[tei-council] Comments on Chapter 18, Tables, Formulae, and Graphics.

David J Birnbaum djbpitt+tei at pitt.edu
Wed Sep 12 23:14:15 EDT 2007


Dear Lou (cc Council),

Unless I am also missing something, making it impossible for someone to 
put <lb> elements between <cell> elements would be an improvement 
greatly to be wished.

Best,

David

Lou's Laptop wrote:
> David J Birnbaum wrote:
>>
>>>>
>>> The content model for cell would become ambiguous if we allowed 
>>> <table> within it explicitly, since <table> is a member of 
>>> model.inter, which is permitted within macro.paraContent.
>>
>> Does this mean that <table> is already allowed directly inside <cell> 
>> because <table> is a member of model.inter, which is part of 
>> macro.paraContent, which is allowed directly inside <cell>? 
>
> Yes!
>
>> If so, does this mean that a <row> may contain a mixture of <cell> 
>> and <table> elements as its immediate children, and the <cell> 
>> elements may themselves contain <table> elements? 
>
> Yes!
>
>> If so, since I can't think of how those two ways of representing an 
>> embedded subtable would reflect different semantics, should we remove 
>> <table> as a possible immediate subelement of <row>, so that users 
>> who want a subtable will have to embed it inside a <cell>?
> This was my thought too.
>
>
>> Or am I still confused?
>>
> Far from it, on present evidence.
>>> That said, I think the content model for these elements is quite 
>>> strange: in particular the way model.global is permitted between 
>>> cells and rows. Consider the following:
>>>
>>> <row><note/><cell><note/>stuff</cell><note/><cell/><note/></row>
>>>
>>> To which element is each of the four notes above attached?
>>
>> A very good question, indeed! I suppose the notes are attached to 
>> their parents, so #1 refers to the <row>, #2 to the cell, #3 also to 
>> the row, and #4 also to the row. Why, then (I hear you cry), are the 
>> three notes attached to the <row> located in different places, and 
>> how would it affect the semantics of the document if they were moved 
>> elsewhere within the <row> element?
>>
> Quite so. And what would we lose by saying "this is silly" and 
> disallowing model.global anywhere except inside <cell> (via 
> paraContent, vide supra)?
>
> We'd lose the ability to put e.g. <lb/>s between cells, rather than 
> inside them. I don't think that would make me lose much sleep, but 
> maybe I;m missing something here?
>




More information about the tei-council mailing list