process.

Sometimes the community is focused on the organizing principles learned from software. Copyleft has been powerful in growing free software, and is regularly insisted on by online communities that build data, or community science content – not because it makes legal or technical sense for data or community science content, but because it’s a security blanket known well from software.

The problem is that in a lot of cases, share-alike breaks the interoperability of data and content, in a way that it doesn’t in software. We’ve covered this elsewhere at Science Commons. It’s a pretty simple problem: copyleft tends to require redistribution of derivatives under the same license that the original was licensed under, but if you have two works mixed together with two different copyleft licenses, you cannot legally comply with either license. They are incompatible.

Another problem is that much scientific work doesn’t fall under copyright. Facts are free, so to speak, and can’t be made more free by the use of a license or contract. They can only become encrusted with legal language that kills integration and interoperability. And stuff that is unclear in terms of copyright – metabolic pathway diagrams, for example – can get encumbered with contracts that make its integration into databases, semantic networks, ontology federations, and more nearly impossible from a legal perspective.

The choice really depends on what the community wants. Some open communities want the copyleft, whether what they do is copyrightable or not, and don’t worry about their integration into other products being chilled. This is certainly fine, though it does create a bias for integrationists away from their content. Of course, closed communities don’t even address the question at all! But some communities are deciding to embrace integration as a first principle, and looking to innovative ways to protect the norms under which they work without creating legal crust around their content.

With all of that as background, here’s to WikiPathways, who have just re-licensed their content under the most liberal terms possible at the moment: CC-BY. On top of that, they’re going to use trademarks to reward the people who abide by their terms of use rather than using contract and copyright to punish people. That’s a tremendous change in attitude, one that I think makes a lot more sense in the data world than the security blanket of copyleft, or contractual re-constructions of copyleft…it’s definitely going to give WikiPathways the legal edge when someone’s got to choose whose pathways to integrate into larger systems, because you know you won’t get sued for some arcane license violation.

They’re also providing stable identifiers for pathways. The nerd in me says w00t and wonders if we can convince them to create neurocommons-compliant URIs.

A deep bow to the team at WikiPathways for their choice to make their work legally available and technically stable!


  1. #1 Egon Willighagen
    December 3, 2008

    “under the most liberal terms possible at the moment”…

    Would that not be one of the OpenData licenses which sort of put the data in public domain by explicitly waiving all rights?

The site is currently under maintenance and will be back shortly. New comments have been disabled during this time, please check back soon.