nuxeo-localizer
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Nuxeo-localizer] translation process vision


From: Juan David Ibáñez Palomar
Subject: Re: [Nuxeo-localizer] translation process vision
Date: Tue, 13 Aug 2002 16:18:14 +0200
User-agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.0.0) Gecko/20020610 Debian/1.0.0-1


The process you describe is the same, you just have replaced the
PO by a text file with a simple format and TMX.

The question is not wether to support PO or TMX, both must be
supported.

The question is where to start, and the answer is to do the simplest
thing that gives quick value and feedback.

There's already infrastructure for PO in Localizer, mainly the export
and import facilities of the MessageCatalog; I would have to implement
it for TMX; so it's easier to start with PO.

I also would get inmediate benefits with PO, because I've tools in my
Debian installation that work with them, but I don't have any tools
here to manage TMX.

So I'll start with PO and implement TMX later. Of course, I will welcome
a TMX implementation.


In a related note. Gettext is intended for translating software, but it's
not much different than a memory translation system. Actually, the poxml
tools of the KDE project use it this way, to manage the translation of
the documentation.

Usually a translation memory system provides fuzzy matching to search
for translations and propose them to the translator, as Gettext does.

But fuzzy matching is just an automatic translation system, just the
simplest one you can build. So it could be posible to plug more complex
automatic translators, as I actually did in the LLEU project when I
worked in Spain.

The difference is that the concrete audience addressed by Gettext is
different than the audience addressed by the TMX standard, then the
available tools are also somewhat different. In one side you've a word
macro that lets to translate from MS-Word, easy for translators; in
the other hand you've an emacs mode to edit PO files, only for developers.


PS: I'm going to reply your other message now :-)




Luistxo Fernandez wrote:

The process, in my vision.

I will follow David´s steps.

(...) Which will be the process?

- The content creator provides the original version of the document;

Step 1. Ok.

For an exanple, we will call this Document A1_en (type: whatever, as in David´s 
proposal). It's an original document.

- The content creator clicks a button in the user interface, the
 action of this button is to parse the document, extract the
messages
 and feed a MessageCatalog object with them;

step 2. The content creator clicks a button in the user interface, the action 
of this button is to parse the document, and feed a MessageCatalog object with 
them; it simultaneously creates a .txt text document where messages are simply 
paragraph after paragraph.

That is the A1_en.txt document. For instance:
-------
About EuroZope

EuroZope is the community of Zope users, developpers and companies in Europe.

EuroZope is focused on evangelization, localization and setting up Zope events 
during computer exhibitions in Europe.

We agreed on a statute, which you can read or download "here.":statute.pdf
-------


- The translator translates the messages with the MessageCatalog
 (he won't have to deal with text formatting!);

Step 3. The translator receives the A1_en.txt document and translates it with 
whatever he uses. He returns the A1_es.txt translated document.
-----
Acerca de EuroZope

EuroZope es la comunidad de usuarios, desarrolladores y empresas que trabajan 
con Zope en Europa.

Los objetivos de Eurozope son la evangelización, y la organización de eventos 
en torno a Zope en ferias informáticas de Europa.

Hemos acordado unos estatutos, que puedes leer o bajarte desde 
"aquí.":statute.pdf
------


- The content creator (or the translator) clicks another button, it
 creates a new language version for the document based on the
original
 one and on the translations available in the message catalog;

step 4. The content creator (or the translator) clicks another button, it 
creates a new language version for the document; then uploads the new A1_es.txt 
text document, which is counter-parsed to reintegrate its messages/paragraphs 
into the message catalog.


- If needed the translated version can be modified.


Now, it´s updating time.

A1_en has to be modified. It will become A1b_en.

---
About EuroZope

EuroZope is the community of Zope users, developpers and companies in Europe.

EuroZope is focused on world domination. We have a well thought out scheme to 
undermine the usage of proprietary software accross Europe.

We agreed on a new statute, which you can read or download "here.":statute.pdf
---

Paragraphs 1 and 2 are not altered. Paragraph 3 is different. Paragraph 4 only 
varies in one extra word (´new´).

So, following the previous process, the translator would receive the A1b_en.txt 
file. But that´s not the only thing he should receive. He should also get a 
translation-memory file, in which the previously translated paragraphs are 
aligned.

This would have required that after step 4 of the previous process, another 
automatic step should gave been taken:

a) the alignment of the A1_en.txt and A2_es.txt files to create a A1_en_es.tmx 
memory.

or alternatively:

b) the alignment of messages in the Message Catalog to create a MC_en_es.tmx 
memory.

So, when the translator receives the A1b_en.txt he also receives the 
A1_en_es.tmx (or MC_en_es.tmx) memory, and proceeds to translate with a good 
basis. He returns A1b_es.txt and the process is once again complete.

Luistxo






_______________________________________________
Nuxeo-localizer mailing list
address@hidden
http://mail.freesoftware.fsf.org/mailman/listinfo/nuxeo-localizer




--
J. David Ibáñez, http://www.j-david.net
Software Engineer / Ingénieur Logiciel / Ingeniero de Software






reply via email to

[Prev in Thread] Current Thread [Next in Thread]