>...
> I always thought it would be interesting to distribute 
> indexes of the DOAJ content more widely.
>...

Fwiw:

The NSDL currently harvests metadata, crawls content, and indexes both for nine of the
seventeen DOAJ subjects.  We gather science, math, engineering, and technology related
DOAJ subjects:

Subject            (item counts)
---------------------------------
Technology and Engineering (15,248)
Physics and Astronomy (5,712)
Mathematics and Statistics (10,726)
Health Sciences (88,726)
Earth and Environmental Science (12,227)
Social Sciences (27,952)
Chemistry (12,089)
Biology and Life Sciences (45,987)
Agriculture and Food Sciences (21,923)


The NSDL search service is described here:

  http://ncore.nsdl.org/index.php?menu=services&submenu=services!search

The search UI and API are documented here: 

  http://wiki.nsdl.org/index.php/Community:Search

The above are maintained as separate collections, and the API allows for filtering by one
or more collection.

All NSDL metadata content is available via OAI-PMH here:
  http://ndr.nsdl.org/oai


-Tim

Timothy Cornwell, Programmer/Analyst
National Science Digital Library (http://nsdl.org)
301 College Avenue
Ithaca,  NY 14850
(607)255-3297



> -----Original Message-----
> From: Code for Libraries [mailto:[log in to unmask]] On 
> Behalf Of Eric Lease Morgan
> Sent: Tuesday, February 17, 2009 8:12 AM
> To: [log in to unmask]
> Subject: Re: [CODE4LIB] ticTOCs makes its data available to developers
> 
> On 2/17/09 6:47 AM, "Laurence Lockton" <[log in to unmask]> wrote:
> 
> > You might be interested to know then that Lund University 
> Libraries, the
> > people behind the DOAJ, have been doing this for years, since before
> > publisher RSS feeds and NGCs. They make it available to 
> other libraries too
> > (for a fee) with your holdings indexed so you can search 
> for full text
> > articles only. See
> > <http://www.lub.lu.se/en/search/information-about-elinlund.html>
> 
> 
> Very interesting. Thank you for bringing this to our 
> attention, and I'd like
> to comment on "make it available to other libraries" part.
> 
> I always thought it would be interesting to distribute 
> indexes of the DOAJ
> content more widely. More specifically:
> 
>   * create (Lucene) indexes for all the OAI DOAJ article sets
>   * distribute the indexes through something like BitTorrent
>   * provide a way to mix and merge the selected indexes
>   * provide tools to search the result
> 
> In such a way a library could select subsets apropos to their 
> collection
> development policy, and maybe integrate the whole thing into 
> their "NGC".
> There is no reason why the indexes would have to be limited 
> to OAI DOAJ
> sets, but it could also include other OAI-accessible content, 
> or content
> extracted from mirrored journal literature.
> 
> In short, library as bibliographic index publisher.
> 
> -- 
> Eric Lease Morgan