Re: [TF-dev] CTI projects completion and restart today

Hi Jean-Francois,

I restarted the projects and now the files are split into smaller files, 
so e.g. for IDE, there are now around 150 smaller files which could help 
you better access the files.

In the meantime, we are looking into other solutions for better 
translations.

Options considered now:
*1) Idiom Desktop Workbench*
-- similar to OmegaT but better handling of fuzzy matches
-- possible usage of terminology database
-- direct import of translation memory into our system
-- under evaluation now
*2) OpenCTI*
-- enhanced CTI that would allow:
-- search function
-- fuzzy matches
-- possibly terminology database (tbc)
-- possible live version in mid of September 09

Please do let me know your feedback on these smaller files, hopefully 
this will solve the problem a bit?

Thank you! :-)
Petra

Jean-François El Fouly wrote:
> Back from short holidays and attempting to resume work on "ide" in CTI 
> I find the situation even worse than I imagined browsing this thread.
>
> Lacking a search functionality, bug fixing has become impossible.
> I had developed a quick hack - which I shared on Kenai - that allowed 
> at least to find the file in which the problem could be. The new 
> structure makes this unusable and I can't imagine an alternative 
> solution now until a good search tool is available in CTI, which has 
> always been the most desirable solution anyway.
> Even for a "first translation" of what's missing, the system has 
> become pretty much unusable. Take ide, which I'm currently working on. 
> Now it has got 467 screens. Suppose I want to translate top to bottom 
> and I'm currently at page 250 (which is some kind of realistic 
> scenario since the translation is 50%). I come back from lunch and 
> relog on CTI. How do I reach page 250 ? It takes 25 clicks and 
> something like 5 minutes just to reach the page I was working on and 
> it's going to be worse when the translation is in progress.
> I can use the filters available from CTI 1.1 to see only the segments 
> that are not translated but I lose the context (segments that are 
> already translated and have been carried on from a previous version) 
> so the probability of an inconsistent result improves tremendously.
>
> I just don't know what to do about this and am just feeling bad. Not 
> sure I'll resume the work right now, I'll probably wait to see what 
> will happen to improve the situation.
>
> Sorry.
>
> Jean-François
>
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [email protected]
> For additional commands, e-mail: [email protected]
>