[tei-council] Internationalization

[Daniel Paul O'Donnell]

Thanks David, this is touching on exactly the question I've been trying 
to ask: how would i18n translations of the TEI handle the elements they 
*don't* translate. You are assuming the solution would be like tite (two 
namespaces). But is it wrong to copy the non-translated ones to the same 
namespace as the corrected ones?

I can see the problem--it implies that the non-translated ones are not 
the same as the canonical elements in the main TEI namespace and that 
this might encourage diversion over time. But I can also see people 
finding it confusing to use two names spaces from what is (from a human 
perspective) "really" an attempt at a single, albeit (partially) 
localised, TEI encoded document.

Moreover, if an element in a localisation is left with the same name as 
the canonical element, it may be simply because the same name works in 
the localiser's language as well: if I rename tei:choice to tei-fr:choix 
because choix is French for choice, but leave tei:biblStruct the way it 
is because that also works in French (maybe better even than in 
English), is it really true to say that this is tei:biblStruct rather 
than tei-fr:biblStruct?

Anyway, this may really be theology, since we have a working model for 
Tite, and in the Tite context this confusion shouldn't arise since we 
are treating it like a tool to be used by accomplished encoders--and 
emphasising that it shouldn't be used for archiving.

We should probably make sure that the documentation says this before the 
benefit (assuming it works out) is announced in November.

David Sewell wrote:
> I don't think it is a problem in the context of Tite. But the question
> was about internationalization. Suppose you have a French customization
> where roughly half of the standard TEI elements are renamed, e.g.:
>
> 	add	->	ajout
> 	del	->	supp
> 	choice	->	choix
>
> and you are then editing a file that looks like this:
>
> <p>Qui a picqué <choix><supp>la plume</supp><ajout>le
> bic</ajout></choix> de ma <rs key="001">tante</rs>?</p>
>
> <p> and <rs> are in the TEI namespace, the others in the French
> namespace. But this is not evident from the local context. Is that bad?
> For a newcomer to TEI, it could certainly be a source of confusion.
>
> On Fri, 10 Jul 2009, Sebastian Rahtz wrote:
>
>   
>> David Sewell wrote:
>>     
>>>  so a human editor of a file,
>>> pre-processing, has no way to know which elements pick up their
>>> namespacing from the DTD without looking at the DTD. Hence it may be
>>> unclear which elements are canonical TEI elements and which are renamed
>>> or otherwise custom elements. Whether or not this is truly a "problem"
>>> depends on the context, I suppose.
>>>
>>>       
>> I really can't see this as problem. why is this human editor and why does she
>> care
>> whether elements are canonical or not in Tite? the end user editor will have
>> had it
>> translated to real TEI,  and the digitizing editor doesnt care.
>>
>>
>>     
>
>   

-- 
Daniel Paul O'Donnell
Associate Professor of English
University of Lethbridge

Chair and CEO, Text Encoding Initiative (http://www.tei-c.org/)
Co-Chair, Digital Initiatives Advisory Board, Medieval Academy of America
President-elect (English), Society for Digital Humanities/Société pour l'étude des médias interactifs (http://sdh-semi.org/)
Founding Director (2003-2009), Digital Medievalist Project (http://www.digitalmedievalist.org/)

Vox: +1 403 329-2377
Fax: +1 403 382-7191 (non-confidental)
Home Page: http://people.uleth.ca/~daniel.odonnell/