[tei-council] Certainty within milestoneLike

Gabriel BODARD gabriel.bodard at kcl.ac.uk
Tue Jul 31 10:45:50 EDT 2012

At Ann Arbor I was asked to do a survey of the TEI Guidelines to see 
what sort of changes would be involved in adding the certLike elements 
(certainty, precision and respons) to the content model of milestoneLike 
elements, as discussed in ticket <http://purl.org/tei/fr/3416130> (tl;dr 
version: this would allow the use of @match='..' to point to these 
elements and make various kinds of certainty/precision statements about 
various aspects/attributes of these elements). I have surveyed the 44 
currently empty elements in the TEI, and have the following recommendations:

MILESTONE (cb, gb, lb, milestone, pb): these are the element that my 
feature request as written proposes to add certLike to.

TRANSCRIPTIONAL (caesura, handShift, pause, redo, shift, undo, move): 
these elements involve aspects of the transcription of the text and so 
are potentially subject to interpretation, uncertainty or other 
qualification. For consistency's sake, you might argue that these should 
be treated like the milestone elements too.

TRANSCRIPTIONAL, SPANNING (addSpan, anchor, damageSpan, delSpan, 
lacunaEnd, lacunaStart, witEnd, witStart): these are also 
transcriptional elements, but since they all mark the beginning and end 
of ranges or spans, there may be less features of such elements that are 
subject to interpretation, so the argument for allowing certLike inside 
them may be weaker. I am agnostic.

LINKING (alt, link): in principle these are not transcriptional but 
stand-off elements, so probably don't need certLike in any event. (Cf. 
however <join>, which is already only pseudo-empty.)

ODD (attRef, catRef, classRef, elementRef, equiv, macroRef, specDesc, 
specGrpRef): obviously not. These should remain empty.

FS (binary, default, fsdLink, iff, numeric, symbol, then, when): 
presumably not either, although I don't pretend to have any idea what FS 
is about.

ADMIN (graphic, ptr, variantEncoding, refState): for want of a better 
label--these elements are not really transcription but rather support 
the markup, so I suspect they don't need certLike at all.

DICTIONARIES (oRef, pRef): my instinct is to say these don't either, but 
is this module ever used to transcribe existing print dictionaries, and 
so subject to the sorts of uncertainty that other transcription elements 
are? If so, then yes, they deserve in as much as TRANSCRIPTIONAL (above) do.

*My Proposal:* to accept the ticket and add certLike to the content 
model of the milestoneLike and transcriptional elements listed above. If 
anyone feels strongly that the transcriptional/spanning elements and/or 
the dictionary elements should also be added to this group, then please 
shout now.

Objections and counter-proposals welcome, of course.



Dr Gabriel BODARD
(Research Associate in Digital Epigraphy)

Department of Digital Humanities
King's College London
26-29 Drury Lane
London WC2B 5RL

Email: gabriel.bodard at kcl.ac.uk
Tel: +44 (0)20 7848 1388
Fax: +44 (0)20 7848 2980


More information about the tei-council mailing list