Re: Updated Working Draft "Best Practices for XML Internationalization"

Martin Duerst wrote:
> [first, with chair hat on: Mark and others, please try to reduce
>  your quoted content.]
>
> At 03:45 07/07/01, Mark Davis wrote:
>   
>> The title appears misleading. There are multiple ways to internationalize XML documents. Only a few of the practices are general; the thrust of the document appears to be using ITS to do so, so a more apt title would be
>>
>>     
>>> Best Practices for XML Internationalization
>>>       
>> =>
>> Best Practices for XML Internationalization using ITS
>>     
>
> The document should definitely be general, not limited to ITS.
> So it's not the title that should be changed, but maybe some
> of the contents.
>   

agree and some background: I asked at the beginning of this thread on 
purpose for comments especially on Best Practices 1-6, since the other 
Best Practices statements are not deeply discussed by the WG, and there 
might be additional ones necessary as well.

For the latter aspect: if people on this list have input for new, not 
ITS specific Best Practices which are not in the document, please 
provide them here. We (the ITS Working Group) can't promise to take all 
into account due to time restrictions, but we will do our best.

> Of course, there are several ways to internationalize documents,
> and this should be taken into account. However, ITS is a W3C
> Recommendation, so using lots of examples from ITS and listing
> ITS as the first choice, etc., seem to be appropriate.
>
>
>   
>>> Include <http://www.w3.org/TR/REC-xml/#sec-lang-tag>xml:lang in your DTD or schema to allow to specify the natural language of the content
>>>       
>> =>
>> Where necessary, include <http://www.w3.org/TR/REC-xml/#sec-lang-tag>xml:lang in your DTD or schema to allow to specify the natural language of the content. 
>>
>> [why? because an XML document that just has locale-independent information like inventory counts of part numbers doesn't want to have this. Ditto below.]
>>     
>
> Agreed, but the wording should be different. "where necessary" doesn't
> say anything specific. I'd go for a wording more along the following
> lines:
>
> Include <http://www.w3.org/TR/REC-xml/#sec-lang-tag>xml:lang in your DTD or schema to allow to specify the natural language of the content for all elements that may contain natual language.
>   

sounds good to me. Editorial nit: s/natual/natural/ .

>   
>>> Make sure the <http://www.w3.org/TR/REC-xml/#sec-lang-tag>xml:lang attribute is available for the root element of your document, and for any element where a change of language may occur.
>>>       
>> =>
>> If you documents can contain text of different languages, make sure the <http://www.w3.org/TR/REC-xml/#sec-lang-tag>xml:lang attribute is available for the root element of your document. If it can contain mixed languages, make sure it is available for any element where a change of language may occur. 
>>     
>
> The "changed of language" may easily be misunderstood to have the
> markup mean 'from now on', rather than 'for this nested element'.
>   

+1

>
>   
>> Same changes for other cases, like #2, #7,...
>>
>>     
>>> Best Practice 19: Use CDATA sections with caution 
>>>       
>> I'd like to see this be:
>> => Best Practice 19: Avoid CDATA sections wherever possible
>>     
>
> I'd tend to agree here.
>   

agree. I also agree with the comment from Mark on "Best Practice 21: 
Ensure any inserted text is context-independent".

Felix

Received on Monday, 2 July 2007 00:29:02 UTC