[tei-council] Things needing translation

[Lou Burnard]

On 02/01/14 16:24, Martin Holmes wrote:
> On 14-01-02 06:47 AM, Sebastian Rahtz wrote:
>> On 2 Jan 2014, at 14:10, Lou Burnard <lou.burnard at retired.ox.ac.uk> wrote:
>>
>>> I disagree. Putting @versionDate on <exemplum> is a bad idea, for the
>>> reasons you give above. I would vote for removing @versionDate from all
>>> <exemplum>s
>> if you think its  a bad idea, you want it remove from att.translateable, then?
>>
>> On reflection, I am inclined to agree with you that they simply shouldn’t be dated at all,
>> as the only point of dating is to compare translated with original.
> Surely it would be useful to be able to generate a list of <exemplum>s
> that use element X or attribute Y, and which have not been updated since
> before a change we made to element X or attribute Y?
>
>


By the same argument, every child of a *Spec element should be dated!

If we make a change which invalidates an example, we will detect it (I 
hope) irrespective of any use of versionDate. If we make a change which 
does not invalidate an example, why would we want to check it (other 
than in the normal way of checking that it's still a nice example)?

Introducing versionDate on exemplum gives the completely misleading idea 
that exempla in different languages are simple translations one of 
another. There are many reasons why we might decide to change an example 
in language X: it does not follow *at all* that these would necessitate 
comparable changes in languages Y or Z.

There is also the rather tricky technical issue that <exemplum> is not a 
member of att.identifiable, so you cannot point to it in the same way as 
other components. However, I think that's a separate issue (it would be 
nice if it were identifiable btw, because then you could suck exempla 
into the text in the same way as you currently can for <desc>)