[tei-council] divliminality

[Lou Burnard]

And further still, I can report that we have actually already done some 
of this:

a) signed now has a content model of paraContent. It isn't allowed at 
top of div as well as bottom since this would lead to ambiguous content 
models

d) cell also now has a content model of paraContent



On 17/10/13 21:19, Lou Burnard wrote:
> Further to this, I've now checked and I can find a SF ticket  (433) for
> only the last. Am I missing something or have these
> requirements/proposals not been expressed outside the confines of that
> ODD yet?
>
> On 17/10/13 20:55, Lou Burnard wrote:
>> So, do I understand correctly, your recommendation is that consulting
>> the community in this particular case would be a waste of time, and that
>> all we should do is modify the TEI so as to match the practice
>> exemplified in EEBO-TCP's usage (as interpreted by your good self)?
>>
>> Leaving aside other issues of principal, does this in practice mean simply
>>
>> a) signed gets looser content model (paraContent, not
>>             phraseSeq) and allowed to appear at top of div as well as
>>             bottom
>>
>> b) stage gets placement attributes, and allowed in model.phrase
>>
>> c) trailer gets looser content model to let it contain <gi>l</gi>
>>
>> d) cell gets looser content model (specialPara)
>>
>> e) salute gets looser content model (paraContent), allowing list
>>
>> (I took these from a file called tcp.odd I found -- no doubt some of
>> them have corresponding sf tickets but my wifi connexion is a bit flakey
>> at present so I havent looked for them)
>>
>> Or are there other more substantive issues?
>>
>>
>>
>> On 17/10/13 18:42, Sebastian Rahtz wrote:
>>> On 17 Oct 2013, at 18:18, Lou Burnard <lou.burnard at retired.ox.ac.uk> wrote:
>>>
>>>> Well, I am not sure that I understand your suggestion for an alternative here. If you know what the "alternative models" are, you are free to propose them, obviously, and I hope you will.
>>> I did, you did, Paul did, we all did, in Ann Arbor.
>>>
>>>> My recollection is that after several hours debate the Council was unable to agree on which model should be recommended, whence the suggestion to consult the community. As to "kicking into the long grass" I think this idea is pretty well buried in long grass already so we can't make things much worse by seeking for guidance from the community.
>>> What I care about is being able to use EEBO TCP in TEI P5 before the end of 2014. This is a blocker.
>>>
>>>> Do we have a seriously better plan?
>>> Yes.  follow what EEBO TCP have done, and adjust Gidlines to suit. Argue at leisure about what it all _means_, joining
>>> the other N% of the TEI which is not precisely defined….
>>>
>>> don't get me wrong, I do agree that its a jolly interesting subject, how to encode bits and pieces
>>> at the top and the bottom  (the ladder of law has no top and no bottom, however), but is it worth spending
>>> 2 years watching the angels dance on their pins?
>>> --
>>> Sebastian Rahtz
>>> Director (Research) of Academic IT
>>> University of Oxford IT Services
>>> 13 Banbury Road, Oxford OX2 6NN. Phone +44 1865 283431
>>>