[tei-council] Re: figurin' about <figure>

Wed Jul 25 05:25:30 EDT 2007

Syd sent me and Christian the following musings about <figure> which I 
am forwarding to the list for discussion, along with my immediate reactions:

Syd Bauman wrote:
> In P4, the content model of figure (ignoring globals) was
>  ( head?, p*, figDesc?, text? )
> 
> In P5 it has become 
>  (  model.pLike | model.global | figure | figDesc | model.graphicLike
>    | model.headLike )*
> 
> So first off, it seems that either <text> or <floatingText> needs to
> be permitted somewhere in there. I'd vote for the latter.
> 

certainly not <text>in any case

> But I am also a bit uncomfortable with the content model permitting
> <head> anywhere other than as the first textual element (other than
> globals). The Guidelines explicitly say that the transcription of the
> caption or descriptive paragraph should occur after the <head>. Do
> you want to defend permitting
>    <figure>
>      <p/>
>      <graphic/>
>      <figDesc/>
>      <head/>
>    </figure>
> ?
> 

I think this is preferable to complicating the model.

> I also thought it might be prudent to take advantage of the use of
> classes, and use <ab> where the semantics of <p> aren't really right.
> However, I can't really tell *what* the <p> in one of our examples
> is: "<p>Thou shalt labor till thou returne to duste</p>" in the
> tagdoc for <titlePage>. Can you shed some light?
> 

I agree that <ab> might be more appropriate than <p> for random biyts of 
texts floating about inside an image.

> In any case, I think both in the prose and in an example we should
> recommend using <ab type="caption"> for a caption. After all,
> depending on your defintion of "paragraph", a caption isn't one.
> (Might think of using <label> for captions, I suppose.)
> 

Not sure what the difference between a caption and a head is within a figure

> Lastly, if we permit <figure> to self-nest, we should probably say
> somewhere what that is for, and perhaps give an example of its use.
> Is there a good use case for permitting it to self-nest anywhere, or
> would it be OK to stick it in at the end?

I thought there was an example of this: it's for things like complex 
figures with subfigures in them.

> 
> If the latter, then how about this for a content model:
>   ( 
>     model.global*,
>     ( model.headLike, model.global* )*,
>     ( model.graphic, model.global* )*,
>     ( figDesc, model.global* )*,
>     ( model.pLike, model.global* )*,
>     ( floatingText, model.global* )?,
>     ( figure, model.global* )*
>   )
> 

I don't think the added complication of this model is worth the effort, 
frankly.

> If the latter (<figure> needs to be able to self-nest anywhere) then
> we'd want to change every "model.global*" to "( figure | model.global
> )*", I guess. That gets kinda complicated, though.
> 
I don't think the added complication of this model is worth the effort, 
frankly.

> 
> I also don't wonder whether <figure> should permit model.egLike in
> addition to model.graphicLike. One can imagine wanting to associate a
> heading or a coption with an example.
> 

This could work. Or could make model.egLike a subclass of model.graphicLike