Re: relationship to iCalendar

Hi Martijn,

> 
> Hrm. what's wrong with sticking close to iCalendar? I'm working on
> Net::ICal, a perl implementation of rfc2445, and I was thinking
> of making it capable of outputting rdf as specified by your schema,
> so we can use xsl to convert it to html/other xml formats. But
> if you're going to move away from iCalendar, I'd have to come up
> with my own xml format.

We fully intend that documents written in rdf-cal format (or whatever we
decide to call it) will be mappable to iCalendar format and
vice-versa. 

In terms of differences, what's most likely to happen is that there will
be several different parts to the rdf-cal schema, with different
namespaces, representing various parts of the iCalendar rfc 2445. So
this is a trivial way in which they will differ. I am also wondering
whether on close inspection some of the names used in iCalendar rfc 2445
map well on to the classes and proprties they need to become in
RDF. Also the capitalisation of names in iCalendar will probably
change in rdf-cal because the convention in RDF is to give classes a
first capital letter and have properties in lower case. These again are
all fairly trivial.

More fundamental is the difficulty that iCalendar may not give us all
that we may want from descriptions of events. For example, in iCalendar
you have a VEVENT which has DTSTART and a DTEND. Now it maybe that if
the VEVENT is a meeting, libby has to leave early. This implies that she
was an ATTENDEE at the meeting for only part of the time of the duration
of the meeting. We can't express this in iCalendar as it stands, but I
think it's an important aspect of archiving event descriptions. It's
also something SKIcal have been looking into. See a preliminary (and
partial) rdf calendar issues list at

http://ilrt.org/discovery/2001/07/cal-issues/

But this is an addition, and so should not affect translation to and
from the iCalendar format. It is possible that we will find other
similar issues as we work through examples for each property and class.

> 
> Another question, why are <ical:VEVENT>s wrapped in a <ical::VEVENT-PROP>?
> From the example data you have
> (http://ilrt.org/discovery/2001/06/content/swws2001-07-30.rdf) it looks
> like it's just a gratuitous extra tag.

This is part of the way RDF works. The striped syntax for RDF means that
you have to have class - property - class - property. ical:CALENDAR is a
class and ical:VEVENT is a class so you need a property in
between.  Perhaps I could have given it a better name though.

cheers,

Libby

Received on Monday, 13 August 2001 11:29:32 UTC