- From: Dan Brickley <danbri@w3.org>
- Date: Mon, 17 Mar 2003 16:06:52 -0500
- To: Jonathan Borden <jonathan@openhealth.org>
- Cc: "Roger L. Costello" <costello@mitre.org>, www-rdf-logic@w3.org
* Jonathan Borden <jonathan@openhealth.org> [2003-03-17 16:02-0500] > Dan Brickley wrote: > > > > * Jonathan Borden <jonathan@openhealth.org> [2003-03-17 15:31-0500] > ..> > > > > ex:Speeder foo:fingerprint urn:uuid:12345... . > > > ex:Robber foo:fingerprint urn:uuid:12345... . > > > > > > foo:fingerprint a owl:InverseFunctionalProperty . > > > > > > => > > > > > > ex:Speeder owl:sameIndividualAs ex:Robber . > > > > Actually there is a property in FOAF that fits this usecase. Not > > 'fingerprint', but foaf:dnaChecksum. It was intended as a JOKE! but > > also a warning of technology potential... (see also various concerns > > many have with TIA, eg. http://www.warblogging.com/tia/). > > > > See http://xmlns.com/foaf/0.1/ (view src) for > > foaf:dnaChecksum definition. I didn't actually specify an algorithm or > > even concrete syntax, needless hopefully to say... > > > > Actually there is some interest and progress on using OWL for bioinformatics > i.e. representing a gene sequence as a typed RDF description (Jim Hendler > has done some work on this) but ... even that is frought with issues, for > example, truly *identical twins* are different individuals that might have > the same dnaChecksum... > > nice try though :-))) Good point! I worried more about collisions in the hashing algorithm making it not strictly inverse-functional... cheers, dan
Received on Monday, 17 March 2003 16:06:56 UTC