Re: [BioPAX-discuss] LSID Best practices...

EJ>I think the DOI system got this right; they specify that an identifier 
such
EJ>as doi:10.1000/182 can be resolved using http://xyz/10.1000/182, where 
xyz
EJ>is your favorite resolver server.

EJ>Any chance to see this being standardized, as a light weight 
alternative to
EJ>the web service resolution mechanism?


I think it could be standardized if we could decide what it resolved to. 
Data or metadata (and who's). Are we more interested in retrieving RDF 
describing it or the bits? Perhaps some variation in the URL could specify 
that..

http://xyz/data/urn:lsid...  and http://xyz/metadata/urn:lsid..  or 
http://xyz/3rdpartymetadata/urn:lsid... etc?

Create verbs to describe what you want the proxy to return?

I am not sure that one should strip out the urn:lsid text otherwise how 
would ones software recognize it as such in a URL? Is this even important?

Kindest regards, Sean

--
Sean Martin
IBM Corp.





Eric Jain <Eric.Jain@isb-sib.ch> 
Sent by: public-semweb-lifesci-request@w3.org
04/07/2005 11:42 AM

To
Eric.Neumann@sanofi-aventis.com
cc
public-semweb-lifesci@w3.org
Subject
Re: [BioPAX-discuss]  LSID Best practices...







Eric.Neumann@sanofi-aventis.com wrote:
> http://lsid.biopathways.org/resolver/data/urn:lsid:
> ncbi.nlm.nih.gov.lsid.biopathways.org:genbank_gi:5851672

Really useful! However, someone else may use

http://xyz/resolve?lsid=
urn:lsid:ncbi.nlm.nih.gov.lsid.biopathways.org:genbank_gi:5851672

or

http://xyz/
ncbi.nlm.nih.gov.lsid.biopathways.org/genbank_gi/5851672

or who knows what else...

I think the DOI system got this right; they specify that an identifier 
such
as doi:10.1000/182 can be resolved using http://xyz/10.1000/182, where xyz
is your favorite resolver server.

Any chance to see this being standardized, as a light weight alternative 
to
the web service resolution mechanism?