[OSM-legal-talk] ODBL enforcement: contract law and remedies

[Ed Avis]

Matt Amos <zerebubuth at ...> writes:

>these sites are in non-compliance with the license
>http://wiki.openstreetmap.org/wiki/Lacking_proper_attribution

Would switching to ODBL (or any licence) solve this particular problem?

>in any case, it's not useful to talk about people "stealing" the data
>- anyone can take a copy of the data without depriving us of the
>original copy. there are people who aren't in full compliance with the
>license, but in all of these cases i'm pretty sure it's an oversight
>more than it's deliberate.

Yes.

>as a thought experiment, what would happen if i took the latest planet
>and put it up on my server (let's assume that both i and my server are
>in the US) with a PD license?

Personally, I would say nothing; wait for someone to start distributing it
under non-free terms and then contact them.  If they made an honest mistake
thinking the data really was in the public domain, they will stop; if not,
it is just the same situation as someone taking the planet file and using
it directly for some non-free distribution.

So what then?  One of the copyright holders would have to sue.  To have
grounds for copyright infringement he or she would need to show some
'spark' of creativity in the mapping work (at least according to
<http://en.wikipedia.org/wiki/Feist_Publications_v._Rural_Telephone_Service>).
The lively discussion on mailing lists and wikis of how to tag things,
and the commonplace addition of 'logical' elements (such as intersection nodes)
which do not exist in the real world, looks to me like prima facie evidence
of such a 'spark' or 'minimal degree' of creativity.

Even if the case is not cut and dried, there is certainly enough here to
keep the lawyers busy for a while.  Which, IMHO, is a strong enough deterrent
for anyone thinking of misusing the data.  Consider how much time and money
the SCO - Linux case has taken up so far, on a far flimsier basis.

One thing which weakens the case is that there is not a single copyright
holder.  Certainly copyright assignment to a single entity such as the
OSMF would make it easier to sue.

(As discussed earlier, even if the USA declines to recognize any copyright
interest in OSM data, there are other jurisdictions, and few US companies
would want to use data they had to keep strictly within the USA's borders
or risk a lawsuit.  I just don't believe it would happen.)

>i think we're agreed that all licenses have flaws. the sticking point
>seems to be that i'm of the opinion that CC's flaws are so great that
>the hassle of moving to a better license is the lesser evil. you
>appear to be of the opinion that the hassle is greater than any
>potential benefits. is that an accurate statement?

That is fair; I might even go so far to say that with 50k contributors,
a licence change is 'totally unworkable' and 'not an option', to borrow some
of the phrases used earlier.  However I would like to be pleasantly surprised
about this.

I'm also not a big fan of the ODBL, particularly the contract-law provisions,
but it seems I'm very much in the minority on this.

-- 
Ed Avis <eda at waniasset.com>