[open-bibliography] New group for bibliographic data on CKAN

Mon Mar 1 20:39:37 UTC 2010

On 1 March 2010 16:49, Jim Pitman <pitman at stat.berkeley.edu> wrote:
>
>> > I've just started a new CKAN group for open bibliographic data:
>> > http://ckan.net/group/bibliographic
>> > Anyone else on the new Working Group interested in helping to maintain
>> > this? If you want to be an admin, please register an Open ID on CKAN
>> > and let me know details off thread.
>
> I dont really want to be an admin. But I do want to push this group to eat some bibliographic dogfood
> and make this list machine-readable and useful.  Preferably JSON, but XML or RDF or whatever.
> And please include attributes like number and format of records.
> This is a tiny list right now, just provide a complete JSON or XML export at e.g.
> http://ckan.net/group/bibliographic/json
> http://ckan.net/group/bibliographic/xml

CKAN has a full RESTful API and search interface that talks JSON. See
<http://www.ckan.net/api/>

So, for example, to get all packages  from the bibliographic data
group you could do:

<http://ckan.net/api/search/package?groups=bibliographic>

To get it with all package attributes:

<http://ckan.net/api/search/package?groups=bibliographic&all_fields=1>

> Browsing around the CKAN site I see no API, no XML or JSON export, no feeds, nothing machine-readable.

I was about to say haven't you seen the "API" link the main navbar
when I double-checked about found it had disappeared! A quick check of
the codebase revealed the link had been removed 3 weeks ago during a
bit of routine refactoring (accidentally I suspect). Anyway it is now
back. (BTW google is your friend here as a google for "ckan api" seems
to work quite well!).

I also note there is a command line called datapkg which communicates
with CKAN via that REST API:

<http://knowledgeforge.net/ckan/doc/datapkg/>
<http://blog.okfn.org/2010/02/23/introducing-datapkg/>

> I did find the gzipped dump file http://www.ckan.net/dump/ which is updated daily.
> I would think the above features could be obtained by taking this daily dump and updating an image of it e.g.
> within delicious or bibsonomy which might provide further functionality, visibility  and API access.

As you will see, the good news is that this isn't necessary! Apologies
for the missing API link and let me know if there isn't anything else
I can help with.

Regards,

Rufus
-- 
Open Knowledge Foundation
Promoting Open Knowledge in a Digital Age
http://www.okfn.org/ - http://blog.okfn.org/