Re: Date/time tag??

• Mail folder: WWW Talk Jul-Oct 1993
• Next message: frans van hoesel: "docs vs browsers"
• Previous message: Terry Allen: "OL numbering and presentation"
• In-reply-to: Erik Ostrom: "Re: Date/time tag?? "

Message-id: <9308022233.AA02702@nada.kth.se>
To: www-talk@nxoc01.cern.ch
Cc: psv@nada.kth.se
Subject: Re: Date/time tag?? 
In-reply-to: <9308021822.AA09795@iicm.tu-graz.ac.at> 
Date: Tue, 03 Aug 1993 00:33:28 +0200
From: Peter Svanberg <psv@nada.kth.se>
Status: RO

mvanheyn@cs.indiana.edu:
>    - Dates frequently represent bounded intervals rather than a single
>      point in time.  Should there be a way to represent "from 3:00 until
>      8:00 PDT on 30 Nov 95" in these?
> 
>    - Dates vary in terms of their quantity of significant "terms"; e.g.
>      "November, 1993" should be representable and it's certainly not the
>      same as "1 Nov 1993 00:00:00 GMT".
> 

eostrom@fiicmds04.tu-graz.ac.at (Erik Ostrom):
> So, suppose we allow for leaving out sections:
> 
>    <date at="1993">1993</date>
>    <date at="Nov 1993">this month</date>
>    <date at="1 Nov 1993">the first of November, 1993</date>
>    <date at="1 Nov 1993 00:00:00">midnight</date>
> 
> Length of field is enough to distinguish between these.

We _could_ avoid reinventing the wheel here: The ISO standard
ISO 8601:1991 "Data elements and interchange formats -
Information interchange - Representation of dates and times"
was designed to solve all the above (and other) problems.
Please take a look at it before you invent something else.

The full format for a time+date spec is aprox. as formely
exemplified:

	1993-08-03T00:21:33+0200 (my time zone)
	1993-08-02T22:21:33Z (UTC)
	19930802T222133Z (UTC, shorter form)

I don't have the standard at hand now but can quote more from
it tomorrow if you are interested.
---
Peter Svanberg, NADA, KTH		    Email: psv@nada.kth.se
Dept of Num Analysis and Comp. Science,
Royal Institute of Technology		    Phone: +46 8 790 71 40
S-100 44  Stockholm, SWEDEN		    Fax:   +46 8 790 09 30