[tei-council] Dates and calendars

[Lou Burnard]

Exactly why I argued against introducing @calendar.

The only logical thing I can see would be to impose a schematron 
constraint which says that if @calendar is present, the other 
normalizationss are not permitted.


On 10/08/12 22:07, Martin Holmes wrote:
> Hi all,
>
> I've just been working on date/calendar issues, and I've come up against
> what looks like rather an anomaly. Our default values for @when etc. are
> this:
>
> data.temporal.w3c =
>      xsd:date
>    | xsd:gYear
>    | xsd:gMonth
>    | xsd:gDay
>    | xsd:gYearMonth
>    | xsd:gMonthDay
>    | xsd:time
>    | xsd:dateTime
>
> So I went to the W3C site to find out about calendar issues, and I
> discovered this:
>
> "The ·primitive· datatypes duration, dateTime, time, date, gYearMonth,
> gMonthDay, gDay, gMonth and gYear use lexical formats inspired by [ISO
> 8601]... [ISO 8601] "specifies the representation of dates in the
> proleptic Gregorian calendar and times and representations of periods of
> time". The proleptic Gregorian calendar includes dates prior to 1582
> (the year it came into use as an ecclesiastical calendar).
>
> [...]
>
> "gYearMonth represents a specific gregorian month in a specific
> gregorian year."
>
> [...]
>
> "gYear represents a gregorian calendar year."
>
> <http://www.w3.org/TR/xmlschema-2/#date>
>
> So what on earth are we doing when we encode something like this?
>
> <date when="1424-02" calendar="#julian"/>
>
> The XSD specification for @when insists that it's Gregorian, yet our
> @calendar says something different.
>
> This is problematic, surely?
>
> Cheers,
> Martin
>