From: Erkki I. Kolehmainen (eik@iki.fi)
Date: Fri Dec 21 2007 - 03:22:35 CST
It is commonplace to define the locale by language and country (region), and
possibly script.
In addition, variants can be defined and selected by the users for certain
items. For this item, the desire to be able to select a given variant would
seem to be relatively strong. These variants should have fairly natural
names, and the vetting process should decide which one would be used as the
CLDR default for a given language etc. environment.
P.S. As per Finnish standard SFS 4175:2006, the full stop is the recommended
separator between hours and minutes in text (4.19.5), and the colon is only
used when the presentation needs to adhere to ISO 8601 (4.9.5).
Regards, Erkki
Erkki I. Kolehmainen
Tilkankatu 12 A 3, FI-00300 Helsinki, Finland
Puh. (09) 4368 2643, 0400 825 943; Tel. +358 9 4368 2643, +358 400 825 943
-----Alkuperäinen viesti-----
Lähettäjä: unicode-bounce@unicode.org [mailto:unicode-bounce@unicode.org]
Puolesta Jukka K. Korpela
Lähetetty: 21. joulukuuta 2007 9:46
Vastaanottaja: unicode@unicode.org
Aihe: Re: CLDR Usage of Gregorian Calendar Era Terms: BC and AD -- Can we
please have "CE" and "BCE" ?
Erkki I. Kolehmainen wrote:
> The open question remains: how to identify the alternatives? -
> Secular, non-secular?
The difference is not between secular and non-secular but between a
notation that identifies the Christian calendar with a reference to
Christ and one that identifies it as the calendar. One might call it a
difference between explicit and implicit (in expressing which calendar
is used).
I understand that this can be a big issue to some people, but it is
really an isolated dispute among all the settings, so I cannot see why
this cannot be left to be selected by the user separately, when desired.
(I don't think there are many people who really care about it, and it's
relevant in few applications only, like extracts from historical data
bases.)
In setting up user preferences, it is natural to select the display
language (in the ordinary, common sense of naming just, say, "English"),
then perhaps the country if it seems to matter, and maybe some
_separate_ choices, like "Use BCE/CE instead of BC/AD". These choices
can then be encoded as needed, if and when they need to be stored and
transmitted.
If such choices by users should be expressed by locale names, then these
names should be genuinely multi-dimensional, allowing a set of
independent choices to be encoded. But I find it much more natural to
use a locale name to identify one set of settings, then different
settings (and names and codes for them) to override some of the settings
in the locale, if desired.
To take another example, sometimes heavily debated but with different
tones, many people prefer or actually use the colon ":" as separator
between hours and minutes in time denotations, even when the rules of
the language require the full stop ".". In Finland, the colon is the
official separator in text, but the full stop is very common too. Should
such a choice be given a name of its own, and should it be encoded into
a locale name, so that a name would express that the user prefers
Finnish as written in Finland with BCE/CE notation and with the colon
notation and...? Where would this end? If some people strongly feel that
the name of a country is wrong in a locale (a quite possible scenario),
should this also be encoded as a difference between named locales? Or
_named_ at all?
Jukka K. Korpela ("Yucca")
http://www.cs.tut.fi/~jkorpela/
This archive was generated by hypermail 2.1.5 : Fri Dec 21 2007 - 03:24:30 CST