LibreOffice 3.3 End Of Life and will be removed from Pootle

Hi,

There will not be any more releases from the LibreOffice 3.3 code
line. LibreOffice 3.3.5 was planned, but apparently due to lack of
developer's interest, only 1 bugfix went in after 3.3.4, so it does
not make sense to release it. I would like to remove the related
projects from Pootle soon. We will save some resources and new
contributors will not be confused by those old files. If anybody needs
the files, I let a few days to archive them.

Thanks,
Andras

Hi,

There will not be any more releases from the LibreOffice 3.3 code
line. LibreOffice 3.3.5 was planned, but apparently due to lack of
developer's interest, only 1 bugfix went in after 3.3.4, so it does
not make sense to release it.

I'm glad to hear, although that's probably my commit for the 3.3 branch.

I would like to remove the related
projects from Pootle soon. We will save some resources and new
contributors will not be confused by those old files. If anybody needs
the files, I let a few days to archive them.

Any need to archive them ? Can't they be pulled from the 3.3 git branch ?

Kaplan

I meant, if some teams wanted to archive for themselves, they should
do now. In git, it is a mess, SDF files in l10n repo + additional PO
files in build repo. Well, the information is there in git, it is just
hard to extract. :slight_smile:

Best regards,
Andras

It may sound nooblish but could you explain what do you mean by "archive"

TIA

Hi,

I meant, if some teams wanted to archive for themselves

It may sound nooblish but could you explain what do you mean by "archive"

Download the full set of po files for your language (as zip) and store it at your favourite place for archiving files :wink:

regards,

André

Þann fös 4.nóv 2011 07:12, skrifaði André Schnabel:

Hi,

I meant, if some teams wanted to archive for themselves

It may sound nooblish but could you explain what do you
mean by "archive"

Download the full set of po files for your language (as zip)
and store it at your favourite place for archiving files :wink:

... which can then be read into your Translation Memory of choice for being able to get suggestions based on older strings while translating offline ... :slight_smile:

Sveinn

Thanks for the explanation :slight_smile:

I thought that LibO already does that for us?

Þann fös 4.nóv 2011 08:19, skrifaði Nguyễn Vũ Hưng:

Download the full set of po files for your language (as zip)
and store it at your favourite place for archiving files :wink:

... which can then be read into your Translation Memory of choice for being
able to get suggestions based on older strings while translating offline ...
:slight_smile:

Thanks for the explanation :slight_smile:

I thought that LibO already does that for us?

Not really; Pootle can create a terminology file which you can then edit. Terminology is mostly a word-by-word aide to have consistent terms in your texts and can be used in Pootle itself (as well as in many offline editors).

Translation Memory Database (TM) works with one or many words (segments) so that one can have suggestions on longer text strings - some translation software lets you decide how close matches you want as suggestions, and to construct rules about word order and such.

Normally I work with several different TM's; one for business texts, another for UI-strings and Help, and a specific one for networking jargon. Depends on your language and workflow.

You can create those TM's with some of the Translate Toolkit on the CLI (e.g. po2tmx); you can also do it with some of the offline-editors which either let you add all opened files to the TM or even are capable to parse through a folder structure of choice.
This is where it can be handy to keep archives of older files.

Best regards,
Sveinn í Felli

<http://en.wikipedia.org/wiki/Translate_Toolkit>
<http://socialsourcecommons.org/toolbox/show/1107>
<http://en.wikipedia.org/wiki/Computer-assisted_translation>

Not really; Pootle can create a terminology file which you can then edit.
Terminology is mostly a word-by-word aide to have consistent terms in your
texts and can be used in Pootle itself (as well as in many offline editors).

Yes, if one uses pootle to translate; which is a perfect place to
quick fix translation bugs.

I prefer poedit for big po file transation.

Normally I work with several different TM's; one for business texts, another
for UI-strings and Help, and a specific one for networking jargon. Depends
on your language and workflow.

That is amazing and I'd want you to share the tools and the workflow of yours.

You can create those TM's with some of the Translate Toolkit on the CLI
(e.g. po2tmx); you can also do it with some of the offline-editors which
either let you add all opened files to the TM or even are capable to parse
through a folder structure of choice.
This is where it can be handy to keep archives of older files.

Now I understand why we need (various) "archived" po files at the end
of release cycles.

<http://en.wikipedia.org/wiki/Translate_Toolkit>
<http://socialsourcecommons.org/toolbox/show/1107>
<http://en.wikipedia.org/wiki/Computer-assisted_translation>

Which ones are you using?

Op Vr, 2011-11-04 om 16:22 +0700 skryf Nguyễn Vũ Hưng:

> Not really; Pootle can create a terminology file which you can then edit.
> Terminology is mostly a word-by-word aide to have consistent terms in your
> texts and can be used in Pootle itself (as well as in many offline editors).

Yes, if one uses pootle to translate; which is a perfect place to
quick fix translation bugs.

I prefer poedit for big po file transation.

I agree that Pootle is ideal for quick fixes and for many aspects of
team work and review. It is also great for searching through all files
in a project. An offline editor can be useful in other cases. See also
my other email about future features of Pootle and how we try to narrow
the feature gap.

> Normally I work with several different TM's; one for business texts, another
> for UI-strings and Help, and a specific one for networking jargon. Depends
> on your language and workflow.

That is amazing and I'd want you to share the tools and the workflow of yours.

Virtaal stores all local TM in a database which can be swapped out if
you want (a single database file in your profile). It also allows to
enable/disable different plugins for translation memory as you prefer. I
don't think a lot of people are interested in this kind of
customisability if they only translate software GUIs and help, but of
course some people might want to do so anyway.

