Hi Friedel and all :) Sorry I couldn't answer this before.
On 16/01/2007, at 1:05 AM, F Wolff wrote:
Most of my work on [the OpenOffice.org Pootle processes] is finalised, I believe. It is not deployed, andwill probably need another 2 hours of work and testing, but drop me a line when we need to start on this again. I hope I can fit it in.
I'm not able for this yet, but hope to be before too long. Thanks very much for continuing to work on this, Friedel. :)
I (very) quickly had a look at one file in helpcontent2, and I must say,it is an abomination!
Perfect description. ;)
The markup is more than the translatable text.
MUCH more. There are a lot of strings which don't have any translatable content at all! When I complain about them, I'm told I should spend some money and get a "real translation editor" which would cope with all these bad strings.
Furthermore it seems as if many of them have a simple pattern like <tag with detail>translatable text</tag>
Yes, indeed. Even more simply use the <emph> tags to highlight names of menu items etc.
which in my opinion should not be coming to the translator at all (unless there are attributes that should be translated).
There are a few strings which require us to translate the "name" attribute, but most don't. We should be able to switch those on or off.
All of thismeans that you cannot (easily) use your UI translations as a TM to alignyour terminology (which is supposed to be trivial)
Ha! Trivial indeed. You are exactly right. What's more, because OpenOffice.org doesn't operate on the unique string principle, you end up translating very common strings over and over. I believe the huge word count of the OpenOffice.org Help could be reduced significantly, by as much as three quarters, if we cut out all the untranslatable strings and unnecessary code. And _that_ would encourage more translators. How many translators are silly enough to try and translate strings muddled inside half a million words of heavily escaped code?
and the problems with syntax as you mention. It also seems as if they would be able to merge some consecutive strings if they were able to strip the markup.
Oh yes...Not to mention things like repeating strings many times simply to increment the number (Long text... Page 1, Long text...Page 2, Long text...Page 3 etc.).
It seems that angular brackets and quotes are permanently escaped whichjust means more hassle for the translator.
It creates a lot of errors. I've got used to the escapes, but they do create more errors, and they make the whole task that much more visually complex.
If I'm not mistaken, this isn't even consistent with the rest of OpenOffice.org, is it? In otherplaces the angular brackets and quotes are used bare, so there seems tobe different rules for helpcontent2.
I guess I should now be offering to allocate time to write a separate storage class and converters for helpcontent2 which will help a little bit, but in my opinion a more disruptive change from their side might help more.
So far, I have not found it easy to achieve any change at OpenOffice.org. While there are people who are active and keen to help within their sub-projects, there does seem to be considerable difficulty in communicating between sub-projects in some cases. So, for example, changes are already happening within the l10n sub- project (thanks especially to people like Rafaella Braconi and Pavel Janik), but when I mention translation issues in the Help, I'm told that's a Documentation issue, so I have to join more lists and talk to different people. i18n seems to be split between the different subprojects responsible for the original code, not concentrated in its own project, for some reason.
I'm reporting typos etc. in the Help, and hopefully can build up a positive relationship with the Docs project.
I do want to talk to you guys in detail about the OpenOffice.org release process and lowering barriers for translation, but I will have to do that when I have more neurons in one place.
I can fully understand why your team got frightened away. In fact, I'm not quite sure why you are not (well, I have an idea ;-) You should feel free to translate offline and to update the translations on theserver. You can simply upload a ZIP file of the updated files. Just takecare to use the same directory structure so that Pootle can merge them with the existing files.
I've actually persisted on Pootle, mostly because I haven't been well enough to think properly, so I've just been doing what strings I can. I'm lucky the switch between English and Vietnamese still operates automatically in my head. ;)
Currently Pootle will always try to merge the files on upload, which might not always be what you want, but I should have some news on developments in this area real soon.
I notice we now have two buttons: Merge and Overwrite, which is cool. Thankyou!
from Clytie (vi-VN, Vietnamese free-software translation team / nhóm Việt hóa phần mềm tự do)
http://groups-beta.google.com/group/vi-VN
PGP.sig
Description: This is a digitally signed message part
------------------------------------------------------------------------- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier. Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________ Translate-pootle mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/translate-pootle
