- From: Geoff Chappell <gchappell@intellidimension.com>
- Date: Thu, 12 May 2005 13:14:49 -0400
- To: "'Eric Jain'" <Eric.Jain@isb-sib.ch>
- Cc: <public-semweb-lifesci@w3.org>
> -----Original Message----- > From: Eric Jain [mailto:Eric.Jain@isb-sib.ch] > Sent: Thursday, May 12, 2005 11:54 AM > To: Geoff Chappell > Cc: public-semweb-lifesci@w3.org > Subject: Re: Uniprot RDF in RDF Gateway [...] > I've been reluctant to include explicit data types in the instance data, > because they increase the data volume considerably, and can be inferred > from the ontology anyway (as you do). Alternatively, a query system might > even decide to detect basic data types (number, boolean, date, string) on > its own. I imagined this was probably the case. It might be worth highlighting for folks on your site, since it's not strictly legal (i.e. the value spaces of plain literals and all of the types used -- except xsd:string as Pat Hayes pointed out -- are disjoint). It certainly seems like an appropriate/pragmatic approach nonetheless considering the size of the files. > Regarding the date issue: Where the ontology specifies an xsd:date, you > may > find any of '2005-05-12' (xsd:date), '2005-05' (xsd:gYearMonth) or '2005' > (xsd:gYear). I changed the 'date' property back to a plain literal for the > next release, and until I find a better solution. Yeah, it's unfortunate there isn't a meaningful common superclass of those (since the they all describe ranges on the timeline). I suppose you could make the range the union of the three types. -Geoff
Received on Thursday, 12 May 2005 17:15:17 UTC