Re: places and lists of coordinates [was: priority bug, to libbyand dan]

On Thu, 2004-04-08 at 19:31, Masahide Kanzaki wrote:
> At 0:22 PM -0500 04.4.8, Dan Connolly wrote:
> >On Thu, 2004-04-08 at 10:40, Masahide Kanzaki wrote:
> >[...]
> >
> >> So, I think it's not good idea to define ical:geo as the list of floats.
> >
> >I can see your point, but I'm not yet convinced. What would you prefer?
> >
> >> Remember, we also want to use RDFical vocabulary with other vocab, such as
> >> RSS, FOAF or even XHTML. Strict round trip .ics <-> RDFical is only
> >> relevant when RDFical is generated from iCalendar, and doesn't make much
> >> sense when the vocab is used in, say, FOAF file.
> >>
> >> Wouldn't it be enough to say something like 'when converted from .ics,
> >> ical:geo should be expressed as the list of floats so that strict round
> >> trip is possible' ?
> >
> >I don't think so... not without the sort of confusion discussed in
> >  http://esw.w3.org/topic/ThingsVersusTheirNames
> >
> >Either the range of ical:geo is a list of floats
> >or a place. RFC2445 says it's a list of floats. I wouldn't have
> >done it that way, but they did. Similarly, ical:location takes
> >a string value.
> 
> I agree. How about assigning place (spatialThing) as the range of ical:geo
> ?

Well, I don't see how to derive that from RFC2445.

>  This makes it clear that ical:geo should be used to describe place as a
> resource, not its name. Both list of floats and geo:lat/long can be
> properties of that resource. And, notes on round trip stay the same...

I don't understand. If the list of floats is a property of the place,
then it looks like:

 :event1
   ical:geo [
      :what_property_here (23.23 42.1);
      geo:lat 23.23;
      geo:lon 42.1;
   ].


So what property would you use for :what_property_here ?

> >It's straightforward to relate the place to the list of floats ala:
> >
> >{ ?E ical:geo (?LAT ?LONG) } <=>
> >  { ?E cyc:eventOccursAt [ geo:lat ?LAT; geo:lon ?LONG ] }.
> >
> >and it's straightforward to relate a place to its name:
> >
> >{ ?E ical:location ?PLACENAME } <=>
> >  { ?E cyc:eventOccursAt [ rdfs:label ?PLACENAME ] }.
> 
> Sure. It's simple if we give up to use ical:geo to describe place with
> other vocabularies. But is this what most of us desire ?

I don't understand what you mean. RFC2445 says that ical:geo relates
an event to a list of floats that give the latitude and longitude
of a place where the event happens. If you want a property
that relates an event to a place, you need a different property, no?

Why do we need... or even want... to use the ical vocabulary to
relate events to places? Why not cyc, SUMO, or a new vocabulary?

We haven't even found any iCalendar software that produces
nor consumes GEO: in any non-trivial way.

> 
> regards,
-- 
Dan Connolly, W3C http://www.w3.org/People/Connolly/
see you at the WWW2004 in NY 17-22 May?

Received on Friday, 9 April 2004 10:34:07 UTC