[okfn-discuss] freebase: another new open knowledge/open data service

[Rufus Pollock]

rob at robmyers.org wrote:
> Quoting Francis Irving <francis at flourish.org>:
> 
>> Regarding licensing, I don't know anything about law, but perhaps as
>> well as the Copy Right it is possible to license out the Database
>> Right :)
> 
> The Talis licence is a DB right copyleft license, for example. But the 
> DB right
> only applies in the EU, so it is not useful for global projects. And I

There is a DB right in many other countries either explicitly or 
implicitly (i.e. under copyright), for example Canada and Australia both 
have a common-law like protection and the Feist decision in the US is 
not nearly as broadly exclusionary as many people think. For more on 
this see this post summarizing a talk by Professor Hugenholtz:

http://lists.okfn.org/pipermail/drn-discuss/2005-November/000086.html

As well as this article by the man himself:

http://www.ivir.nl/publications/hugenholtz/fordham2001.html

> personally think that it's not a "right" that we should have in the EU, 
> never
> mind be trying to use anywhere else. :-)

I have to say I am not so sure on this point.

Firstly, databases are fairly easy to lock up using secrecy and 
restrictive licensing -- particularly in the digital age. Thus, the 
alternative to the right is not open dbs but dbs which are closed by 
access restrictions rather than using an IP right. Furthermore for those 
projects which *are* open having a db right provides a means to protect 
the 'data' commons using share-alike provisions (which without the 
underlying right would be meaningless). In such circumstances the 
existence of an explicit DB right is good for open knowledge and not 
bad. As I wrote to Jamie Boyle:

<quote>
Finally I wish to play devil's advocate and ask what you think is wrong 
with the following proposition.

Assuming that:

   a) we can use an exclusive, 'closed', right to build an open 
'commons' (as with 'copyleft' for software)

   b) in the long run open knowledge drives out closed

   c) prior to the DB directive there were already db-type rights in 
various EU member states as well as in many other common law 
jurisdictions (I also note that, when asked why the US had not created 
DB rights as in the EU at a seminar in Cambridge in November, Professor 
Hugenholtz expressed the view that this was largely because US DB 
companies were able to protect themselves by recreating IP type rights 
through contract)

Then, in the long run, the introduction of a harmonized DB right will 
promote the openness of databases rather than hinder them.
</quote>

Secondly, I would point out that even if a CC license were not *legally* 
effective it might have social value in making explicit what the social 
contract was for access to a given dataset.

For example, in a community of scientists even without an enforceable 
right there might be many other social sanctions which could be used to 
encourage (and sanction) people in regard to sharing (and not-sharing).

In such circumstances while not legally enforceable the license makes 
explicits the basis on which the community will be working on the 
dataset and what is expected of those who use it. (For more on this see 
this email to the open-data list 
https://mx2.arl.org/Lists/SPARC-OpenData/Message/100.html).

Regards,

Rufus