Date: Sat, 9 Jul 1994 09:07:24 -0400
Message-Id: <199407091307.JAA92477@inca.gate.net>
To: sallyh@Ludwig.intel.com (Sally Hambridge), ccoprmm@oit.gatech.edu,
From: hoymand@gate.net (Dirk Herr-Hoyman)
Subject: Re: URC Author Section
At 8:47 AM 7/8/94 -0700, Sally Hambridge wrote:
>Micheal - In general, the URC draft is very nice. I do have one
>quarrel, and I think it's larger than nit-sized.
>
>I suppose it's the librarian in me coming out, but I really have a
>problem with this construction of the author's name. Suppose I have a
>client, and I request that it fetch me URC's with the subject of
>"Shakespeare" and then sort those by the author's last name. Given
>this construction, how does the software figure out what the last name
>is?
>
Sally (and others),
I share your concerns. And I truly appreciate your input as a librarian,
this is crucial to the ultimate sucess of the URI effort.
However, the URC spec cannot address these concerns directly. Basically,
we are describing what the container called URC looks like, not the
contents. There have been efforts to look at exactly how the meta-data
might be structured (such as the oft-mentioned non-existant Data Elements
Working Group :-). There is an RFC out on FTP templates (IANA), which does
define some aspects.
In a general sense, the URC spec does not define any fields per se, it just
shows one how to define them. Exactly what fields are used and what goes
in them, is something that will be defined elsewhere. We are primarily
concerned here with defining a mechanism for providing a bit of meta-data
along with URNs/URLs.
Naturally, there does need to be a few well known fields. That much is
clear. And I have previously suggested that we might utilize the template
mechanism of whois++ to create a well known template for URC. But, we also
must provide for other uses and other fields that we cannot predict.
Again, we define a mechanism for creating the meta-data, it's going to up
to the information providers and client s/w developers and end users (i.e.,
the marketplace) to determine exactly how this mechanism gets used.
So, talk to use about how the fields get used. The discussion of repeating
elements, such a Authors, is relevant. But, we cannot afford to get bogged
down in a discussion of how to format author names, or how to format the
meta-data in general. This is important, but not our work here.
-- Dirk Herr-Hoyman <hoymand@gate.net> | Practice CyberBeach Publishing | random acts of kindness * Internet publishing services | and senseless beauty Lake Worth, Florida, USA | Home Page: <URL:http://www.gate.net/~hoymand/cyberbeach.html> Phone: +1.407.540.8309