[tei-council] handy

[Syd Bauman]

> I think the idea is that while <handDesc> is used to describe the
> general writing in a manuscript and therefore doesn't necessarily
> need to mention singular hands in it, <handNote> refers to the
> description of a particular hand (be it The scribe's A Hand or one
> of the many interventions of a scribe/author over time on the same
> manuscript).

Right. But my question is why when <handNote>s (or prose descriptions
about handwriting) are stored as a descendant of <msDesc>, their
parent is <handDesc>, but when they're stored as a descendant of
<profileDesc>, their parent is <handNotes>? My gut instinct (which
may well be wrong) is that we could use <handDesc> for both purposes. 


> > <handDesc> has a hands= attribute; if we really don't want that
> > on the <handNote> wrapper when MS is not loaded, then we should
> > put it in an attribute class defined by MS.
> 
> My understanding was that if you loaded PH, than you would get
> <handDesc> which has the @hands attribute. The attribute @hands as
> currently defined wouldn't make sense in <handNote>.

You are correct on both counts. But what I'm trying to poke at is not
the effect of having hands= on <handNote>, but rather of having
hands= on whatever element is used as the <handNote> wrapper:
currently either <handNotes> or <handDesc>.


> Personally I am not in favour of the use of <p> for things that
> 'semantically' can be paragraphs as well as anything else we want
> them to be in the header especially since we have good wrappers. So
> I think we should avoid - if possible (although in MS we have
> paragraph everywhere...sob...).

Not sure I grok what it is you're not in favor of, here. I don't
suppose it will help that you could use <ab> if you don't think it's
semantically a paragraph? Or do you think the descriptive prose
should be allowed as a direct child of, say, <handDesc>? (Or do you
think that descriptive prose should not be allowed, and people should
be forced to use the specialized elements?)


> > If so, I think I'd still be in favor of combining these two
> > elements, and using Schematron to issue a warning if
> > profileDesc/handDesc/p occurs.
> But the whole point was to allow people who don't want to load MS
> to be able to list descriptions of specific hands and point to them
> from chunks of the main text.

And a good idea that is ... but why not use the same wrapper element
-- I'll go out on a limb and just say <handDesc> -- in both cases?
(Yes, <handDesc> would have to be declared in the header module, but
I thought Lou's ingenious plan already did that, no?)


> fI am aware that probably not many projects would use PH without
> MS, but what if they wanted to? what would the @hand attribute
> point to?

I was imagining it would point to a 
/TEI/teiHeader/profileDesc/handDesc/handNote/@xml:id.

BTW, I don't think the idea of using PH w/o MS is at all unlikely,
and moreover that wanting to describe hand writing when using PH but
not MS is also quite reasonable -- I can see it being used for
handwritten additions and notes on typescripts.