At 1:17 PM -0500 7/18/97, Diane I. Hillmann wrote:
>On Fri, 18 Jul 1997, Robin Wendler wrote:
>
>>
>>However, it may be that Date and Coverage are the two places where we
>>simply cannot live with the Canberra Qualifiers Rule. This pains me
>>somewhat, because it's such an elegant rule...
>
>Actually, I don't think we have to go that far. As a recent convert and
>proponent the "date pristine," I hasten to point out that even if several
>DC.date.whatevers reside in a particular record, if the search engine
>doesn't understand the "whatever" and just treats them all as the kind of
>date it prefers, what's the harm? At least from a discovery point of view,
>that is.
We need to define a *default* date to be used. For documents with only one
associated date (most homepages, for example), this date would be used.
For documents with many dates associated with them, there must be one date
subcategory that is also put under "DC.date" by itself. For example, a
book might use the publishing date, an e-mail correspondance the date it
was sent, etc. People can thus be as specific as they want, while still
providing one basic category that all engines will recognize.
--------------------------------------------------------
[ Jordan Reiter ]
[ mailto:[log in to unmask] ]
[ "You can't just say, 'I don't want to get involved.' ]
[ The universe got you involved." --Hal Lipset, P.I. ]
--------------------------------------------------------
|