RE: Can servlet know the timezone of the browser?

My feeling is that timezone info is borderline.  There are
certainly apps where it makes sense to include timezone info
along with locale.  However I prefer to err on the side
of a narrower definition of locale.

The point that Mark makes is that the current locale initiative
is not meant to solve the general problem of communicating
preferences.  That is still up to particular apps to solve.
I sympathize with the feeling that "if I still need
to build a proprietary mechanism to communicate preferences then
what good does the locale initiative do for me?"

I think we know the answer to that.  The locale initiative will
help us communicate - not all - but a large number of cultural
preferences with shorthand notation and have apps treat those
preferences in a standard fashion across apps and across platforms.

That is huge.  And its success is furthered by erring on the side
of a narrow definition of locale.

-Paul
-----Original Message-----
From: Mark Davis [mailto:mark.davis@jtcsv.com]
Sent: Saturday, July 26, 2003 12:09 PM
To: Carl W. Brown; www-international@w3.org
Subject: Re: Can servlet know the timezone of the browser?



I am familiar with Olson IDs, of course. It still doesn't mean that
timezone is appropriate for a "locale". It is certainly a setting, and
certainly in many contexts important to convey, but that doesn't
necessarily mean it belongs in a locale (however that is defined).

Mark
__________________________________
http://www.macchiato.com
$Bcv%3(B  $Bc`W&(Bppur si muove$Bc`Ycw'î(B
----- Original Message -----
From: "Carl W. Brown" <cbrown@xnetinc.com>
To: <www-international@w3.org>
Sent: Saturday, July 26, 2003 07:56
Subject: RE: Can servlet know the timezone of the browser?


 >
 > Mark,
 >
 > > People may differ in what they think of as a locale. We tend to
take a
 > > narrow view, that it is principally items that differ according to
 > > language, thus excluding other items like preferred timezone,
 > > preferred currency, preferred character set, smoker/non-smoker
 > > preference, meal preference (vegetarian, kosher, etc.), music
 > > preference, religion, party affiliation, favorite charity, etc.
 > >
 > > It is not that these other items may not be important; they may
very
 > > well be, depending on the application. And thus one may need to
 > > communicate them. But they don't necessarily belong in an
 > > all-encompassing 'locale'.
 >
 > You are right that you have to draw the line somewhere.  I used to
think of time zones as separate from locales until I became more
familiar with Olsen time zones.  He establishes times zones that are
unique subdivisions within a country.  This makes sense since each
country has political control over time zones.  The time zone
parameters are regional data within the country data.  It is
sub-country data like language within a country may vary to produce
sub-languages.
 >
 > Carl
 >
 >
 >
 >
 >

Received on Sunday, 27 July 2003 21:59:48 UTC