[tei-council] Preparation for F2F meeting

[James Cummings]

Hi all,

So this would be less of a validator but more of a TEI document 
'checker' which not only listed whether it was valid TEI (Or 
noted that it was P4 and stopped there) but listed things in 
other namespaces, and other convenient information (I could see 
things like exposing appInfo, prefixDef, what elements it uses, 
rendition styles etc.)

As part of TEI Simple we're supposed to develop a way to 
document/display TEI Performance Indicators -- by which I'm told 
we mean things like whether we've marked up names (and to what 
granularity?) or not in this document, or whether there just 
aren't any in this document instance, etc.  So informing users 
what they could potentially use in this document.  I'm completely 
unclear on how that might work or be formulated, but if so, then 
such a 'checker' could also list this kind of information?

-James


On 16/11/14 15:14, lou wrote:
> Apologies, I should have been clearer: if a document has elements from
> other namespaces, it can obviously still be valid TEI: validity is
> assessed with reference to the TEI namespace elements only.
>
>
> On 16/11/14 18:42, Hugh Cayless wrote:
>> Yeah, it’s most definitely worth discussing. As to Lou’s point: we do permit customization and the use of other schemas in TEI docs. Not being valid against TEI All doesn’t mean something isn’t TEI, it just means it isn’t only TEI, or is TEI that’s been augmented. If we’re talking about this as a tool for people who aren’t able to validate documents on their own, I’m not sure I’d want to be too dogmatic about it, particularly if we aren’t giving detailed information about why it doesn’t validate.
>>
>> Hugh
>>
>>> On Nov 16, 2014, at 11:47 , Peter Stadler <stadler at edirom.de> wrote:
>>>
>>>
>>> Am 16.11.2014 um 11:37 schrieb Hugh Cayless <philomousos at gmail.com>:
>>>
>>>> I worry a bit about a simple validation tool: what if someone checks a P4 document against it, for example? It might be perfectly valid P4, but come back as invalid. It seems to me that’s not very helpful…
>>> Yes, that’s a good point.
>>> I think we need something in between valid and invalid. (Maybe ‚conformable‘ ;)
>>> In my validation service I plan to implement various fallbacks. If a file is not valid against the current tei_all than try with 2.6.0, etc.
>>> Other schemas such as tite or P4 can easily detected by looking at the root node.
>>> If those files validate against the respective schema, I mark them ‚yellow‘ (and still lack a name for it).
>>>
>>> In parts, it’s a political issue what to call ‚valid‘ but I think we should discuss it …
>>>
>>> Best
>>> Peter
>>>
>>>
>


-- 
Dr James Cummings, James.Cummings at it.ox.ac.uk
Academic IT Services, University of Oxford