Re: How to extract PDF of translated documentation?

classic Classic list List threaded Threaded
22 messages Options
12
Reply | Threaded
Open this post in threaded view
|

Re: How to extract PDF of translated documentation?

Federico Bruni-2
Il 13/01/2013 11:45, Francisco Vila ha scritto:
> Oh sorry! you mean PDF of the translation of the web document.
>
> http://www.lilypond.org/development.de.html  links to web.de.pdf but
> you are right, this is a dead link.

the dead links to web.xx.pdf occur only on certain languages: fr, de, cs, hu

I'm CCing translation.
Can anyone have a look at that?

--
Federico


Reply | Threaded
Open this post in threaded view
|

Re: How to extract PDF of translated documentation?

Francisco Vila
2013/1/13 Federico Bruni <[hidden email]>:
> Il 13/01/2013 11:45, Francisco Vila ha scritto:
>
>> Oh sorry! you mean PDF of the translation of the web document.
>>
>> http://www.lilypond.org/development.de.html  links to web.de.pdf but
>> you are right, this is a dead link.
>
>
> the dead links to web.xx.pdf occur only on certain languages: fr, de, cs, hu

for 'es' the link is alive but points to untranslated web.pdf
--
Francisco Vila. Badajoz (Spain)
www.paconet.org , www.csmbadajoz.com


Reply | Threaded
Open this post in threaded view
|

'make doc' should generate translated PDF files of website [WAS: Re: How to extract PDF of translated documentation?]

Federico Bruni-2
Il 13/01/2013 11:58, Francisco Vila ha scritto:

> 2013/1/13 Federico Bruni <[hidden email]>:
>> Il 13/01/2013 11:45, Francisco Vila ha scritto:
>>
>>> Oh sorry! you mean PDF of the translation of the web document.
>>>
>>> http://www.lilypond.org/development.de.html  links to web.de.pdf but
>>> you are right, this is a dead link.
>>
>>
>> the dead links to web.xx.pdf occur only on certain languages: fr, de, cs, hu
>
> for 'es' the link is alive but points to untranslated web.pdf
>

Same for italian and nederlands.
Curiously, spanish doc v2.15 links to web.es.pdf (therefore it's broken):
http://lilypond.org/doc/v2.15/Documentation/web/development.es.html

Anyway, translated versions of PDF files were never generated (going
back up to 2.13):
http://lilypond.org/doc/v2.15/Documentation/

We would expect to have the translated PDF, as in any other manual.
I've opened an issue in the tracker:
http://code.google.com/p/lilypond/issues/detail?id=3107

I've CCed also lilypond-bug (sorry for the big cross-posting).

--
Federico


Reply | Threaded
Open this post in threaded view
|

Re: How to extract PDF of translated documentation?

Harmath Dénes
In reply to this post by Francisco Vila
As responsible for the Hungarian translation, what are the exact steps I should perform in order to solve this?

2013/1/13 Francisco Vila <[hidden email]>
2013/1/13 Federico Bruni <[hidden email]>:
> Il 13/01/2013 11:45, Francisco Vila ha scritto:
>
>> Oh sorry! you mean PDF of the translation of the web document.
>>
>> http://www.lilypond.org/development.de.html  links to web.de.pdf but
>> you are right, this is a dead link.
>
>
> the dead links to web.xx.pdf occur only on certain languages: fr, de, cs, hu

for 'es' the link is alive but points to untranslated web.pdf
--
Francisco Vila. Badajoz (Spain)
www.paconet.org , www.csmbadajoz.com



Reply | Threaded
Open this post in threaded view
|

Re: How to extract PDF of translated documentation?

Federico Bruni-2
2013/1/14 Harmath Dénes <[hidden email]>
As responsible for the Hungarian translation, what are the exact steps I should perform in order to solve this?

Hi Darmath

I don't think it's your duty, the problem is in the build system.
Honestly, I don't know why the links change depending on the language. The first step is waiting for some developer to fix this bug:

Then we'll see if there's any error in any language
Reply | Threaded
Open this post in threaded view
|

Re: How to extract PDF of translated documentation?

David Kastrup
In reply to this post by Harmath Dénes
Harmath Dénes <[hidden email]> writes:

> 2013/1/13 Francisco Vila <[hidden email]>
>
>     the dead links to web.xx.pdf occur only on certain languages:
>     fr, de, cs, hu
>    
>    
>     for 'es' the link is alive but points to untranslated web.pdf
>
> As responsible for the Hungarian translation, what are the exact steps
> I should perform in order to solve this?

