I'm Carmen from Friesland, The Netherlands. I am currently working on
GNOME to make the GNOME desktop available in Esperanto. At the current
rate, GNOME 3.22 will have its core release (standard apps, core
libraries) completely translated into Esperanto, which I am fairly proud
Unfortunately, from the big distributions, only Debian (and its
derivatives) have Esperanto available as a choice out of the box. I
want to make sure that Fedora also gets that option.
Could someone give me a few steps in the right direction to get that
ball rolling? I imagine an Esperanto team must be created on Zanata
(which is why I have CC'd Michael Moroni, who is listed as coordinator
at <https://fedoraproject.org/wiki/L10N_Teams>;). I also imagine some
packages need adjusting.
I have some experience with RPM packaging from openSUSE's Build Service,
so I can probably do that just fine on my own, but I would really
appreciate if someone provided me with an overview of the core language
packages (or how to find such an overview). I imagine it would be okay
to just send a patch to those packages to enable/add Esperanto? Or is
this all done automatically? That truly would be magical.
A little bit about myself finally: I'm a free software advocate, avid
D&D player, software engineering student, FSFE volunteer, and a couple
of other things. You can find me at the FSFE booth at SFScon in
Bozen/Bolzano, 16 November.
Thanks and with kindness,
Join us on irc.freenode.net in #fedora-meeting-1 for the Fedora 29
Final Release Readiness meeting. This meeting will be held on Thursday,
2018-10-18 at 19:00 UTC.
We will meet to make sure we are coordinated and ready for the Final
release of Fedora 29. Please note that this meeting will be held even
if the release is delayed at the Go/No-Go meeting on the same day two
You may receive this message several times in order to open this
meeting to the teams and to raise awareness, so hopefully more team
representatives will come to this meeting. This meeting works best
when we have representatives from all of the teams.
For more information, see
View the meeting on Fedocal:
Fedora Program Manager
Name: Qiyu Yan
Location: Haidian District, Beijing
About You: Student from UCAS. Knows some C/C++,
You and the Fedora Project: Using Fedora on various devices, and want
to contribute my own effort
GPG KEYID and fingerprint:
[Yan@Karuboniru ~]# gpg --fingerprint 78F2F78AD61E7448
pub 2048R/D61E7448 2018-09-26
密钥指纹 = 18BD 8B07 636E 0C2F D3B1 BD70 78F2 F78A D61E 7448
uid Qiyu Yan <Yanqiyu01(a)gmail.com>
sub 2048R/886C2AC3 2018-09-26
I haven't contributed for years now but still checking several things.
The websites are building hourly (please check me Robert and Co if I
am wrong) and has been crashing for 73hours now.
Due to the Danish incorrect translation.
Can someone fix this?
msgfmt -c -o po/da.mo po/da.po
po/da.po:262: number of format specifications in 'msgid' and 'msgstr'
does not match
msgfmt: found 1 fatal error
The string is there at Zanata
Even Zanata mark this translation as bad but users don't know that it
fails our websites rebuild...
I dunno how to 0) mark as incorrect under Zanata, 1) get the username
who last translated.. I have just used the "reject translation" button
but don't know its real meaning...
@websites shouldn't we improve our Makefile to: "msgfmt -c " in order
to avoid the stacktrace (I might send you a patch :/) continue with
default po template for that broken language ? Or open a BZ ticket
against that team? Or send an email to the mailing list?
It just feel bad to be notified that websites build is broken without
Le 2018-10-01 16:20, Petr Bokoc a écrit :
> Adam, the proposal looks reasonable. Regarding where to store
> translated sources: based on the Antora issue for multilang
We decided to have different repositories because Adam think it is more
convenient for his CI work and convinced me it will work fine ;)
As one repository for all translation file would create multiple issues,
we decided to have dedicated repositories to store POT/PO.
Please read this page for more details:
> Of course, that requires that someone to periodically in the POs from
> our translation platform, run the script to apply them to the sources,
> and push the translated sources in the correct directory. The process
> was fairly manual before with DocBook/publican, too, so it shouldn't
> be a huge issue.
The "fairly manual" process was a real issue in the past. We had a lot
of translation that never been published :(
What Adam is working on is: automation.
So, we should have something like this:
English source -> POT files -> PO files -> translated source ->
English source => Doc repository ==> (D)
POT files -> PO files => localization repository ==> (L)
translated source => "web" repository ==> (W)
When there is a change in (D), it will call a job, and update pot files
When there is a change in (L), translation platform will know and get
When there is a change in translation platform, it will update content
every day or night, content from (D) and (L) are used to generate
content for (W), and publish it
To be open: I offered to run a test using my own server as translation
platform with Weblate and this is a personal choice for the development
What we design is and should be translation platform agnostic.
We have to discuss between translators if we want to use Weblate as a
real world test or Zanata for the coming weeks.
There is good and bad: Zanata will be stable but will require way more
time for automation (I won't go into details here).