Re: [EP-underground] Fwd: BOUNCE [email protected]: Non-member submission from [Thomas Krichel <[email protected]>]

"Dublin Core is ten times better than nothing"

oai_dc is fine for "resource discovery" - people searching for something
accross several archives can type in a word and find some matches.

The problem with high quality data is that it costs money - as you have to
train your people more strictly and proof read it more. Also if you have 
people entering the metadata who feel they've got something better to do
than learn your (seemingly, to them) over fussy field definitions.

EPrints2 can support other datasets than unqualified dublin core (oai_dc)
but doing so is probably more effort than many people can be bothered with.

I'm considering adding a "filter" option to the OAI system as I have to 
set up an archive which may contain records with either no document or
no document publically available. Exporting these via OAI is just annoying.

On Tue, Jun 18, 2002 at 08:39:27PM +0100, Christopher Gutteridge wrote:
> The mailing list bounced this address (It's now being strict about
> posters being on the list, to reduce spam)
> 
> Aplogies to Mr Krichel - I've added [email protected] to the list of
> people allowed to post.
> 
> ----- Forwarded message from [email protected] -----
> From: Thomas Krichel <[email protected]>
> 
>   Roy Tennant writes
> 
> > Here it is at the EconPapers archive in Sweden:
> > http://econpapers.hhs.se/paper/cdloplwec/38.htm
> > 
> > Here it is in the WoPEc archive in the UK:
> > http://netec.mcc.ac.uk/WoPEc/data/Papers//cdloplwec38.html
> 
>   There URLs come through RePEc project, see http://repec.org.
>   RePEc provides a few more services of this style. They all
>   deliver to a common pool of logs that has abstract view
>   and downloading statistics, see http://logec.repec.org
>   This can show authors with figures at hand how well they 
>   are doing. Most RePEc services contribute to this common
>   pool.  A set of current awareness lists that are filtered by subject
>   editors, see http://nep.repec.org, does not do this at the moment.
> 
> > Formerly, we talked about the possibilities of OAI in the abstract to
> > our faculty. Now we can demonstrate it in reality. That, as you might
> > imagine, is a powerful thing.
> 
>   Yes, but David Cahill is right that you can not build many
>   good services with the oai_dc metadata. For your data, we
>   rely on massaging cdl data into RePEc's internal format to
>   deliver the services that we do.
> 
>   We really need better data and better metadata. 
> 
> 
>   Cheers,
> 
>   Thomas Krichel                                   mailto:[email protected]
>                                               http://openlib.org/home/krichel
>   CORRECT private phone: 1-718-507-1117   RePEc:per:1965-06-05:thomas_krichel
> 
> ----- End forwarded message -----

-- 

 Christopher Gutteridge                   [email protected]
 ePrints2 Coder, Support and Stuff        +44 23 8059 4833