This is first of all a general script problem since the files in
question are not even produced, so you can't fix the link to them before
they are even created.

But I have to admit at being disturbed by this response for an entirely
different reason.  I remember a similar question from the maintainer of
the Chinese translation around the time of the release of 2.16 which
I more or less ignored because I am not regularly involved with the
translator list and which I partly just attributed to confusion not
relevant to the project.

But together with this response, I can't help the impression that we
have, for several languages, people who feel responsible for maintaining
the translations but are unaware that many translations are severely
outdated and not really all that useful any more for working with the
current version of LilyPond: partly because they describe things that
don't even work in the described manner any more, partly because the
English documentation has seen a lot of improvements making things much
clearer.

The only translations that appear to be thoroughly maintained are the
Spanish and French translation, with the German translation following
close behind.

It would a lamentable state of affairs if we had translations that are
not maintained because of a lack of translators and manpower.  But this
and some other mailings on the translator list rather suggest to me that
we _might_ have manpower and persons feeling responsible, but that the
information what they could be doing for the project does not even
arrive at their doorstep.

And that would not be lamentable, but stupid.  If we have people
enlisted and willing to do a job, we should not let their availability
go to waste.

Are there people willing to do translations, particularly those who
actually _are_ already subscribed to the translator list, that are
waiting for instructions or whatever else before they know how to get to
work on making their language better supported with LilyPond?  If so, we
need to organize this effort better and more transparently.

--
David Kastrup


Reply | Threaded
Open this post in threaded view
|

Re: How to extract PDF of translated documentation?

Federico Bruni-2
2013/1/14 David Kastrup <[hidden email]>
The only translations that appear to be thoroughly maintained are the
Spanish and French translation, with the German translation following
close behind.


Well, the italian translation is far to be completed but it _is_ maintained.

 
It would a lamentable state of affairs if we had translations that are
not maintained because of a lack of translators and manpower.  But this
and some other mailings on the translator list rather suggest to me that
we _might_ have manpower and persons feeling responsible, but that the
information what they could be doing for the project does not even
arrive at their doorstep.

And that would not be lamentable, but stupid.  If we have people
enlisted and willing to do a job, we should not let their availability
go to waste.

Are there people willing to do translations, particularly those who
actually _are_ already subscribed to the translator list, that are
waiting for instructions or whatever else before they know how to get to
work on making their language better supported with LilyPond?  If so, we
need to organize this effort better and more transparently.

Probably the CG should be updated, improved and simplified.
I think that at the moment it would be a bit hard for a new translator to get in.
Reply | Threaded
Open this post in threaded view
|

Re: How to extract PDF of translated documentation?

Federico Bruni-2
In reply to this post by David Kastrup
2013/1/14 Werner LEMBERG <[hidden email]>
I wonder what's the best way to update outdated documentation.
Reading the whole documentation and checking the translation is
probably quite hard and needs a lot of time.  Would it be better
instead to use a git repository viewer, walking over the English
documentation commits and translating them commit by commit?  Or do
you recommend to check the differences between development releases?

Hi Werner
the maintenance is described here:

I use only check-translation.

If you have questions let us know.
Reply | Threaded
Open this post in threaded view
|

Re: How to extract PDF of translated documentation?

David Kastrup
In reply to this post by Federico Bruni-2
Federico Bruni <[hidden email]> writes:

> 2013/1/14 David Kastrup <[hidden email]>
>
>     The only translations that appear to be thoroughly maintained are
>     the
>     Spanish and French translation, with the German translation
>     following
>     close behind.
>    
>    
>
> Well, the italian translation is far to be completed but it _is_
> maintained.

I apologize for forgetting you.

> Probably the CG should be updated, improved and simplified.
> I think that at the moment it would be a bit hard for a new translator
> to get in.

And I think we could make of several new ones or, probably needing
similar measures, reactivate some old ones.

It would also be nice to have some redundancy.  I know that I am not
particularly fond of the German translation since its language is partly
awkward, in all likelihood partly due to the need of thinking of German
language structure and vocabulary independently from the English one:
the language flow and word choice is often stumbling a bit too close to
the English version.  Good "literary" translations are actually
something that does not come naturally even to people who are perfectly
able to express themselves well in either language in separation.

So it would actually benefit the material if a second person went over
everything again with the focus of getting it to read well independently
from the material, and then a third person checked that the original and
translated version still corresponded _factually_.

I can't actually speak for more than German here, but I would guess that
other translations might also bit from some independent polishing up by
people able to focus just on the target language rather than the
translation process.

