Re: About translating documents (.xml/.sgml) in GNOME

[Bernd Groh <bgroh redhat com>]

Malcolm,

>Just to let people know my thinking (particularly now that I've written
>more code to implement the design): I am not going to worry about this
>particular facet for the moment. I am not ruling it out, but there are
>other things that should be implemented first to cover the 95% case.
>
>I know that I was the one who originally thought this might be a
>sticking point, but based on the feedback I have seen, it seems to be a
>low priority issue. In a subsequent version I will revisit this issue.
>In passing, I think the "two formats" method is a bit doomed, since it
>runs counter to the principle of keeping things familiar and simple, but
>it may turn out to be necessary.
>

I'm just being selfish, I want the one way of PO's -- not two, and at 
the same time, I don't want anyone else to add too many options that *I* 
don't really need. I rather have the essentials finally working 
properly, i.e. extracting all the right tokens from an SGML DocBook 
document. I believe this to be far more important.

>Based on the other pertinent posts in this thread, it appears nobody has
>come up with any other fundamental sticking points. A few people with
>experience seem to have implicitly approved the idea, so I'm just going
>for it. We'll see what happens.
>

As said, I believe the only fundamental thing is a parser to parse in 
SGML DocBook and to identify all the tokens properly and split them up 
nicely, eliminating all tags you don't really need. I would always 
approve anything like that, since *I* do need it! :-)
Don't get me wrong, we've got versions that cover most tokens, but we 
don't have one yet that does all of them and is able to handle malformed 
tags properly, as well as any depth of nesting. This, I believe, is the 
hard part.

And sure, just go ahead! :-)

Cheers,
Bernd

-- 
Disclaimer: http://apac.redhat.com/disclaimer