Re: [syndication] envelopes vs. payloads, calsch,

[Dan Brickley <daniel.brickley@bristol.ac.uk>]

On Sun, 9 Jul 2000, Aaron Swartz wrote:

> Phil Wolff <pwolff@bigfoot.com> wrote:
> 
> > Do you think there is room in the next syndication release to further
> > separate envelope from payload?
> 
> I think this is a must for whatever we do next. This ties in very well with
> the modularization stuff -- if we come up with an envelope system, it will
> be extremely helpful in moving syndication forward.
> 
> > Can the spec allow just about anything to be syndicated, beyond news
> > items? 
> 
> We should allow anything to be syndicated. Once we agree on a syndication
> envelope, to syndicate new types of data would just require an XML->HTML
> converter for each new type. This would make it really easy to syndicate
> anything.

This could probably be done with XSLT plus some additional metadata
(taxonomy etc) describing what the XSLTs do...


> > I agree about syndicating events; they are a natural. About the
> > Calendaring and Scheduling standard, (1), I'm not sure if the ietf
> > calsch working group ever finished, but it appears so.
> 
> Looking at the mailing list, it looks like there's still active discussion
> going on. They just posted an Action Item listing yesterday:
> 
> http://www.imc.org/ietf-calendar/mail-archive/msg05128.html

For info: there was also some discussion of an RDF representation of
iCalendar (and SKiCal, which sits on top of iCalendar) in the RDF Interest
Group[1], including some examples put together by Jonas Liljegren and
Greg FitzPatrick[2], started off by an initial note from Tim Berners-Lee
proposing an XML/RDF representation of iCalendar[3]. From a rather
different perspect (metadata mappings,  dublin core interoperability
etc -- eg see draft at [4]) I've become convinced that a common
vocabulary for event descriptions is an important goal. Characterising
events (eg. meetings, publication events, document processing events
etc) is a useful strategy for describing stuff since events don't change
over time, which means that event-descriptions don't need to be packaged
with a whole load of 'true from; true until' context. This contrasts
interestingly with other varieties of metadata (eg. document
descriptions) where the object of interest changes over time.

I guess 'events' in the sense of this thread is somewhat more restricted 
(eg. TV shows, concerts...?) than the fairly generic model I'm inclined
towards. That said, I'd be interested to see an attempt at scoping the
kinds of event descriptions that might usefully be interchanged with
RSS++, since 'event' is rather broad ('things that happen'?).

Dan

ps. check out this fun paper on SKICAL and event modeling[5] if the idea
of event syndication over RSS appeals...

[1] http://www.w3.org/RDF/Interest/
[2] http://lists.w3.org/Archives/Public/www-rdf-interest/2000Feb/0041.html
[3] http://lists.w3.org/Archives/Public/www-rdf-interest/2000Jan/0089.html
    http://www.w3.org/2000/01/foo
[4] http://www.ilrt.bris.ac.uk/discovery/harmony/docs/abc/abc_draft.html
[5] http://www.metamatrix.se/presentationer/html/inet.htm