Trevor is doing a tremendous job casting technobabble and information
into comprehensible English, but that is quite an undertaking with slow
progress, and I consider it somewhat comparable as it is also sort of an
English->English process.  While he is more freely rewriting information
than a "mere" translation usually would, the latter still needs a solid
dose of willingness to make more sense in the translation than a literal
translation would produce.

So how can we improve the efficiency of our current translation work
force, and how can we make it easier for people to help in a way that
does not step on anybody's toes and actually leads to better results?

--
David Kastrup


Reply | Threaded
Open this post in threaded view
|

Re: How to extract PDF of translated documentation?

Federico Bruni-2
2013/1/14 Werner LEMBERG <[hidden email]>

> So how can we improve the efficiency of our current translation work
> force, and how can we make it easier for people to help in a way
> that does not step on anybody's toes and actually leads to better
> results?

Yeah.  Let's assume that I would like to quickly improve various
places of the German translation (for example, just fixing a typo).
What should I do?  Directly committing to `staging' doesn't seem
appropriate...


    Werner

All the translation work should be made in the translation branch.

git checkout -b translation origin/translation

and push changes to translation.

If you just fix a typo and don't update anything, _don't_ change the committish in the document.

Reply | Threaded
Open this post in threaded view
|

Re: How to extract PDF of translated documentation?

David Kastrup
In reply to this post by David Kastrup
Werner LEMBERG <[hidden email]> writes:

>> So how can we improve the efficiency of our current translation work
>> force, and how can we make it easier for people to help in a way
>> that does not step on anybody's toes and actually leads to better
>> results?
>
> Yeah.  Let's assume that I would like to quickly improve various
> places of the German translation (for example, just fixing a typo).
> What should I do?  Directly committing to `staging' doesn't seem
> appropriate...

We have a translation branch, but for typos, I think (correct me if I am
wrong) that of little relevance as we have back-and-forth merging.

However, this implies that

a) no significant reformatting occurs, as that might have negative
impact on parallel work
b) the "typo" is an actual typo rather than, say, disagreement over the
spelling reform to deploy

Factual errors occuring only in one translation should be fixed there.
Factual errors inherited from the English version should be fixed there
first.  If they are already fixed there, then the _whole_ relevant
passage should be translated, and commit ids adapted, namely then the
job should be done right, and likely in the translation branch.

I have no big clue here, but that is more or less my gut feeling.  I
don't know how one would approach retranslating or reworking an existing
passage.  Do the regular translators get to see Rietveld reviews?  A
retranslation by a non-translator would seem to warrant proper review,
but it would be somewhat pointless if the actual main translator only
notices the review once it has been committed.

--
David Kastrup


Reply | Threaded
Open this post in threaded view
|

Re: How to extract PDF of translated documentation?

Francisco Vila
In reply to this post by David Kastrup
2013/1/14 Werner LEMBERG <[hidden email]>:
> I wonder what's the best way to update outdated documentation.
> Reading the whole documentation and checking the translation is
> probably quite hard and needs a lot of time.  Would it be better
> instead to use a git repository viewer, walking over the English
> documentation commits and translating them commit by commit?  Or do
> you recommend to check the differences between development releases?

Werner: we have means to know exactly what needs updating in a
translation. John Mandereau built the check-translation method
Federico mentioned in his response. CG explains it. If a file does not
change, keeps unnoticed for years. You don't have to read the
documentation to check it.

This is my workflow for updating Spanish translations:

 - Open a terminal in ~/source/lilypond/Documentation
 - move git to translation branch if not there already.
 - git pull the translation branch
 - make ISOLANG=es check-translation > ct.log

A bash script counts lines in previous version of ct.log, then
performs the make command above, then counts and prints its length
again. Thus, I have a shrinking number and my goal is reaching 1 (one)
for it.

This log is usually very long. Examine it with less. It contains all
differences for all outdated files since they were last updated.
Choose a file, e.g. notation/chords.itely

Now, updating this file comprises two stages. One is structure; for
this I use the meld utility this way: (I am in Documentation)

 - meld notation/chords.itely es/notation/chords.itely

I have a script that does this:

  $ mymeld.sh chords.itely

Now, I have to clone the node/section structure of the English file
(left) to the translated file (right) using clickable arrows in meld.
Copy whole sections, menus, paragraphs, lilypond snippets, single
lines, whatever. No matter if they are in English. Save your file and
exit meld.

