> In comparing WWW vs Gopher and WAIS, I admit
> that WWW is a more powerful data representation model.
> But, its ability to discover information is only as good as
> the links are and the links are made by hand.
As marc says, you can certainly run a wais index on
the data you serve up with w3, and serve up that index
as a w3 server. Just look at the famous unix manual pages
in the mosaic menu ... that is indexed hypertext.
You don't have to use the wais indexing tool, of course, you
could use anything.
> Gopher and WAIS have global information discovery tools
> which know more than any one person does about available
> resources, e.g.,
> WAIS's directory-of-servers
> Gopher's Veronica
> Gopher's ts/tb or jughead
> and which are created by software.
> Is anyone working on similar tools for WWW?
Now we're talking indexes across servers.
I agree there is a need for an index of resources. I think a few
people are thinking of it, but as far as I know there isn't one yet,
and while that is the case, there is an opening for someone who
would like to put together a quick hack on the public WAIS code
with www as front end. Contributions please! Using www to collect
the data for indexing means that the same indexer would be able to
run over not only native w3 servers, butalso anything else (wais,
gopher, etc) which it was pointed at, and the result would be an
index which would cover all these forms. This would be really
useful, as one doesn't in general know what sort of a server has
the data one wants. So the new index will INCLUDE the directory
of servers and veronica. All on one swell foop. Anyone going to
go for it?
(I know some scandinavians hava funding proposal in for this, but if
they are beaten to the post I am _sure_ they will find ways of
spending the money!)