[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-i18n-doc
Subject:    Re: Languages without correct plural declaration
From:       Chusslove Illich <caslav.ilic () gmx ! net>
Date:       2005-12-25 11:41:48
Message-ID: 200512251241.49052.caslav.ilic () gmx ! net
[Download RAW message or body]


>> [: Luciano Montanaro :]
>> Actually, while Italian has a simple singular/plural form, to properly
>> compose a phrase containing a number preceded by a determinative(?)
>> article (for example, "take the %n items...") the "the" article can be
>> translated "i" or "gli" depending on the following number. Actually,
>> it's "i", unless the number is 8\d* or 11(\d{3})*, that is 8, 11,
>> eightysomething, eight hundredsomething, ... eleven thousand something,
>> and so on.
>>
>> Would it be possible to use scripting to to handle this situation?

> [: Abel Cheung :]
> Some personal pronouns may or may not change due to plural
> form, such as "your" (of course, expressed in chinese).
> So whether to use plural form or not is decided by sentence
> structure, and can't be categorized simply. Is there any plural
> infrastructure existing that can accomodate these kinds of
> situation?

I'd say that no special infrastructure is technically needed in both these 
cases. It is just that for Italian you could use 3 plural forms (last two 
being always the same, except when there is definite article), and for 
Chinese 2 plurals forms (both same most of the time, except when you need 
to modify possessive pronoun). Even for Serbian/Bosnian/Croatian, and I 
imagine some other Slavic languages, an extra form for number 1 is 
sometimes needed (typically when English singular itself doesn't contain a 
number, like "Delete this file?" / "Delete these %n files?"), and that 
would prop our number of plural forms to 4.

One could however debate whether the overhead to translators, of having an 
extra plural form although most of the time it is not needed, is 
justified. If not, then scripting would of course be a viable alternative, 
if and when it becomes available in KDE 4. On the other hand, the evidence 
is mounting that Gettext's own plural handling may need some expanding.

-- 
Chusslove Illich (Часлав Илић)
Serbian KDE translation team

[Attachment #3 (application/pgp-signature)]

[prev in list] [next in list] [prev in thread] [next in thread]