[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

Re: limits of the generic

• To: John Cowan <jcowan@r...>
• Subject: Re: limits of the generic
• From: Jeni Tennison <jeni@j...>
• Date: Sat, 28 Sep 2002 23:21:14 +0100
• Cc: dareo@m... (Dare Obasanjo), uche.ogbuji@f..., jonathan.robie@d... (Jonathan Robie), <xml-dev@l...>
• In-reply-to: <200209282114.RAA07578@m...>
• Organization: Jeni Tennison Consulting Ltd
• References: <200209282114.RAA07578@m...>
• Reply-to: Jeni Tennison <jeni@j...>

John,

>> For example, I think that "What is 2002-09-28 plus a year and a
>> day?" is also a simple question with a well-defined answer; I don't
>> think it should be an error to try to ask it.
>
> Ah, but what is 2004-02-29 plus a year and a day?
>
> Time labeling is an impossible compromise between incommensurable
> astronomical cycles, and can never be really clean.

Sure, that means sometimes it has to be a bit arbitrary, but it
doesn't mean that it's impossible to create algorithms that give
consistent (if not absolutely "correct") results.

The algorithm in the W3C XML Schema Datatypes Recommendation adds the
year/month component first (which would give 2005-02-29), then "pins"
the result of that to give a legal date (i.e. 2005-02-28), and then
adds the day and time components of the duration (to give the final
answer of 2005-03-01).

Cheers,

Jeni

---
Jeni Tennison
http://www.jenitennison.com/

• Follow-Ups:
  • Re: limits of the generic
    • From: John Cowan <jcowan@r...>

• References:
  • Re: limits of the generic
    • From: John Cowan <jcowan@r...>

• Prev by Date: Re: limits of the generic
• Next by Date: Re: limits of the generic
• Previous by thread: Re: limits of the generic
• Next by thread: Re: limits of the generic
• Index(es):
  • Date
  • Thread