Re: More Work on iCalendar RDF Schema

hi Aaron. Many thanks for having a look at this.

I should perhaps clarify at least what my intentions were in writing the
iCalendar in RDF. In particular I wanted to stay as faithful to
iCalendar as I possibly could, even if this meant that I ignored
potentially better ways of doing it. I figured that once we had a
relatively clear iCalendar model in RDF, this could form the basis of
further work to perhaps separate out different parts into diffeent
namespaces, or possibly make simplifications, or hook into different
existing schemas. So as far as RDF calendar taskforce goes, we shouldn't
see something like the hybrid calendar as the end of the work but just
an attempt at a solid basis for the start. A next step is to start
comparing other similar schemas using examples. 

> 
> Why THEANNOYINGNAMES? I have difficulty distinguishing between 
> Classes and properties, and there is no good reason for keeping 
> them in ALLCAPS.
> 

This is down to me - Micheal used lower case caps/no caps for classes
and properties as is more usual. I started off doing that too. However
the iCalendar rfc is such a complex document that I wanted to minimise
the number of mistakes that crept in in transcribing it to RDF. For that
reason it is better to be consistent with the iCalendar usage of
capitals even though it's not vrey aesthetically pleasing. We can always
change this later.


> Why the funky dates? Can't we just use W3CDTF like everyone else?
> 

This is consistent with the iCalendar RFC usage. I didn't want to
arbitrarily change anything at this point.


> Why duplicate ical:DESCRIPTION and dc:description ?
> 

Michael's reply to this is a good one.

> As a test, I tried transcribing 
> http://ilrt.org/discovery/2001/06/content/track2.rdf into N3 
> with the simplifications I propose above. Look at how much 
> simpler it is:
> 

Aaron, it's confusing to compare one example in XML and another in N3,
since N3 looks much less complex anyway. One complexifying factor with
the example I made are the timezone objects (which I'm not sure I have
right). One thing that could be done here is a fairly authoritative
conversion of some of the data at www.dateandtime.com (say) into RDF, so
people would not have to define their own timezones, but could just
refer to previously defined ones. 

I don't see that your idea is so much simpler. It seems that dates are
now strings rather than objects (my N3 isn't so great). This is ruled
out if we want to use Timezones. Otherwise changing ical:description to
dc:description isn't such a big change. Have I missed something else...?

thanks again Aaron - I'm very grateful for your hard work.

Libby


> @prefix util: <http://ilrt.org/discovery/2001/06/schemas/ical-util#> .
> @prefix foaf: <http://xmlns.com/foaf/0.1/> .
> @prefix dc: <http://purl.org/dc/elements/1.1/> .
> @prefix event: <http://test.example.com/events/> .
> @prefix speaker: <http://test.example.com/speakers/> .
> @prefix : <http://ilrt.org/discovery/2001/06/schemas/ical-
> full/hybrid.rdf#> .
> 
> [ a :VCalendar ;
> 	dc:description "2" ;
> 	:vEventProp event:137 .
> ] .
> 
> event:137  a :VEvent ;
> 	:dtStart "20010525T141500" ;
> 	:dtEnd "20010525T141500" ;
> 	:description "The State of XML" ;
> 	util:presentation [ a foaf:Document
> 		dc:creator speaker:40  ;
> 		dc:title "The State of XML" ;
> 		dc:description "This talk surveys the current state of 
> XML - from standards and standards-makers through to XML's role 
> in the architecture of the next generation of computer systems. 
> XML's influence is being felt ever more broadly: not only on the 
> Web and in business, but also in more unexpected places such as 
> operating systems and embedded devices. Aided by an energetic, 
> creative and sometimes controversial community, the state of XML 
> is looking better than ever." ] .
> 
> speaker:40 a foaf:Person, :CalAddress .
> 
> --
> [ "Aaron Swartz" ; <mailto:me@aaronsw.com> ; <http://www.aaronsw.com/> ]
>

Received on Friday, 22 June 2001 06:35:26 UTC