On 23/11/11 20:25, Mylonas, Elli wrote:
> Three years ago, there was one answer, from James Cummings,
> (http://listserv.brown.edu/?A2=ind0811&L=TEI-L&P=R6280) who suggested
> that he would
> "...create the msDesc
> reflecting the medieval manuscript but with that one msItem with the
> proper origdate (of say, last tuesday).  The msItem would, one assumes
> provide additional discussion of the false nature of that item."
>
> This relies on a date inside a<p>  inside the<msItem>, that forms
> part of a textual gloss. It's not very structured, and in any case,
> our dates are in the history/origin or history/provenance section.

Hrmmm, I might disagree with myself at this point. (We all change 
over time...) From what you say, I think having it in the history 
section (whether origin or one of the provenances) might make 
more sense.

> Relying on a date is also not enough information, as the fake might be
> not be modern, and the corpus might cover a long span of time. This
> works for display or readability, but raises questions for for
> searching or faceting.

Maybe you should be using @period if all that is known is it was 
done in this or that vague time period?

> I've been using a lot of @ana on elements, referring to a set of
> taxonomies that contain controlled vocabularies. Perhaps I could use
> that attribute, either on<msItem>  to follow along with James, or on
> <origin>.

That would be a way of classifying the analysis of that 
particular msItem or origin, so might make sense.  If you know 
that it was faked in the 13thC though, I'd prefer to see 
@notBefore="1200" @notAfter="1299" in preference.

> Does anyone else have fakes, either mss or other objects? How are you
> encoding this information?

Nope. Not that I've been involved with producing msDescs of at 
least.  Can you give us more details of the original (however 
fake) again? Certainly an interesting question!

-James

-- 
Dr James Cummings, InfoDev,
Computing Services, University of Oxford