RE: Referencing time resources

Hi Libby:

Whoops, sorry! - maybe it was a trick question. ;) But if I wanted to
unambiguously refer to a time resource as a first-class object then don't I
have to accord it a URI, e.g. (just for example)

	info:date/2003-01-15T18:00:00Z

(hate to do that but it is one possibility).

The problem with bNodes, of course, is merging graphs. Even if all the
properties can be matched, the bNode ID is local to the graph and so how can
resources from different graphs (models, whatever) be compared?

Tony


-----Original Message-----
From: Libby Miller [mailto:Libby.Miller@bristol.ac.uk]
Sent: 17 October 2003 14:31
To: Hammond, Tony (ELSLON)
Cc: www-rdf-interest@w3.org
Subject: Re: Referencing time resources



How do you mean Tony? time as in date-time?

For Dates, dateTimes, Times, we've been using an RDF version of
iCalendar (developed by the RDF interest group calendaring taskforce):

http://www.w3.org/2002/12/cal
http://esw.w3.org/topic/RdfCalendarDocumentation
http://www.w3.org/2002/12/cal/test/

e.g.

<Vevent>
    <dtstart rdf:parseType='Resource'>
      <dateTime>2003-01-15T18:00:00</dateTime>
      <tzid>/softwarestudio.org/Olson_20011030_5/Europe/London</tzid>
    </dtstart>
</Vevent>

(Notice the timezone reference, which in iCalendar (RFC 2445) has to
reference a timezone in the same file. Notice also that we use bnodes).

There's a problem with this approach:

http://ilrt.org/discovery/chatlogs/rdfig/2003-08-20.html#T16-57-54

There's also a problem with the W3C datetime formats note's approach to
timezones (http://www.w3.org/TR/NOTE-datetime) in the sense that it is not
possible to be completely accurate about date-times using formats like:

1994-11-05T08:15:30-05:00

in the case of some timezone changeover periods.

There's lots more information on the www-rdf-calendar@w3.org list
(http://lists.w3.org/Archives/Public/www-rdf-calendar/)

There is a guy generating urls from times and dates, but I can't find it
offhand, sorry. I'll try again if that's what you were after.

or did you mean something else? :)

Libby


On Fri, 17 Oct 2003, Hammond, Tony (ELSLON) wrote:

>
> Hi:
>
> Please excuse general level of ignorance, I'm just curious, but how does
one
> reference time resources - unambiguously (i.e. not by bNodes)?
>
> Thanks,
> Tony
>
>
>

Received on Friday, 17 October 2003 10:15:22 UTC