[OSM-legal-talk] ODbL and duration of IP protection

[Mike Collinson]

At 11:47 PM 18/08/2010, TimSC wrote:
>Hi all (here comes trouble, again),
>
>Another issue which occurred to me is the issue of the lapse of exclusive rights over the OSM dataset, including copyright, database rights and other IP. The issue is a really long term consideration but it is important to get the foundations correctly laid. The exact duration of protection is probably horribly complex due to different legal jurisdictions. Ideally, the database would be protected for a finite period and would come out of protection across the globe simultaneously. The latter is necessary for legal predictability.
>
>As I understand it, EU/UK law has a 15 years database right which renews if there are more recent substantial changes. If OSM continues in the long term using ODbL, this is effectively a perpetual exclusive right. For areas which rely on contract law, it seems that OSMF again retains a perpetual right over the data. I will try to argue this is not good.
>
>The public domain is important for all created works, as it provides a "no strings attached" pool of cultural and intellectual resources. This removes the burden of legal compliance on future works (which is definitely non-zero effort). Check out Lessig's book Free Culture in which he argues that long or perpetual copyrights create a permission based culture which is detrimental to creators. For that reason, many exclusive protection rights have LIMITED duration. This is acknowledged in the US constitution: "To promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries." Obviously, OSMF are intending to make the data available under an "open license". This doesn't remove the high transaction costs and license un-interoperability of ODbL in the long run [1] of decades and centuries.
>
>To some extent, these issues were not of our making, but we at least try to address it.
>
>So the issues are:
>
>1) The duration of the lapse of OSMF's exclusive right is probably different in various jurisdictions. This creates complexity in reuse for PD and therefore legal uncertainty.
>
>2) The duration claimed under the ODbL in some jurisdictions seems to be perpetual, which is imposes an unacceptable transaction and license interoperability burden on future creators.
>
>One possible solution is to explicitly state the duration of protection under ODbL in the license.
>
>Btw, the solution that "OSMF will fix it later" is not really good enough for me; what makes the future better for fixing problems when we have the present? In fact, all creative common's comments on ODbL really apply [1]. Given the fairly abstract nature of my concern, I guess many will think "who cares?". Well, I care. As a side note, I notice that I can't get an answer to a simple legal question [2], so for this is horribly complex issue, I don't hold out much hope on a straight answer!

I think there is actually a straight simple answer on this one! The Contributor Terms.

As the current objective is protect the Share-Alike nature of OSM data and require attribution, the complexity is a good thing not a bad thing. FUD works in our favour for once. ODbL fires three guns copyright, contract and database rights. If someone wants to circumvent Share-Alike then it is up to them to figure it all out and be sure that we won't be jumping up and down. My daughter is training to be a lawyer; gosh, she could earn a fortune.

If the general open geodata environment converges on Share-Alike, then we need do nothing.

If, as Frederik suggests, the general open geodata environment moves away from Share-Alike (and perhaps attribution chains) the Contributor Terms have the mechanism ready by design [1]. Someone proposes a license change variant "You allow the OSMF to also publish historical data dumps older than x years under CC0".  That is all it takes.  If 2/3 of the future mapping community are happy with that, then the OSMF can. If they are not, then it can't.

<rant>

I also personally agree with Lessig.  A 25-year old mapper today may well live to 100. Using life plus 50, that means that in some jurisdictions, OSM data is potentially copyright-restricted until 2135!! I think that is just plain crazy ... but I could be plain wrong.  Many respondents in recent exchanges appear to be completely missing the importance of this and why it is s-o-o import to get beyond worrying about maximising what you can import or derive from right now to intelligently setting up a simple system that allows OSM to self-adjust and remain relevant for a very, very, very l-o-n-g  time.

</rant>

Mike

[1] And this may help explain why the "free and open license" bit and why Clause 2 of the new Contributor Terms is so much broader than what the OSMF can actual do with your data under Clause 3.  It is nothing to do with evil machinations, plots or magic spells. It allows the future contributor community, not the OSMF, to decide for themselves in a practical, democratic environment.
http://www.osmfoundation.org/wiki/License/Contributor_Terms