ubuntu-manual team mailing list archive

[Ask Hjorth Larsen <asklarsen@xxxxxxxxx>]

On Tue, Jan 11, 2011 at 6:25 PM, lafeber-dumoleyn2
<lafeber-dumoleyn2@xxxxxxxxx> wrote:
>
> Op 10-01-11 12:21, Valter Mura schreef:
>
> In data mercoledì 22 dicembre 2010 13:06:11, c7p ha scritto:
>
> Hi Hannie,
> the fact that lp isn't acting as we would like as translators is a known
> and the only thing we can do is copy n paste the lucid-e1 string to
> lucid-e2 string and translated the minor change.
> Lp is speciallized for translating software messages/strings not paragraphs
> of a book and because of that we are trying to find an alternative solution
> for example a new translation service or platform for the manual editions
> to come.
>
> What am proposing you to do is not to translate the lucid-e2, since the
> changes are minor and i don't think that they worth the effort of
> retranslating the book again -that's my point of view-.
>
> I hope we will find the solution soon, if you have to propose a solution or
> an idea, fire it away!
>
> For me the alternative is only one, at the present time:
>
> - download the package, translate offline and create a translation memory, then
> upload again the file.
> - the next time the file will be updated, you have the old tm ready to be used
> in your CAT tool, so minor changes will be minor headaches for us :)
>
> Choose the CAT tool you prefer, I personally use Lokalize for Interface and
> Help KDE/OOo/Libò files, and OmegaT for other stuff.
>
> I wonder if Launchpad will reintroduce the fuzzy state option (marking the
> string with a percentage of match), very useful in these specific cases.
>
>
> Best regards,
>
> Hello Valter,
> The manual is a special case. It is an advantage that it can be translated online through Launchpad as long as we have to do with the first time strings are translated. In our case that was Ubuntu Manual Lucid Lynx. But Launchpad was not able to transfer the translation of the same strings to the next version of the manual, in this case Maverick Meerkat. Now we get a kind of strange situation which I will illustrate using the following example:
>
> English string in Lucid-e1:
> \newglossaryentry{router}{name={router}, description={A router is a specially designed computer that using its software and hardware, routes information from the internet to a network. It is also sometimes called a gateway.}}
> Dutch translation in Lucid-e1
> \newglossaryentry{router}{name={router}, description={Een router is een speciaal ontworpen apparaat dat, met behulp van speciale software en hardware, informatie van het internet naar een netwerk stuurt. Het wordt soms ook wel een gateway genoemd.}}
> Translated by Kenneth Venken on 2010-08-13
> Reviewed by Redmar on 2010-08-22
>
> Dutch translation in Maverick Meerkat
> \newglossaryentry{router}{name={router}, description={Een router is een specifiek ontworpen computer waarmee, gebruikmakend van zijn software en hardware, informatie gerouteerd wordt van het Internet naar een netwerk. Het wordt soms ook wel een gateway genoemd.}}
> Translated by kwoot on 2010-08-26
> Reviewed by Kenneth Venken on 2010-09-28
> Look at the names and dates and you can see what happened here (even though you do not understand Dutch, the differences are clear). And this is just ONE example! I think this is a waste of energy and it leads to different translations, even if the same translator is involved.
> Now, about your suggestion: working offline has the same disadvantage. Since Lucid and Maverick are not merged you can only transfer translations from LL to MM manually, and that is just as much work as copying in LP from LL to MM.
> I am still just as unhappy about the situation as I was when I started this thread, but I do appreciate the many reactions and suggestions that were put on the list. With that I hope we can get to a solution together.
> Regards,
> Hannie

What do you mean when you say that you cannot transfer them because
they are not "merged"?