Thus spoke Terry Allen (at least at 12:19 PM 9/5/96 PDT)
>As it's a slow day,
Not so slow here, but let me make 2 quick responses to Terry's
message.
First, however much Terry agrees with what he wants me to say,...
I still think that all we should do w.r.t. the IETF and DC is
register a media type. We have a perfectly good SGML encoding for
DC info, lets register it and be done with it. There is zero chance
that the IESG is going to approve a WG whose charter is to "standardize"
the DC. At most, we could write an informational RFC and see if the
IESG will allow it to be published without inserting text at the front
saying that we are full of rubbish.
Second. Terry had
earlier told me of his concerns about the WF not providing the
info about the packages that is needed for a client to determine
which ones are appropriate for a particular purpose. This is one of the
things I am working on. I would like to use MIME's multipart/related (M/R)
content type to implement the WF containers. M/R allows one body part
to be denoted as the "start" of the multipart message, by default
it is the first bodypart in the message. I have discussed having a
"catalog" for the starting body part with another person on this
list, who may or may not want to disclaim all knowledge of such an
effort :-) This catalog would say things like
package A is_bibliographic_info for package B
package C is_terms_and_conditions for package B
package D is_critical_review of package B
package E is_revision_history of package D
in addition to describing the relations between packages, we can
also indicate what sort of information is contained in packages,
such as
package A has_urn_resolution_information
The "catalog" can also contain its own revision history info so that
we can avoid infinite regress.
The hard part about this proposal is coming up with the list of relations
(is_x) and properties (has_y) to be used. The first thing we can do to
simplify that is allow multiple "controlled vocabularies" of those terms.
This will allow us to refine the schemes over time. However, there is still
the hard work of coming up with those lists of relations. Do people have
good lists of relations we can pirate, or do we really need to create
our own?
Later,
Ron Daniel Jr. email: [log in to unmask]
Advanced Computing Lab voice: +1 505 665 0597
MS B287 fax: +1 505 665 4939
Los Alamos National Laboratory http://www.acl.lanl.gov/~rdaniel/
Los Alamos, NM, USA 87545 obscure_term: "hyponym"
|