The most important thing, I think, is to follow a standard.  Software is
generally built around standards and practice and standards allow us to
share information more easily.  As long as you are following a standard
consistently, it is very easy to extract that data and use it in a
different way.

For instance, it would be very easy to do an XML to XML transformation
using XSLT substring function to convert the ISO 8601 encoded elements
into the separate attributes as below.

Mark Carlson
Computer Support Analyst,
Manuscripts, Special Collections, University Archives
University of Washington Libraries
Seattle, Washington, USA

On Wed, 18 Jun 2003, Bill Landis wrote:

> I guess I don't see what the big difference is between Matt's examples:
> > <unitdate normal="19010904-19670327">4 Sept. 1901 - 27 March 1967</unitdate>
> [sidebar: though in the above normalization, I hope Matt meant to use
> the ISO range divder "/" instead of a dash?] and
> > <unitdate firstnormal="19010904" lastnormal="19670327">4 Sept. 1901 - 27
> > March 1967</unitdate>
> I'm an archivist and not a programmer, so I don't know if it is easier
> from a programming perspective to work with the range specified by two
> separate attributes (or tags, as in Matt's other example), and a
> normalized date range encoded to the specifications of the W3C DTF
> profile of ISO 8601. They both represent a range of dates with first and
> last somehow specified. Using 2 attributes to specify the range seems a
> bit more verbose, and certainly ups the encoding overhead a bit. I'd
> love to better understand, from a programming perspective, why the
> 2-attribute (or 2-tag) option is preferable to some.
> Matt also said:
> "The inability of EAD to encode a formal distinction between first and last
> dates (for date range searching) has always been a stumbling-block for me
> whilst trying to develop various EAD-based applications over the years. Yes,
> there is the ability to encode a machine-readable, normalised version of
> dates - but no actual *distinction* between first and last dates is catered
> for."
> EAD does allow for the formal distinction, in the same way ISO 8601 allows for the formal distinction between the start and end of a date range. I don't really see how labeling the distinction using 2 attributes or tags in any way improves the designation of the start and end of a range. If it were done consistently and predictably throughout in one way or the other, why wouldn't that be good enough?
> Bill
> --
> | Bill Landis
> | Manuscripts Librarian, Special Collections and Archives
> | The UCI Libraries, University of California
> | P.O. Box 19557, Irvine, CA 92623-9557
> | 949 824.3113 Voice | 949 824.2472 Fax
> | [log in to unmask]