Re: Mapping...

Hi all,

Just some remarks and a proposal on "local ITS":

- We have not yet tried to relate the differences between mapping,
pointing, ... to what is happening with local usage of ITS
- Local usage of ITS follows always the pattern "select s.t. and add
information to it", just like the global usage of ITS in

<its:translateRule its:select="//p" its:translate="yes"/>

The only difference being that local usage of ITS works only with
default selections.

IMO we should try to make that commonality clear.


- Felix


Yves Savourel wrote:
> Hi Sebastian,
> 
>> If/when we do return to mapping,
>>  <its:langMap its:mapName="myLang"/>
>> isn't sufficient to deal with my examples in the
>> Mandelieu Communique.
> 
> 
> After talking with Felix today I realized that for language
> identification, we can have a 'normal' rule. Actually the proposed notation:
> 
> <its:langRule its:selector="//*" its:langMap="@myLangAttribute"/>
> 
> Is fine (except for the langMap name). But I see it as a normal
> slection+information rule, not a mapping.
> 
> It simply says: For any node of "//*" go fetch the language identifier
> information in @myLangAttribute and apply it to the node.
> 
> Just like <its:locInfoRule its:selector="//p" its:locInfoMap="@myNode"/>
> says: for any node of "//p" go fetch the note in @myNote and apply it to
> the node.
> 
> So, we certainly can have a langRule without involving 'mapping'.
> 
> As for mapping, I'm not sure at all how to express it. I think we would
> first need to know how it would be utilized besides using it for what
> the <zzzRule> elements already provide.
> 
> Cheers,
> -yves
>