From: Jukka K. Korpela (jkorpela@cs.tut.fi)
Date: Fri Dec 21 2007 - 04:00:13 CST
Erkki I. Kolehmainen wrote:
> It is commonplace to define the locale by language and country
> (region), and possibly script.
Yes, and among these, language is probably the only one that intuitively
clear to users. The country selection is confusing, since it is far from
obvious what is being chosen. But I digress.
> In addition, variants can be defined and selected by the users for
> certain items.
The question is whether these variants need to be named and identified
as locales, in the sense of being alternatives selectable as a whole.
The more factors you bring into the variation, the more artificial the
naming and identification becomes.
> For this item, the desire to be able to select a given
> variant would seem to be relatively strong.
It is sometimes expressed strongly, but usually even people who see it
as an issue don't really make much noise about it. Perhaps most
importantly, virtually everyone who knows (and possibly prefers) "CE"
and "BCE" knows and understands the old alternatives "AD" and "BC". This
makes it reasonable to make the latter the default and the other one a
selectable alternative, with no need for a name for either the choice
(it needs to be _described_ in terms of what it means but it needs no
_name_, no identifier) or the collection of user preferences, i.e. a
locale.
To take yet another matter of preference, everyone who understands "42
%" also understands "42%" (and, for such a simple presentational issue,
vice versa as well). Thus, even if there were a greate debate in a
community over the question whether the "%" should be preceded by a
space or not (and there are such debates), we don't need a name for this
choice, or for locale variants created by it. It is sufficient to let
people decide on it on per-setting basis.
> These variants should have fairly natural names,
Why would they need names in the first place? When we start naming such
things, we cannot expect to find _natural_ names, and the odds are that
the names themselves will become a considerable dispute.
Jukka K. Korpela ("Yucca")
http://www.cs.tut.fi/~jkorpela/
This archive was generated by hypermail 2.1.5 : Fri Dec 21 2007 - 04:01:14 CST