[open-bibliography] OKFN blog: Bibliographica, an Introduction

[Jim Pitman]

Tom Morris <tfmorris at gmail.com> wrote:

> > I'd love to be able to extract out of OL the subset of book titles and
> > authors of interest to academic researchers e.g. in mathematics or statistics or economics or some
> > other discipline, but I dont have the resources to do this and maintain the results. 
> > Anyone able to help with that?

> I actually see the uncoordinated duplication of data as part of the problem.  

I see it as part of the solution, provided we empower agents who actually know something
about the data to help clean the parts they care about, and keep their own copies of the clean
data exposed back to a larger pool for all to see.
There can be many overlapping sets of more or less well-curated biblio collections.  
Curators can acquire reputations for quality.  


> The Open Library data is pretty low quality right now, but hopefully it will be improving 
> (hint to OL: people would help you with this if you let them).  

Let me strongly reinforce that. If I and others had easy access to selected subsets of a 
few thousand titles of OL data in fields they cared about,  we could machine match those 
titles against various other sources, and publish the results back to the web of linked data. 

> How will BibKN get the updates with those improvements?  

I think its up to BibKN nodes to query OL from time to time with ids to see if 
there is any change. If OL can provide node-specific feeds, thats very convenient. BibKN
nodes can then do what they want with such feeds.

> If BibKN goes ahead and cleans up their little extracted subset, how do those changes get 
> back to OL? 

That is entirely up to OL. OL can facilitate matters for users by linking from their records 
to a BibKN node they trust for further info. Users can chase that link if they care to.

> How are the two sets of changes reconciled if they conflict?

I see no need for global reconciliation, and the cost of facilitating that seems
far too great. 
OL admins and users can make whatever corrections they deem necessary from their perspective. 
BibKN node admins and users can do the same.

> I think acknowledging that there will be multiple repositories of
> bibliographic data with different focuses and figuring out how they exchange updates 
> cooperatively would be a useful thing.  

I strongly agree.

> Having strong provenance for the data and modifications to it is a key enabler, I think.

Again, strong agreement. 

> The main thing is to get the dialog established.  It sounds like there
> have been some bilateral discussions.  Perhaps it's time to widen the scope.

Fine with me. I welcome further contributions to the discussion.

--Jim

> open-bibliography mailing list
> open-bibliography at lists.okfn.org
> http://lists.okfn.org/mailman/listinfo/open-bibliography