Now comes 2nd stage: actually translating the file. I edit my files
with emacs. I use alt-tab to switch to the terminal which has 'less'
showing differences for the file. One difference at a time, I
translate those words/lines/paragraphs. Then update the texinfo menus
with emacs. Save the file and exit. Now, 'git diff' shows what
editings I made to the file.

Red blanks in 'git diff' show trailing spaces. Remove them.

We have to mark the file as updated. I extract the version ID "the
committish" of the original file from 'git log notation/chords.itely'
and copy it in my file. Again, a bash script does this all, giving
only "notation/chords.itely" as argument. The file is updated and
marked as such.

 - git add es/notation/chords.itely
 - git commit -m 'Doc-es: <description of the commit>'
 - make ISOLANG=es check-translation > ct.log

Now the file should not appear in this log.
 - less ct.log and choose another file.

Repeat until output of 'make ISOLANG=es check-translation' is void.
--
Francisco Vila. Badajoz (Spain)
www.paconet.org , www.csmbadajoz.com


Reply | Threaded
Open this post in threaded view
|

Re: How to extract PDF of translated documentation?

Francisco Vila
In reply to this post by David Kastrup
2013/1/14 David Kastrup <[hidden email]>:
> [...] I can't help the impression that we
> have, for several languages, people who feel responsible for maintaining
> the translations but are unaware that many translations are severely
> outdated and not really all that useful any more for working with the
> current version of LilyPond: partly because they describe things that
> don't even work in the described manner any more, partly because the
> English documentation has seen a lot of improvements making things much
> clearer.

Both reasons are important but we can't ask the translators to
priorize their updates depending on where better improvements have
occurred. Or can we? We in fact do have a priority list. I can reorder
it if needed. The first reason you point is priority zero.

> It would a lamentable state of affairs if we had translations that are
> not maintained because of a lack of translators and manpower.  But this
> and some other mailings on the translator list rather suggest to me that
> we _might_ have manpower and persons feeling responsible, but that the
> information what they could be doing for the project does not even
> arrive at their doorstep.

Your last phrase probably misses a word or two but I think I get it.
Speaking of myself, freezing during 2.16 launch process has cooled my
translations as well, because it lasted too long IMO. Nex time I beg
(yet again) to fork the translation branch at the same time as stable
is forked from master.

> And that would not be lamentable, but stupid.  If we have people
> enlisted and willing to do a job, we should not let their availability
> go to waste.

Very true. In a case (Czech translation by Pavel Fric) he has done an
amount of work which never got published due to a pile of
technicalities and problems such as structure, xrefs, snippet sync
etc. , along with the fact that many changes were occurring at the
same time in originals.

Pavel, please let's try again.

> Are there people willing to do translations, particularly those who
> actually _are_ already subscribed to the translator list, that are
> waiting for instructions or whatever else before they know how to get to
> work on making their language better supported with LilyPond?  If so, we
> need to organize this effort better and more transparently.

I agree.

--
Francisco Vila. Badajoz (Spain)
www.paconet.org , www.csmbadajoz.com


Reply | Threaded
Open this post in threaded view
|

Re: How to extract PDF of translated documentation?

Francisco Vila
In reply to this post by David Kastrup
2013/1/14 David Kastrup <[hidden email]>:
>> Probably the CG should be updated, improved and simplified.
>> I think that at the moment it would be a bit hard for a new translator
>> to get in.
>
> And I think we could make of several new ones or, probably needing
> similar measures, reactivate some old ones.
>
> It would also be nice to have some redundancy.

Redundancy of contributors, or redundancy in CG to make things much
easier to find?

>  I know that I am not
> particularly fond of the German translation since its language is partly
> awkward, in all likelihood partly due to the need of thinking of German
> language structure and vocabulary independently from the English one:
> the language flow and word choice is often stumbling a bit too close to
> the English version.  Good "literary" translations are actually
> something that does not come naturally even to people who are perfectly
> able to express themselves well in either language in separation.

Yes. Translators: please bear in mind that making a good translation
means, first of all, loving your own language. Read what you wrote and
think whether you would have written that way in the first place. Read
it aloud and decide whether it is 'good ISOLANG'. Also, do not
hesitate to disconnect, whenever it is advisable, the translation
language flow from English language flow.

> I can't actually speak for more than German here, but I would guess that
> other translations might also bit from some independent polishing up by
> people able to focus just on the target language rather than the
> translation process.

Yes but "literary" translations are personal and not always one is
willing to let someone else to amend a work you consider "in a
delicate balance" of word choice, for example. A great capacity of
consensus and team work skills are needed.

