[tei-council] proposed new <entry/> example with xml:lang attributes

[Stuart A. Yeates]

The more I think about it,

 <etym>
  <lang>French</lang> <mentioned xml:lang="fr">oui oui</mentioned>
 </etym>

should actually be

<langUsage>
 <language ident="fr" xml:id="french">French</language>
  ...
</langUsage>

...

 <etym>
  <lang corresp="#french">French</lang> <mentioned xml:lang="fr">oui
oui</mentioned>
 </etym>

cheers
stuart


On Fri, Jul 6, 2012 at 7:05 AM, stuart yeates <syeates at gmail.com> wrote:
> On 05/07/12 22:38, Piotr Banski wrote:
>>
>> Hi all,
>>
>> I somehow managed to miss the @corresp proposal, need to search the past
>> messages, I guess.
>
>
> There's a good long example / discussion at:
>
> http://www.tei-c.org/release/doc/tei-p5-doc/en/html/ref-att.global.linking.html
>
>
>> I had the impression that one would normally go for the same "box"
>> containing the translation (equivalent), whether it corresponds to the
>> <trans> or <dicteg> of old:
>>
>> equivalent:
>>
>> <cit type="translation">
>> <quote>--equivalent--</quote>
>> </cit>
>>
>> example:
>>
>> <cit type="example">
>> <quote>--example here--</quote>
>> <cit type="translation">
>> <quote>--translation of the example--</quote>
>> </cit>
>> </cit>
>>
>> On this approach, there is probably (?) no need for @corresp, because
>> the relevant relationship is expressed by containment.
>
>
> If that is true (and it may be) then we can write xslt to regularize these
> dictionary entries to standard TEI, but I'm not seeing that level of
> consistency.
>
>
>> Or am I missing the point? (scanning mails ultraquickly today, apologies
>> in advance)
>
>
> The point is that I'm trying to be consistent, consistent identification of
> parallel texts, consistent with the use of xml:lang in other kinds of XML.
>
> cheers
> stuart