Subject: Re: Merge GILS
Eliot Christian (echristi@usgs.gov)
Date: Wed, 11 Feb 1998 13:45:48 -0500
Message-Id: <3.0.3.32.19980211134548.00b150a8@130.118.4.2> Date: Wed, 11 Feb 1998 13:45:48 -0500 To: gils@cni.org From: Eliot Christian <echristi@usgs.gov> Subject: Re: Merge GILS In-Reply-To: <34E1E1F9.319B@wln.com>
Hi Phil
On 2/11/98, Philip Coombs <pcoombs@wln.com> wrote:
>
> I appreciate the effort to merge GILS into bib-1. May I offer a
> suggestion?
>
> For those of us embedding the GILS attributes in the <HEAD> of HTML
> documents, use of technical and often cryptic attribute terms like
> "uncontrolled term" is counter-productive. First, the authors and
> Webmasters coding their documents haven't a clue what these are.
> Second, the major search engines ignore the value's increased importance
> in these unrecognized metatags. Why couldn't we call it "keywords"?
> ditto for subject-terms-controlled --> "subjects". ditto for abstract
> --> "description". As long as we have some license to modify our
> attribute set, can we jump on the opportunity to improve it?
GILS is fairly open about how you might name an element. Search
interoperability through the GILS Profile is defined during the
actual search and at that point the well-known attributes are
identified only by number.
Gathering contents that will be made searchable through GILS is
outside of the scope of the GILS Profile. In my experience, server
software provides some mapping process to associate a particular tag to
the appropriate attribute number. Your mapping could use "description"
instead of "abstract", or a French tanslation of the tag, or really
anything you want. (Your server may have a similar mapping process to
create the SUTRS format on the presentation side.)
I would add a note of caution, though. Your ability to support search
against well-known attributes depends on the quality of the cataloging.
If people populate metadata without a "usage guideline" (aka,
cataloging rules), the bad metadata detracts from the usability of the
entire set. For this reason, I'd be careful not to use very common
terminology lest people just assume they know what it's supposed to be.
(For instance, I'm afraid "keyword" is not always applied solely in the
sense of "subject term".)
Eliot Christian, US Geological Survey, 802 National Center, Reston VA 20192
echristi@usgs.gov Office 703-648-7245 FAX 703-648-7112 Home 703-476-6134
This archive was generated by hypermail 2a16 : Tue Mar 23 1999 - 03:55:41 EST