> Trevor is doing a tremendous job casting technobabble and information
> into comprehensible English, but that is quite an undertaking with slow
> progress, and I consider it somewhat comparable as it is also sort of an
> English->English process.  While he is more freely rewriting information
> than a "mere" translation usually would, the latter still needs a solid
> dose of willingness to make more sense in the translation than a literal
> translation would produce.

I understand. Sometimes I have encountered changes that did not
require any change at all in the translation.

> So how can we improve the efficiency of our current translation work
> force, and how can we make it easier for people to help in a way that
> does not step on anybody's toes and actually leads to better results?

I don't have an immediate answer and, frankly, I am a bit short of
time to think about this. Output of 'make ISOLANG=es
check-translation' is 9729 lines long.

fr --> 254
nl --> 326
it --> 943
ja --> 8346
de --> 24989 (!)
--
Francisco Vila. Badajoz (Spain)
www.paconet.org , www.csmbadajoz.com


Reply | Threaded
Open this post in threaded view
|

Re: How to extract PDF of translated documentation?

Francisco Vila
In reply to this post by David Kastrup
2013/1/14 David Kastrup <[hidden email]>:
> We have a translation branch, but for typos, I think (correct me if I am
> wrong) that of little relevance as we have back-and-forth merging.

Right.

> Factual errors occuring only in one translation should be fixed there.
> Factual errors inherited from the English version should be fixed there
> first.  If they are already fixed there, then the _whole_ relevant
> passage should be translated, and commit ids adapted, namely then the
> job should be done right, and likely in the translation branch.

We can use the translation branch as safety net in any case, provided
that it gets merged into staging once a week at most. I used to do
this but you have done a few merges and now I don't know if I'm still
responsible of it.  Result of make translation-status has to be
committed, also.

> I have no big clue here, but that is more or less my gut feeling.  I
> don't know how one would approach retranslating or reworking an existing
> passage.  Do the regular translators get to see Rietveld reviews?

I think not in general.

> A retranslation by a non-translator would seem to warrant proper review,
> but it would be somewhat pointless if the actual main translator only
> notices the review once it has been committed.

Please don't forget to contact/cooperate directly with the main
translator first. It would be good to have more listed, regular
translators/retranslators, so the main translator should pose no
obstacle to this, but ask first in any case.

--
Francisco Vila. Badajoz (Spain)
www.paconet.org , www.csmbadajoz.com


Reply | Threaded
Open this post in threaded view
|

Re: How to extract PDF of translated documentation?

David Kastrup
In reply to this post by Francisco Vila
Francisco Vila <[hidden email]> writes:

> 2013/1/14 David Kastrup <[hidden email]>:
>>> Probably the CG should be updated, improved and simplified.
>>> I think that at the moment it would be a bit hard for a new translator
>>> to get in.
>>
>> And I think we could make of several new ones or, probably needing
>> similar measures, reactivate some old ones.
>>
>> It would also be nice to have some redundancy.
>
> Redundancy of contributors, or redundancy in CG to make things much
> easier to find?

Redundancy of contributions.  We currently have "one person does the
translation once and nobody including himself looks at it ever after".

As you aptly observed, that's the best you yourself can barely manage
for Spanish.  But I can't believe you to be the only LilyPond user
natively speaking Spanish.  The main qualification we should be
requiring from translators and translation revisors is good command of
language.  Not ability to juggle with git and other technical stuff.

So how can we move in a direction where translation work is less
technically challenging?  I believe that with regard to the program
string translation (gettext, po, pot files or whatever), there is some
sort of process that detaches the language teams from the technical
details.

Maybe we can move more in that direction for the manual texts as well?
I don't have good suggestions here at the moment, but I should think
that there is potential for getting more people involved for making
their own language look better-suited for reading about and working with
LilyPond.

--
David Kastrup


Reply | Threaded
Open this post in threaded view
|

Re: How to extract PDF of translated documentation?

Francisco Vila
In reply to this post by Federico Bruni-2
I think this thread has moved too fast and we could have a problem of
time zones here. Sorry for answering on a one-by-one basis but it
worths it. The thread is asking many questions and I have some
answers. I don't want you to think I am not concerned with the issue.

2013/1/14 James <[hidden email]>:
> Once a lot of checkins to translation have occurred, and someone decides it
> needs to be pushed to staging, does someone do a doc build first or what
> else happens?

Whoever performs the merging, should test a doc build first.

> Is there something that we could add to patchy to help for example with
> testing any new translation builds - I have horsepower and time but not
> translation skills.