> You can create those TM's with some of the Translate Toolkit on the CLI
> (e.g. po2tmx); you can also do it with some of the offline-editors which
> either let you add all opened files to the TM or even are capable to parse
> through a folder structure of choice.
> This is where it can be handy to keep archives of older files.

Now I understand why we need (various) "archived" po files at the end
of release cycles.

> <http://en.wikipedia.org/wiki/Translate_Toolkit>
> <http://socialsourcecommons.org/toolbox/show/1107>
> <http://en.wikipedia.org/wiki/Computer-assisted_translation>

Which ones are you using?

The Translate Toolkit can combine lots of PO files into a TMX file with
po2tmx. You can import the translated strings from any supported file
into Virtaal's TM by saving it inside Virtaal. (Saving might be disabled
if you only openend it, so just add and remove a space to enable
'Save'.)

If you want to import a lot of PO files into the TM database of Virtaal,
you can also use the script from the Translate Toolkit called
build_tmdb.

Keep well
Friedel

Op Vr, 2011-11-04 om 09:15 +0000 skryf Sveinn í Felli:

Þann fös 4.nóv 2011 08:19, skrifaði Nguyễn Vũ Hưng:
>>> Download the full set of po files for your language (as zip)
>>> and store it at your favourite place for archiving files :wink:
>>>
>>
>> ... which can then be read into your Translation Memory of choice for being
>> able to get suggestions based on older strings while translating offline ...
>> :slight_smile:
>
> Thanks for the explanation :slight_smile:
>
> I thought that LibO already does that for us?
>

Not really; Pootle can create a terminology file which you
can then edit. Terminology is mostly a word-by-word aide to
have consistent terms in your texts and can be used in
Pootle itself (as well as in many offline editors).

For reference, here is the information about this functionality in
Pootle:
http://translate.sourceforge.net/wiki/pootle/terminology_matching

Similar terminology matching is available in Virtaal.

Translation Memory Database (TM) works with one or many
words (segments) so that one can have suggestions on longer
text strings - some translation software lets you decide how
close matches you want as suggestions, and to construct
rules about word order and such.

For the upcoming version of Pootle we have support for translation
memory during translation. The new version isn't quite ready for release
yet, but we have already used it successfully at some localisation
events and it looks very promising. It uses an online Translation Memory
service that we can keep up to date with the latest translations of
things like LibreOffice.

This service is already available with Virtaal 0.7. I don't think the
current TM database has the very latest LibreOffice translations, but we
can hopefully do a refresh of the data quite soon.

If people are able to help in testing or giving the last bit of
refinement for the upcoming Pootle release, that would be great to speed
up the process. A testing server for the upcoming release is available
here:

http://test.locamotion.org/

It is loaded with some old copy of the database of our main Pootle
server. All data entered there will be lost at some stage, but feel free
to play around. You can see a translation memory result here, for
example:

http://test.locamotion.org/fr/virtaal/fr.po/translate/#unit=187419

Keep well
Friedel

Hi

Pulling this thread from long time ago... Had translation memory been
activated/loaded in our instance of pootle?

I was told other similar office projects had it enabled.

Any news on it?

Thanks

Olivier

Op Vr, 2011-11-04 om 09:15 +0000 skryf Sveinn í Felli:

Þann fös 4.nóv 2011 08:19, skrifaði Nguyễn Vũ Hưng:

Download the full set of po files for your language (as zip)
and store it at your favourite place for archiving files :wink:

... which can then be read into your Translation Memory of choice for being
able to get suggestions based on older strings while translating offline ...
:slight_smile:

Thanks for the explanation :slight_smile:

I thought that LibO already does that for us?

Not really; Pootle can create a terminology file which you
can then edit. Terminology is mostly a word-by-word aide to
have consistent terms in your texts and can be used in
Pootle itself (as well as in many offline editors).

For reference, here is the information about this functionality in
Pootle:
http://translate.sourceforge.net/wiki/pootle/terminology_matching

Similar terminology matching is available in Virtaal.

Translation Memory Database (TM) works with one or many
words (segments) so that one can have suggestions on longer
text strings - some translation software lets you decide how
close matches you want as suggestions, and to construct
rules about word order and such.

For the upcoming version of Pootle we have support for translation
memory during translation. The new version isn't quite ready for release
yet, but we have already used it successfully at some localisation
events and it looks very promising. It uses an online Translation Memory
service that we can keep up to date with the latest translations of
things like LibreOffice.

This service is already available with Virtaal 0.7. I don't think the
current TM database has the very latest LibreOffice translations, but we
can hopefully do a refresh of the data quite soon.

If people are able to help in testing or giving the last bit of
refinement for the upcoming Pootle release, that would be great to speed
up the process. A testing server for the upcoming release is available
here:

http://test.locamotion.org/

It is loaded with some old copy of the database of our main Pootle
server. All data entered there will be lost at some stage, but feel free
to play around. You can see a translation memory result here, for
example:

http://test.locamotion.org/fr/virtaal/fr.po/translate/#unit=187419

Keep well
Friedel

- --
Olivier Hallot
Founder, Board of Directors Member - The Document Foundation
The Document Foundation, Kurfürstendamm 188, 10707 - Berlin, Germany
Gemeinnützige rechtsfähige Stiftung des bürgerlichen Rechts
Legal details: http://www.documentfoundation.org/imprint
LibreOffice translation leader for Brazilian Portuguese
+55-21-8822-8812

http://www.mail-archive.com/l10n@global.libreoffice.org/msg07191.html

doesn't work currently because the TM-Server doesn't support https and
modern browsers block loading stuff from http servers when the
mainpage is accessed using https

Chromium browser will display the Shield in location bar where you can
allow it to load the "script from unsecure location" (i.e. the request
from the Translation-Memory server)

Other browsers should have similar notification or an option to
whitelist the TM-server.

ciao
Christian