reconsidering timezones in light of hCalendar and CALSIFY

In the Sep 2005 RDF calendar note, we have a section on shop
hours and timezones...

[[
While those rules are an accurate model of the timezone in New York at
least as far back as 1970, the Energy Policy Act of 2005 is intended to
change them in 2006. While the Olson database is likely to reflect these
changes in due course, the copies in all the iCalendar data out there
will fail to accurately represent the timezone rules for New York.

One approach, exemplified by the datetime design pattern in the
microformats community, is to not use iCalendar timezones, but only UTC
datesdelt.
]]
 -- http://www.w3.org/TR/rdfcal/#L21805


In that design pattern page...
  http://microformats.org/wiki/datetime-design-pattern

you'll find that I was convinced...

Aug 18 16:05:57 <DanC>	        Tantek's right, Kragen; iCalendar looks
like it solves the local timezone
                                problem but doesn't.

Timezone related interoperability problems are all over the place.
I just heard about one with google calendar today in #swig discussion.

In my own work on synchronizing from my sidekick/hiptop PDA
with iCalendar tools, I found that the hiptop
  - stores start dates of recurring events in Z time syntax
  - records a timezone
I implemented the conversion to the iCalendar way of doing
things, which is to store local time and the timezone, but
then I found that events were off by an hour when I looked
at them with iCalendar tools.
I learned that the timezone that gets recorded is not really the one
relevant to the event; it's just the timezone that the gizmo
was in when the event was recorded. I was in Helsinki when
we scheduled the RDF Data Access WG meetings, so that's
the timezone that got recorded.

Most recently, I re-architected my PDA sync stuff; the new
version is
 http://dev.w3.org/cvsweb/2001/palmagent/hipsrv.py
and in there, I implemented converting the time of recurring
events to local time and then just stripping off the timezone.
This is less expressive; it doesn't allow my colleagues in
other parts of the world to see when my recurring events happen
in their timezones. But it does work per the specs.

By the way, I note the new(ish) IETF CALSIFY WG has a draft where
RECUR is taken out.

Basic Internet Calendaring and Scheduling Core Object Specification
                           (iCalendar Basic)
December 24, 2005
http://www.ietf.org/internet-drafts/draft-royer-ical-basic-04.txt


There's also a draft on a timezone registry linked
from the WG materials. http://ietf.webdav.org/calsify/
I haven't read it closely.

The more I think about it, the more I hope that the semantic
mediawiki project turns Wikipedia into a home for timezone data.
http://meta.wikimedia.org/wiki/Semantic_MediaWiki


There's also an early draft on timezones by Jeff Pan,
being kicked around in the SemWeb Best Practices and Deployment WG.

Time Zone Resource in OWL 
W3C Editor's Draft 20 September 2005
http://www.isi.edu/~pan/SWBP/time-zone-note/time-zone-note.html

I have skimmed it. I hope to look into it more closely.


I intend to take another pass at our RDF calendar tests with
all this in mind.
  http://www.w3.org/2002/12/cal/#dev

and I hope to think more about connecting hCalendar to RDF
Calendar with GRDDL while I'm at it, updating the transformation
as I go.
http://www.w3.org/2002/12/cal/glean-hcal.xsl

I'm working with Ryan King on microformats tests. I have
been having so much fun with this new mercurial/hg distributed
version control system that I convinced them to use it.

http://hg.microformats.org/tests

See:
22 Mar: using mercurial/hg to share microformats code, tests
http://microformats.org/discuss/mail/microformats-dev/2006-March/000070.html

The current version of the hCalendar tests, for example, is:
http://hg.microformats.org/tests?cmd=manifest;manifest=9a7237c61adbf4a96ff840403fe638aa9eaee846;path=/hcalendar/

Unfortunately, hgweb doesn't (yet) support browsing in such a
way that you can link to "the up-to-date /hcalendar/ directory".


Ooh... I see somebody else is working on hCalendar test
materials...
  From: Mark Mansour <mark@lifelint.com>
  Date: Wed, 8 Feb 2006 21:47:01 +1100
  http://lists.w3.org/Archives/Public/www-rdf-calendar/2006Feb/0000.html
  -> http://www.lifelint.com/blog/2006/02/07/hcalendar-test-cases/

I think I missed that somehow.


For (my?) reference,I put some of this todo stuff in
  http://esw.w3.org/topic/DanConnolly
I'm thinking about moving it around to places in the wiki
that are more shared, like...
  http://esw.w3.org/topic/MicroModels
  http://esw.w3.org/topic/RdfCalendar

Oh... speaking of wiki stuff... if you haven't seen
the SPARQL calendar demo, please do check it out.
It's a great example of what you can do when
you mix calendar data with foaf:interest data and such.
  http://esw.w3.org/topic/SparqlCalendarDemo

-- 
Dan Connolly, W3C http://www.w3.org/People/Connolly/
D3C2 887B 0F92 6005 C541  0875 0F91 96DE 6E52 C29E

Received on Friday, 14 April 2006 06:25:53 UTC