This site has been retired. For up to date information, see handbook.gnome.org or gitlab.gnome.org.


[Home] [TitleIndex] [WordIndex

Translation teams survey

Rationale

There is an interest on knowing how the GNOME Translation teams work so that we can improve their way of work by knowing how the others are working. See Og Maciel's mail to the gnome-i18n mailing list.

Ubuntu translation community has already made something similar (see their survey).

Survey Results

Gil Forcada, with the feedback from other community members, conducted the GNOME I18N Survey by sending a questionnaire on August 13 to every GTP language coordinator, and collecting answers for two weeks.

On October 16, Gil presented results of the survey. A brief analysis of the results was included in his message and is excerpted below. 36 out of 119 coordinators replied, which represents 30% coverage.

Quick links:

Two remarks by Gil:

  1. You will notice that there are 4 teams that only have data in the ods file and there isn't a counterpart in raw files. Their team leaders asked to just publish the summary and not the raw data.
  2. If you wonder, the teams order is based on the time I got the reply, so the first team was Belarusian and the last one Danish.

== SUMMARY ==

- Teams range from 1 or 2 to +15 members (average to 3 members per team)

- Nearly all work is made by volunteers (only 2 or 3 teams have paid
members)

- All teams use mailing lists to coordinate (and some other tools)

- On the need help/advice question coordinators asked to help work with
the documentation po files and Q&A tools/processes

- 19 out of 36 teams (52%) start working around string freeze (~1 month
before next GNOME release)

- Practically all of them use glossaries and translation guidelines

- To keep track of translation issues most use bugzilla and/or D-L but
Asturian uses a specific software for this (could you elaborate Xandru?)

- Few teams have gone to their language official institution with
miscellaneous results

- 19 out of 36 teams (52%) pass Q&A tools before committing

- 4 out of 36 teams (11%) don't use D-L, all others use it to some kind
of degree

- D-L scores a 3,7!! (from 1 bad to 5 perfect)

- 13 of 36 teams ask for git commit along with other improvements (be
sure to read them all!)

- There's a good relation with downstream translators (mostly Ubuntu,
Fedora and other distros)


== STEPS FORWARD ==

>From the replies above we can say that the coordinators are quite
comfortable and used to D-L.

Most of them didn't gave a 5 (instead they gave a 4) to D-L just because
it doesn't have git commits. Now that we are discussing if switching to
Transifex, keeping D-L or something else, having D-L to support git
commits [5] can weight a lot on D-L side.

Nearly all teams use a glossary so it would be really useful to
create/use a tool to generate an automatic glossary out of the
translations already submitted.

The automatically updated po file downloaded from Damned-Lies could be
improved if the above glossary existed pre-translating (marking them as
fuzzy and maybe from which module comes the string) the po file so that
instead of starting a module with 100% 0% 0% (translated, fuzzy, done)
it's already 80% 20% 0%.

[5] Claude reports that's on sysadmin side, I recall hearing from him
that he has working code.

http://mail.gnome.org/archives/gnome-i18n/2010-October/msg00156.html:

In the same way that there's gtk+ and gtk+-properties, i.e. two po files
for a single module, it would be extremely useful for translators if ALL
schemas were going to a different po file and thus having a double
effect on translators:
- know which strings are actually visible to the users
- reducing a lot [1] the number of strings to translate to reach the
glorious 80%

Survey Draft

Add the questions that are going to be asked to each translation team here


CategoryGnomeI18n


2024-10-23 11:49