> Andrew reports:
> >This is what the WG agreed (11-0):
> >
> >"Typically, Date will be associated with the creation or availability of
> >the resource. A date value may be a single date or a date range. Date
> >values may express temporal information at any level of granularity
> >(including time). Recommended best practice for encoding the date value
> >is to supply an unambiguous representation of the single date or date
> >range using a widely-recognized syntax (e.g., YYYY-MM-DD for a single
> >date; YYYY-MM-DD/YYYY-MM-DD for a date range; YYYY-MM-DDTHH:MM to
> >specify a single date and time down to the minute)."
I'm slightly confused by the sequence of yesterday's postings on
this topic...
I agree that just recommending a "widely-recognized syntax" --
but without recommending that an encoding scheme be specified --
is not a good idea. I prefer Andrew's proposed re-wording:
On Thu, Mar 30, 2006 at 04:41:14PM +0100, Andrew Wilson wrote:
> I'm inclined to propose some slight changes to the WG wording eg. instead
> of: "unambiguous representation of the single date or date range using a
> widely-recognized syntax...", I think this might be better as:
> "unambiguous representation of the single date or date range using a syntax
> encoding scheme, such as the W3CDTF profile of ISO 8601 (which follows the
> YYYY-MM-DD format)".
As Pete points out, this version loses some of what the Date
WG wanted to include. However, we cannot yet point to an
encoding scheme which supports the missing bits:
> It does, though taking Diane's point about naming syntax encoding
> schemes, the example includes a date range ("YYYY-MM-DD/YYYY-MM-DD"),
> but - as far as I can recall, I confess I've lost track a bit! - we
> don't yet have a syntax encoding scheme that covers this case, right?
> i.e. it's not covered by W3CDTF, the DC Date WG hasn't yet proposed a
> new date range scheme, and - again, as far as I can recall - there isn't
> an XML Schema datatype that covers this.
>
> So saying "you could use this representation" and saying also "you
> should indicate explicitly the syntax encoding scheme" kind of begs the
> question of how people can do that.
I suggest we refrain from using the comment to recommend
constructs such as "YYYY-MM-DD/YYYY-MM-DD" that are not yet
covered by an encoding scheme.
Tom
--
Dr. Thomas Baker [log in to unmask]
Director, Specifications and Documentation
Dublin Core Metadata Initiative
|