Bruce points out:
> Making effective use of today's Web requires knowing how information is
> stored and indexed. You need to know the difference between full-text and
> manually-indexed search engines, and how and when to use each. You need
> to know how to dissect URLs, or you spend far too much time chasing down
> false hits.

Sounds like it's already up to the level of functionality of the
technology it's replacing: the Public Library.

Of course a Library is usually staffed with helpful people who can aid
you in your search for information, or even perform the search for you.

The central debate in the world of Library "Science" these days is whether
or not there is a place for Librarians in the world of the Internet.

Today's search engines (AltaVista etc.) are pretty darn impressive (in
fact I don't know if anyone has noticed but AltaVista is now capable of
translating web pages from one language to another for you).  I can't
see how you're going to revolutionarily improve them.  Certainly there
is a place for "Contents" services like Yahoo in addition to the "Index"
services like AltaVista.  For a question like "Computer, tell me about
foo", you would probably go to a "Contents" service and get either a
hand picked set of web sites or a hand constructed summary of the topic.
There is a small danger in summaries though, which I think will keep the
raw index services the most valuable ones:  It's not too hard to see the
question/command "Computer, tell me about foo" incrementally interpreted
as:

"Computer, tell me about foo"

"Computer, tell me everything I need to know about foo"

"Computer, tell me everything you think I need to know about foo"

"Computer, tell me everything your creators/advertisers/government thinks I
 need to know about foo"

"Computer, tell me what to think about foo"

"Computer, determine what I should think about foo, think it for me, and
 tell me what to do as a result."

Personally I don't see this increasing level of functionality as an
improvement :-)

G.