- From: Trish Whetzel <whetzel@pcbi.upenn.edu>
- Date: Wed, 10 Jan 2007 22:57:31 -0500 (EST)
- To: William Bug <William.Bug@DrexelMed.edu>
- cc: Tim Clark <twclark@nmr.mgh.harvard.edu>, Mark Wilkinson <markw@illuminae.com>, dirk.colaert@agfa.com, Daniel Rubin <rubin@med.stanford.edu>, w3c semweb hcls <public-semweb-lifesci@w3.org>, public-semweb-lifesci-request@w3.org
Hi Bill, Is the work that you mention below part of that for BIRN and is there a pointer to a policy? Also, have you found it necessary to be able to re-create the graph within the same version of the ontology on a day-to-day basis or is a version-to-version basis acceptable? Trish > As to Dirk's original point below, this is a requirement for OBO Foundry > ontologies - "the original URI should still point to the old term or concept, > even if it is deprecated" - though I take it not all BioPortal ontologies > will follow the OBO Foundry guidelines. > > There's also been a lot of work on ontology class metadata standards recently > - much focused within the OBI ontology development group - with a specific > eye toward helping to support the link between deprecated classes and those > classes to which the original semantic content covered by the defunct class > has been transferred. To fully represent this evolutionary graph, there will > likely be a need for some ruled-based formalism to deal with scenarios where > the deprecated class has not simply been decomposed into 2 or more newer > classes - or scenarios where multiple deprecated classes map to multiple > current classes. Even under the simple case where 1 entity --> 2 entities, > rules may be required to more fully and formally express the semantic > transference asserted by the ontological evolutionary graph. > > For the time being, we at least are working to include metadata properties to > make it possible to - in theory - reconstruct the ontology graph as it > existed on a particular date.
Received on Thursday, 11 January 2007 03:57:47 UTC