John was administrating the Grenouille system for this. No news is
good news when talking about Grenouille.
--
Francisco Vila. Badajoz (Spain)
www.paconet.org , www.csmbadajoz.com


Reply | Threaded
Open this post in threaded view
|

Re: How to extract PDF of translated documentation?

Francisco Vila
In reply to this post by David Kastrup
2013/1/14 David Kastrup <[hidden email]>:
> As you aptly observed, that's the best you yourself can barely manage
> for Spanish.  But I can't believe you to be the only LilyPond user
> natively speaking Spanish.  The main qualification we should be
> requiring from translators and translation revisors is good command of
> language.  Not ability to juggle with git and other technical stuff.

> So how can we move in a direction where translation work is less
> technically challenging?  I believe that with regard to the program
> string translation (gettext, po, pot files or whatever), there is some
> sort of process that detaches the language teams from the technical
> details.

Yes, contributing to translations is currently very very technically
challenging.

The process you mention is updating a .po file with kbabel,
gtranslator (gnome), poedit (windows) or the po-mode of emacs. Yes, it
makes translating easy for everyone. It is based on strings: you jump
to the next fuzzy or untranslated string, translate it, save the file,
send the file, done. You have contributed one string.

> Maybe we can move more in that direction for the manual texts as well?

If we could convert a texinfo manual in a template for po files (a
.pot), then translators should only have to maintain a po file in an
easy way. Here ends my understading of gettext.
--
Francisco Vila. Badajoz (Spain)
www.paconet.org , www.csmbadajoz.com


Reply | Threaded
Open this post in threaded view
|

Re: How to extract PDF of translated documentation?

Federico Bruni-2
In reply to this post by David Kastrup
Il 14/01/2013 20:32, David Kastrup ha scritto:
>> Redundancy of contributors, or redundancy in CG to make things much
>> >easier to find?
> Redundancy of contributions.  We currently have "one person does the
> translation once and nobody including himself looks at it ever after".
>

I've had a look here:
http://lilypond.org/translated.html

and it seems that this is true for any language except french and italian.

> As you aptly observed, that's the best you yourself can barely manage
> for Spanish.  But I can't believe you to be the only LilyPond user
> natively speaking Spanish.  The main qualification we should be
> requiring from translators and translation revisors is good command of
> language.  Not ability to juggle with git and other technical stuff.
>

My proofreader, Luca, is a huge help for the quality of italian words
and also the choice of musical terms (he studied music, while I'm just
self-taught and very ignorant in notation theory).  He's scared away by
the technical stuff (which I'm happy to handle) and I remember that he
offered to help me provided that he didn't have to learn git, texinfo
and the build system.
So technical stuff can be handled by just one person per language.
Editing the italian words of a text file, ignoring any strange texinfo
command, and producing a diff file, is not an hard task... especially
for a LilyPond user :-)

> So how can we move in a direction where translation work is less
> technically challenging?  I believe that with regard to the program
> string translation (gettext, po, pot files or whatever), there is some
> sort of process that detaches the language teams from the technical
> details.
>
> Maybe we can move more in that direction for the manual texts as well?
> I don't have good suggestions here at the moment, but I should think
> that there is potential for getting more people involved for making
> their own language look better-suited for reading about and working with
> LilyPond.

I've never worked at the translation of a manual which uses gettext.
The Debian Handbook can be a source of inspiration:

http://debian-handbook.info/contribute/
http://anonscm.debian.org/gitweb/?p=debian-handbook/debian-handbook.git;a=blob;f=README.translators;hb=master
https://fedorahosted.org/publican/
http://debian.weblate.org/projects/debian-handbook/

Did you know Weblate (it's GPL)?
http://weblate.org/en/

--
Federico


Reply | Threaded
Open this post in threaded view
|

Re: How to extract PDF of translated documentation?

Francisco Vila
2013/1/14 Federico Bruni <[hidden email]>:

> I've never worked at the translation of a manual which uses gettext.

For example the GIMP manual is translated using gettext.

http://l10n.gnome.org/module/gimp-help-2/

a POT per chapter, a PO per chapter/language

The workflow in GNOME is:
 log in
 book (reservate?) a PO
 download latest PO for your language
 translate strings marked untranslated or fuzzy
 upload the file via web (this closes reservation)

> Did you know Weblate (it's GPL)?
> http://weblate.org/en/

Interesting.
--
Francisco Vila. Badajoz (Spain)
www.paconet.org , www.csmbadajoz.com


12