- From: LYNN,JAMES (HP-USA,ex1) <james.lynn@hp.com>
- Date: Tue, 13 May 2003 06:51:53 -0700
- To: "'Seth Ladd'" <seth@brivo.net>, www-rdf-interest@w3.org
All of these approaches (Sowa's, MKR, reification) seem to make sense. I don't feel a particularly strong opposition to any of them, but putting myself in the role of the typical semantic web programmer (in the not too distant future?) I might say: 1. I need a way to make a simple statement using RDF and OWL that makes my meaning clear as pertains to its truth at the time I make the assertion or at some other specific time or duration. This statement needs to be clear to anyone who accesses my site with "standard tools", not a proprietary solution. 2. Reification is standard enough and would be understood by any decent RDF parser, but that's too much work every time I want to make a statement. Shouldn't RDF provide something that doesn't make my graphs look so cluttered. Again, I don't necessarily believe this is the state of affairs. I guess I'm just playing Devil's Advocate in an effort to elicit opinions on this matter. James -----Original Message----- From: Seth Ladd [mailto:seth@brivo.net] Sent: Tuesday, May 13, 2003 8:55 AM To: www-rdf-interest@w3.org Subject: RE: rdfs:class and rdfs:resource On Mon, 2003-05-12 at 16:16, LYNN,JAMES (HP-USA,ex1) wrote: > This would certainly accomplish the task. I guess my question is - is there > a need for a predifined property (or three?) to express a point in time or a > duration of time in a simply manner. In other words, how common is this need > - enought to warrant an addition to the spec? I believe the Dublin Core has done some work on time and date modeling. Also, you might want to check out what the RDF Calendar people are doing. They are working on the same problem. As for adding to RDF itself, I don't think that will happen. What would you assign the time point to? The predicate? The binding of the predicate to the object? How about the whole statement? Using reification, you can assign other statements to statements. So, one could assign properties such as "The statement was asserted at time t" or "The statement is only good for t to t2", but since reification is talking about some stating, and not a statement, you're really not binding these time oriented statements to the actual primary statement. I think I lost myself there... Seth
Received on Tuesday, 13 May 2003 09:52:10 UTC