[tei-council] witList/Group query

Matthew James Driscoll mjd at hum.ku.dk
Mon May 21 10:21:47 EDT 2007


Dot and you and Matthew. We discussed this in Berlin, I remember, although I
can't now remember what we decided (apart from changing witList to
witGroup). It seems clear enough, though, that witness is essentially a
pared down msDesc, with only those subelements necessary for describing
manuscripts in a specific context or perhaps rather a specific role, as text
witnesses. My feeling is still that as such it's not necessary.

Whether msDesc can be used for TBOs other than manuscripts is obviously a
much larger issue.

But as regards your question, Lou, I think James's answer, that a witness is
only likely to be part of a single witGroup AT ANY ONE TIME is the right
one. Nesting should be able to take of any overlaps.

Matthew

-----Original Message-----
From: tei-council-bounces at lists.village.Virginia.EDU
[mailto:tei-council-bounces at lists.village.Virginia.EDU] On Behalf Of James
Cummings
Sent: Monday, May 21, 2007 1:02 PM
To: Lou Burnard
Cc: TEI Council
Subject: Re: [tei-council] witList/Group query

Lou Burnard wrote:
> Thanks Daniel, but this does not address the question. By "overlapping"
> I did not mean "containing". Let me rephrase the question: if to witness
> A is allocated to witGroup sigma, does that always preclude it
> simultaneously being allocated to witGroup delta?  I have modified the
> text on the assumption that it does, but would just like to check that
> that is indeed the case. (The former state of affairs, using the
> attribute @included, did not enforce such a rule, of course)

I'm not sure, I suppose it makes sense that at any one time a witness is
part of a particular witGroup and that precludes it being part of any other
(non-ancestor) witGroup.  But I'm not confident enough in that to say that
it always precludes it.  I view these groupings as possibly very fluid and
dependent on future analysis of the apparatus in the document instance.

> More significantly, looking at this again, I really want to know why we
> need this element <witness> at all. Why isn't it a (possibly
> abbreviated) <msDesc>?

That is the problem, I believe that Dot and I had with Gautier's proposed
revisions, most of what he wanted to do, I believe, could be done in an
msDesc.  But, the short answer being, I suppose, that msDesc is for
manuscripts, and not all witnesses are manuscripts.  Thus, some people
might feel it is semantically inaccurate to use msDesc for other
text-bearing objects which witness the same text as some manuscripts.

-James

-- 
Dr James Cummings, Oxford Text Archive, University of Oxford
James dot Cummings at oucs dot ox dot ac dot uk
_______________________________________________
tei-council mailing list
tei-council at lists.village.Virginia.EDU
http://lists.village.Virginia.EDU/mailman/listinfo/tei-council



More information about the tei-council mailing list