RE: What does "attaching ITS information" mean?

On Mon, 2006-12-11 at 08:31 -0700, Yves Savourel wrote:
> 
> > sorry, I don't understand: what attributes do you want to use? 
> > Also, in the case of <p its:locNote="Some locale note"> versus 
> > a global localization note: would you put all contents of the 
> > localization note in <locNote_Text> element, no matter if the 
> > note comes from global or a local ITS markup? My question again 
> > would be: were to we describe this in the draft, since it is 
> > different to e.g. its:translate or its:withinText ? I'm also 
> > asking because my "generic" ITS processor does only understand 
> > what is in the draft, so it is not able to produce decide 
> > "should I produce just text (in the case of locNote) or copy 
> > ITS markup? (in the case of its:translate)" ;)
> 
> Ok, I see your point. The draft is not quite clear on what to do with
> the information provided by the data categories: especially, we don't
> really say what 'adding' means in term of results.
> 
> The whole 'adding' part is section 5 is a bit strange. If it refers to
> decorating the tree, it should also do that ('add') with the info it
> 'points' to.
> 
> So, for the test result format we should be able to do anything we want,
> just like a real application.

Sorry, I would put it differently: for real applications we should be
able to do what we want, but for the result format we need to agree on
what is produced. Otherwise, there is no way to compare e.g.
<output><locNote_text> ... 
with
<output its:locNote="...">
or
<output><its:locNote> ...
I'm fine with any way of generating output, as long as we agree upon it
explicitly for each data category. That is not necessary for the spec,
but only for comparability of results.

> 
> I guess, what I'm trying to say is that one cannot have a "generic"
> application that does the same thing for all data categories, since we
> don't define what 'adding' or 'pointing' really means.

I think you could have that, but maybe we don't need to discuss further.
I think we agree that we don't need it for real life ITS applications.

Felix

Received on Monday, 11 December 2006 16:48:12 UTC