At http://www.tei-c.org/release/doc/tei-p5-doc/en/html/HD.html#HD44 I read :
"The <calendarDesc> element is used within the <profileDesc> element to document objects referenced by means of either the @calendar attribute on date or the @dateScheme attribute on any member of the att.datable class. "
Except that there is no such thing as a dateScheme attribute sfaics. att.datable.custom does supply one called datingMethod which is probably what was intended, though the name seems a bit odd.
I take this as an indication that we probably need some way of checking all <att> elements in the source. Again.
Nor indeed is there any such attribute as @calendar on <date> : it's supplied by the att.datable class. I am not clear why we need both. Certainly we need an example to explain the difference, if there is one.
@calendar and @datingMethod do completely different things, and both are required. @calendar tells you which calendar is used for the content of the date element; @datingMethod tells you what calendar is used for the @*-custom attributes. We use both, all the time, on the same elements. I'm pretty sure there's good documentation of the difference.
What is intended here by dateScheme is @datingMethod.
Assigning to Martin Holmes for triage and reporting to council
Fixed the erroneous attribute name in rev 12892. Now looking for the explanation of @calendar vs @datingMethod which I could have sworn Gabby put into the Glines a few years ago...
There seems to be no discussion of @calendar and @datingMethod in the Glines prose, so I've added one to ND, and pointed to it from the relevant section of HD. When Jinks has built the new versions of chapters, I'll post links to my additions here and copy the Council list.
New section is here:
http://teijenkins.hcmc.uvic.ca/job/TEIP5-Documentation/ws/Guidelines-web/en/html/ND.html#NDDATECUSTOM
along with a link to it from the Calendar Description section of the header chapter:
http://teijenkins.hcmc.uvic.ca/job/TEIP5-Documentation/ws/Guidelines-web/en/html/HD.html#HD44
Comments welcome.
I've added one para per suggestion from LB on the list, and I'm now closing this ticket.