> [: Max Brazhnikov :] > Is there a reason to not pack sr@* separately, like ca and ca@valencia for > example?
I'm not sufficiently familiar with ca/ca@valencia situation to make comparison, so here is the reason for sr@* on its own: On the theoretical side. All four sr@* variants are part of single Serbian language standard, both on the official level and in everyday life. All sorts of written materials are published in any of the variants, all over the Serbian-speaking communities. In particular, from elementary school onwards, all children are put in contact with texts in any of these variants. Even on single-person level, many people regularly use more than one variant, depending on where and what they are expressing. On the practical side. During KDE 3, we've had a constant stream of questions "I installed Serbian language, but I get only <variant-A>, how can I get <variant-B>?". Due to the above, people simply expect "Serbian language" to mean the variant they would like (and this is what they had with most Gettext-based non-KDE codes, since they install all languages no questions asked). For a public or semi-public multi-user machine, it makes no sense whatsoever to install one variant and not the others; but this was happening all the time, and people were even less happier when they were told to "find the sysadmin" and ask him to install all variants. Even for a household machine, different members of the hausehold may use different variants. Since we started delivering the unified package with KDE 4, all these questions "where is <variant-X>" disappeared. (Well, except when the language pack was truncated, and except on Ubuntu...) -- Chusslove Illich (Часлав Илић)
signature.asc
Description: This is a digitally signed message part.
_______________________________________________ release-team mailing list [email protected] https://mail.kde.org/mailman/listinfo/release-team
