Seriously, though, there are important aspects of resource
discovery best left to humans, considerations like 'relevance to task
at hand' and 'degree of relatedness to present document holdings' and
'reputation of source author'. Of course, leaving these things to
human judgment means that you have to be at least slightly willing to
accept the judgment of the human involved...and that you probably want
that human to have at least some automated tools.
What has in fact emerged in our field is an informal system of
registration similar to what someone proposed in an earlier message.
At the moment, there are two sites which maintain Net 'kiosks'
focussed on legal information (ours and Washington and Lee). One of
these (ours) maintains a mailing list which is solely for the purpose
of communicating the discovery of or need for new Net-based legal
information. There are also a couple of well-known lists for
legal-tech and law-library people on which such offers and needs
appear regularly. Also important to us -- but not yet in place -- is
a system for registering 'works in progress'. Much of what is most
useful to us at present involves marking up existing public-domain
documents of some size (say the Internal Revenue Code); there is a
rather limited set of documents involved and everyone in the field
would I imagine prioritize them in roughly the same way, so there is a
great chance of duplication of effort if work-in-progress is not
communicated in some way.
Might we not do the same thing for the Web? I plan to put together a
'law page' roughly paralleling what we've done with our Gopher site
soon, and put it together in such a way that it could be partitioned
off to other sites and other editors by sub-field as it grows. To be
truthful, I'm also anxious to pass off the maintenance of this
information to others with more substantive legal expertise as quickly
as possible. I'm even less capable of judging the quality of legal
information than I am of writing software, and moreover I get paid to
Thomas R. Bruce