- From: Graham Klyne <GK@ninebynine.org>
- Date: Fri, 28 Dec 2001 11:27:18 +0000
- To: Dan Brickley <danbri@w3.org>
- Cc: <www-rdf-calendar@w3.org>
This is a difficult problem, and I wouldn't look for a quick-fix solution. I recently had occasion to update an old Internet Draft by Chris Newman that profiled ISO8601 date/time formats [1] for use in IETF protocols. The original draft (now long expired) included a considerable amount of material on timezones and their use for indicating schedule events (while what we actually wanted from this draft was a simple format for expressing timestamps). As part of this exercise, Chris passed me a copy of his voluminous correspondence on the previous draft; it was interesting to note that the vast majority of this correspondence was about the matter of timezones and their relation to scheduled events. I got the impression that a complete solution is not really feasible because it would attempt to codify localized social behaviour, past present and future, in a global protocol. I haven't noticed anything about specifying time-offset details separately from timezone identification -- that seems like a reasonable idea, but I tend to think that it might as well be specified as two distinct values (i.e. separate XML elements or RDF properties). #g -- [1] http://search.ietf.org/internet-drafts/draft-ietf-impp-datetime-05.txt At 11:45 AM 12/23/01 -0500, Dan Brickley wrote: >forwarded from the xmlschema-dev list... > > http://lists.w3.org/Archives/Public/xmlschema-dev/ > http://lists.w3.org/Archives/Public/xmlschema-dev/2001Dec/0233.html > >Can anyone advise? (crossposting between these two lists seems a >reasonable thing to do, btw, if anyone has an answer!). > >cheers, > >Dan > >ps. has anyone tried to use DAML+OIL to define membership rules for >classes of repeating events? see recent W3C Note(s) >http://www.w3.org/TR/daml+oil-reference and nearby... > >---------- Forwarded message ---------- >Date: Sun, 23 Dec 2001 16:32:43 +0000 (GMT) >From: s.livingstone@btinternet.com >To: xmlschema-dev@w3.org >Subject: XML Timezones >Resent-Date: Sun, 23 Dec 2001 11:33:11 -0500 (EST) >Resent-From: xmlschema-dev@w3.org > >I am writing a Windows Scheduling Service in C# that will use data in XML >format to determine Scheduling information. This same file will be >replicated globally and hence will be subject to timezone changes and >daylight saving time. > >[my timezone values are just examples - they may not be correct!!] > >I was originally using the XSD dateTime data type to determine the date and >time to perform a task, but this is limited when applied globally, because >the offset (timezone) is based on local time and as soon as you replicate to >some other timezone, the local time changes and so your scheduling >information goes nuts. > >An example... > >We have a Scheduled item that is to be done on July 16th, 2002 at 2PM >Scottish time. >This is simple to express as July 16th, 2002 at 2PM GMT (summer has no >daylight saving time). >According to the XML Schema stanard >(http://www.w3.org/TR/xmlschema-2/#dateTime) which uses the format >"CCYY:MM:DDTHH:MMZ" >this could be written as : >"2002:07:16T14:00Z" > >However, this standard format doesn't work too great because in the winter >the local time in Scotland changes to 1 hour less than UTC and so the >scheduled time should change to "2002:07:16T13:00Z" or 1PM GMT, but how do >you know this from the above fomat? Sure, it tells you the date, so you can >work out whether Daylight Saving Time applies, but you don't know what >timezone it refers to - it is a bad mistake to assume the local timezone, >because if i moved the Schedule information to Santiago, Chile and we >assumed the local settings to determine the offset then on that date the UTC >time would be "2002:07:16T08:00Z" - because here we are 8 hours behind GMT >on that date. > >There needs to be some way of stating the target timezone so that the >correct Daylight settings relative to UTC can be resolved. Something like >this... > >"CCYY:MM:DDTHH:MM:SSZGXXXX" > >where at the end, GXXXX specifies the locale identifier of the originally >scheduled time. > >Does that make any sense?? Has there been talk of this before... > >I never really thought much about it until i had to use it in my app!! > >Cheers, >Steven >Author "XML Application Development with MSXML 4.0" >www.deltabis.com ------------ Graham Klyne GK@NineByNine.org
Received on Friday, 28 December 2001 18:47:46 UTC