[tei-council] FW: Further update on PH

[Conal Tuohy]

On 18/09/2007, Sebastian Rahtz <sebastian.rahtz at oucs.ox.ac.uk> wrote:
>
> Lou Burnard wrote:
> > 1. Disallow the simplified use cases in which co-ords dont figure (i.e.
> > the cases of <facsimile> containing just <graphic>s, or <surface>
> > containing alternate <graphic>s)
> >
> I am with Lou half-way on this. I don't mind giving up the
> list of <graphic> as direct child of <facsimile>, I think I do mind
> being forced to put each one in a <zone>. It gives an impression
> of precision that may not exist.

We're dealing here with the case where there's more than one image
(<graphic>) of a particular page (<surface>), then, isn't that right?
Because if there were only one image per page then we could just use 
<pb facs="page1.png"/>

So the question is whether the <graphic> elements for the <surface> should
be grouped into <zone> elements, to indicate that they cover the same area
or different areas of the page (whether or not the coordinates of the areas
are actually specified, with a @box). My preference would be to always have
a zone, whereas the earlier version says that if 2 graphics exist as
children of a particular surface, then those graphics are considered to
cover the exact same area ( i.e. it is equivalent to enclosing them both in
a single zone). I guess this would cover cases where you have "master" and
"access" images for the same page, but cases such as high-res detail images
would require a parent zone (since the graphics would cover different areas
of the page).

Anyway ... I've given my justification for why I prefer the explicit markup,
but I am happy to go along with implicit zones (and even implicit surfaces)
as long as their semantics are clear (i.e. what the @box value of the
implicit zones and surfaces could be assumed to be, if anything).

Con