>That is both more complicated and error prone to implement and more
>complicated and error prone to do.

No it's not. It is like saying it is better to have a computer off the 
network than having a fail-over like system on the net. Syncing is not that 
bad at all.

>For every version of a translatable product that ever has been release you
>woulld need to have a separate database of translations in a central
>repository. Who is going to manage this central repository?

In the beginning the ZBabel team is going to manage the Master Tower.

>What happens if that location goes down?

You have mirrors of course; it is no different than any other FTP 
distribution site.

>What happens if you want to something on a PC
>before the IT department have opened a friewall so you are blocked from
>accessing the central repository?

Well, if you are blocked from HTML, then you cannot download products in 
the first place. We also plan to have a way of bundeling modules up into a 
file... In anyway, no HTTP = No Products, in which case that means you do 
not need the translations in the first place.

>  If you find a bug in the translation, how
>to you update the repository?

You contact the maintainer of this module and language. Works exactly like 
software development.

>From: "Stephan Richter" <[EMAIL PROTECTED]>
> > Well, no. Because then the products would get too large. The better
> > approach is to allow an online sync, again, which comes with the next
> > ZBabel version.
>If the translations mean the products get too big (which I honestly doubt, I
>have never ever seen any example of a localization that is more than a
>couple of percent of the whole application) it would be no problem at all to
>have the translations downloadable separately if so desired by the

You see, that is exactly what we are doing. We also offer a nice facility 
for people to go and translate. This way the authors of products do not 
have to worry when their phrases are translated into a new language. And it 
will be much faster for the translator, since he can do several modules at 

> >Well, all the i18n developers will have a meeting early January in Europe
> >with Jim Fulton in a 2-day brainstorming session.
>Excellent! When and where?

Beginning of January somewhere in Europe. However, as I understand it, it 
will NOT be a "public" event, since the goal is to produce workable Zope 3 

BTW, with criticism comes usually constructive input. Speaking for me, most 
of the code was written in my free time (I summed up my time for this year 
and I estimated between 400-500 community hours. That is excluding the free 
software I have produced as contractor). If you want to make a difference 
and influence the way the products are developing, you just have to produce 
the majority (>30%) of the new code and everything will work out for you 
much better.
I have also looked for your name on the Zope site, and all I found were 
some comments. If you have already improvements to some of the Zope code, 
why are you not publishing it then? If you have the problem of not knowing 
any parts to improve I will be happy to provide you with a list.


Stephan Richter
CBU - Physics and Chemistry Student
Web2k - Web Design/Development & Technical Project Management

Zope-Dev maillist  -  [EMAIL PROTECTED]
**  No cross posts or HTML encoding!  **
(Related lists - 
 http://lists.zope.org/mailman/listinfo/zope )

Reply via email to