[tei-council] layers again

[Kevin Hawkins]

On 10/15/11 10:18 AM, Lou Burnard wrote:
>> So here's what I don't like about it.  Until this moment I had thought
>> of the profileDesc as describing the TEI document, not the source
>> document.  Indeed, Wendell Piez, sitting next to me right now, agrees.
>> But when I read section 2.4, and the examples given with the definition
>> of<creation>, it seems that parts of the profileDesc actually describe
>> the source document.  This is all ontologically problematic.  Maybe we
>> don't want to deal with this larger problem right now, but at least I
>> want to make sure we're aware of this ambiguity.
>
>
> I'm not sure what the problem or ambiguity is. That you and Wendell
> should independently (I assume) have reached a wrong conclusion about
> this particular aspect of the semantics of some Header elements is of
> course a matter of concern, so if you can point to any text which led
> you to that conclusion we should certainly try to correct it. The
> profileDesc was originally conceived as a kind of catch all for all
> sorts of non-bibliographic aspects relating to the source, hence the
> presence of both text classification and creation info here.

There is one reason that I previously did not think that profileDesc 
describes the source document:

1) Months ago Lou sent me a correction to 
http://www.ultraslavonic.info/talks/20100407b.pdf (which I've since 
made) yet didn't point out that slides 7 and 8 claim that profileDesc is 
about the TEI document and not the source document.

And there's one reason why I still don't think that profileDesc 
describes the source document:

2) Section 2.4 of P5, and of the elements therein, all make reference to 
"the text".  While this could refer to the source document (as Lou says 
it does), and while I would like it to be more specific if it does, now 
that I think about it, I think that instances of "the text" more 
properly refer to the platonic object represented by either the source 
document or the TEI encoding.  I could imagine wanting to encode 
keywords that apply only to a digital document and not to the source 
(since they may differ significantly, especially if the TEI document 
brings together multiple sources) or imagine using any of the 
profileDesc elements for a document composed by scratch in TEI.  As I 
understand Lou, you shouldn't use profileDesc for either of these.

As for the matter at hand, I'm not sure whether layers/campaigns are 
really related to the platonic object or to the source document.  If 
they relate to the platonic object, then I have no objection to the 
proposal.  (If they relate to the source document, then Lou has no 
problem with the proposal.)