On Wed, 16 May 2007, John A. Walsh wrote:

> On May 16, 2007, at 9:16 AM, Peter Boot wrote:
>> I don't think it is a good idea, for three reasons:
>> * it should not be the business of document encoding to propose intended 
>> renditions (unless writing new documents). Renditions are decided at 
>> display time based on the characteristics of that context (user, medium, 
>> query, etc.);
> The Guidelines state very explicitly that rend is for describing the source, 
> not "intended rendition." This remains the case in P5 and in the proposed 
> change to rend.  It just so happens, however, that languages like CSS and 
> XSL-FO provide a formal mechanism that may be used for explicitly defining 
> styles that appear in the *source* document, especially for print sources, 
> though often usefully for manuscript sources as well.

True, but it is therefore important to preserve in TEI a means to 
preserve the HTML/CSS distinction between literal style instruction and 
indirect reference, in other words the difference between @class and 
@style on an HTML element. In our local tagging practice (and based on 
past TEI-L discussion, we're not alone in this), we use constructions 
like <hi rend="italic"> in parallel with constructions like
<hi rend="CSS(border-bottom: double; padding-left: 1em;)">. Under
the proposed system, it would be necessary to create a style definition
for each and every one of the latter constructions, which seems
thoroughly undesirable.

I would agree that this is such a large change that it's probably 
rushing things to aim at P5 for implementing it, in any case.


David Sewell, Editorial and Technical Manager
Electronic Imprint, The University of Virginia Press
PO Box 400318, Charlottesville, VA 22904-4318 USA
Courier: 310 Old Ivy Way, Suite 302, Charlottesville VA 22903
Email: [log in to unmask]   Tel: +1 434 924 9973