Re: Ontology Working Group Proposal Draft

Agree that this will be critical, esp. with genomic-phenotypic links. 
Clinicians will want to ask, "what are the possible variants/druggable 
pathways associated with my patient's clinical presentation" ?, but with 
limited (to say the least) time to investigate/validate.  Describing the 
phenotype in RDF and then sending it "out there" into medicocyberspace 
will be standard operating procedure.  One of our jobs will be to conceive 
and build the "out there" in terms of service 
advertising/requirements/interfaces.  Web services should also help guide 
our thinking in terms of use cases.

JJ Michon, MD, MS
Duke University




John Madden <john.madden@duke.edu> 
Sent by: public-semweb-lifesci-request@w3.org
02/10/2006 07:58 AM

To
public-semweb-lifesci@w3.org
cc

Subject
Re: Ontology Working Group Proposal Draft







Hey Vipul,

There's a topic I'm not sure whether it is reflected in the draft 
(I'm away from my desktop).

That's the issue of helping to collect/develop some standard web 
service definitions for "ontology services".

Of course there are lots of wonderful "terminology server" API's out 
there going back 20 years. Some of these have been considered 
standards, others were more like demos. But I don't know if anybody 
has yet tried to think about broadly applicable, web-standard WSDL 
work, and I think a set of abstract classes expressed in web service 
language, that people trying to implement web service interface could 
inherit from, would really be a way of promoting webification of 
ontologies.

(Also talking with danbri about this would be a good idea, I think.)

John