After a discussion we had in -boot back in November, which involved
Frans Pop and myself, mostly, I worked on a new version of the
synchronization script (scripts/l10n/l10n-sync in SVN), that allows
splitting messages in "sublevels" for all core D-I packages.
That works by "marking" each translatable string in
<package>/debian/<package>.templates with a sublevel marker indicating
what sublevel the string belongs to:
=========================================
Template: debian-installer/locale
Type: select
Choices: ${LOCALELIST}
# :sl2:
_Description: Choose a locale:
Based on your language and country choices, the following locale
parameters are supported.
=========================================
The marker above is "# :sl2:" and indicates that the following
translatable paterial (_Description and what follows) belongs to
"sublevel 2".
Sublevels are defined this way:
sublevel 1: default installs on common architectures
sublevel 2: general strings not used during default installs
sublevel 3: expert strings (some low prio, some for e.g. RAID)
sublevel 4: specific to less-popular arches (powerpc, mips, sparc) or used in experimental features
sublevel 5: same for high-end (hppa, ia64, s390) and hobby (m68k) arches and old stuff (partconf, partitioner, that are replaced by partman for a while)
The sync script is now able to spread out strings in
packages/sublevelN/*.po where N goes from 1 to 5, depending in the
presence of this marker. Unmarked strings go to "level 1".
I just finished marking all strings in D-I packages with the above
criteria.
A test run of l10n-sync gave:
sublevel 1: 515 strings
sublevel 2: 465 strings
sublevel 3: 397 strings
sublevel 4: 168 strings
sublevel 5: 176 strings
The total of 1721 strings is higher than the current 1682 strings
because some identical strings are now in different sublevels. However
the sync script is intelligent enough to reuse translations from lower
sublevels in high sublevels.
So, now everything is ready for the switch to a 5 sublevels
translation system for "core D-I".
What would change?
-----------------
For existing translations, it means that what is currently in one big
(and often complete) 1682-strings file in packages/po, translators
would have to work on 5 different files in packages/po/sublevelN.
There would be NO possibility to keep the unique bug file (that is
meant to avoid complicating the sync script)
For work in progress, that would allow translators to focus on the most
important strings and, for instance, to target completing sublevels 1
and 2, before moving to the current "level 2" (tasksel, iso-codes...)
before coming back on "sublevel 3" later.
For D-I release managers, a decision should be made about the
completeness we require to activate a language. That could be either
"sublevel 1 and sublevel 2" or "sublevels 1 to 3".
When will this be activated?
---------------------------
I need more input and criticism to decide whether to activate this and
when. So, don't expect this change to take place before the first
week(s) of January.
I'll do my best for it to happen as I'm currently not very happy with
us enforcing the translation of very obscure stuff before activating a
language, which is not the best reward to translators' efforts.
Will the current "levels" change?
---------------------------------
Probably not. We could probably move sublevel 4 and sublevel 5 to
"level 4" or "level 5" as these translation are clearly less
important. That is feasible quite easily and would be more realistic.
--
Attachment:
signature.asc
Description: Digital signature