> here is what I have been kind of toying with mentally. people want to
> approach the net from the point of view of their *interests* often
> coinciding with their *occupation*. now, when anyone who is anyone can
> build up their list of favorite references, and people can find those
I see where you are going. But I think the scale of things is still going
to be too tough. Let's say that 100,000 writers have their own pages. If
there were some mechanism for you to say 'Does Piers Anthony have a personal
pointer page' and then you find one, then you have indeed improved the system.
But let's say that I want to find resouces for writing but I have no idea
for whom I am looking. So I go to the ratings server. I say "What is
the most popular writing page" and the server says 'Page abc has 1 million
accesses, with 100,000 this month'. All that tells me is how many times
someone was wandered there - not how good it is. So perhaps by ratings
you mean a mechanism whereby I can indicate 'Resouce abc rates a 9.7 out
of 10 in terms of usefulness to me.' This server would permit me to
change ratings (as timliness decreases or increases perhaps), query
to see what other related resources have been rated by me (so I can say
'But this one is better than that one, so I want to change their ratings),
etc.
Is that what you have in mind?
> entrance requirements for FAQs are kind of high. I want to make it as
> easy to create Pointer Pages as it is to do an anonymous FTP.
The only difficulty in creating an HTML pointer page is writing the notation -
and that could be mechanised so that one just create plain text and a filter
read the txt and generate HTML. Personally, I would like to find FEWER pointer
pages than more. Not to stifle folks, but because if I go looking for writing
resources and find 100,000 pointer pages for writing materials, what I really
would rather do is have something show me the unique resources. Popular
resources are fine - but it is highly likely that the non-novice has already
found and used them - the non-novice wants to find new, unique resources.
And having more and more pointer pages is going to duplicate the presence
of the common ones and obscure the unique resources.
> notice, however, that we don't have a very good way for people to
> maneuver through cyberspace based on their *interests*. look at all the
> premier technology, and how tedious it is to find good information:
I notice that you do not mention WAIS below. I really think that if you
have not experimented with WAIS you should - it is a text searching tool
which provides _a rating_ of how good the hit is! Sounds similar to
your concept. Of course, there is some more rating info that I am
sure is needed to take your idea to its extreme.
> 2. mailing lists-- how do I find one that I am interested in? if I even
> find the Lists of LIsts, I have to wade through a massive
> alphabetical listing where lists that appeal to me as a say,
I am not sure what you consider an alternative. If there are 100,000
pointer pages, again, you have to either read all of them or do a
hit or miss text search - where hit or miss doesn't mean that you cant
search the entire thing, but that the words you choose for searching
may not be present in the relevant entry. The only alternative to
this of which I can envision is an intelligent agent, with an expert
system knowledge of you and your interests, going out over the net to
do the searching for you. Or perhaps an information broker service.
> 4. Gopher -- a little better. But still, trying to find the `master
I assume that the 'little better' means Veronica. But I don't
find veronica any better than Archie.
> We need a *ratings server* that sorts popular resources by their
> *accesses* so that people can traverse cyberspace based on the most
Here is where we differ. Accesses do NOT mean popular. They just
mean accesses. Perhaps it is because a particular tool defaults to
a place. Perhaps one has to pass thru one document to get to the
truely popular one. Perhaps it is very bad, and no one ever returns,
but everyone thinks it sounds neat and so goes and tries it.
> interested in writing', and it would return in sorted order the most
> popular net resources on that topic, and I could browse them all via
And what about resources which talk about expressing ideas thru
the creation of novels - sounds like it would fit what you were after,
but the description might never mention the word writing. Again,
searching for things , to me, automatically points me to Z39/WAIS.
> And notice that really a good FAQ, the kind that is a `pointer to other
> information' is often nothing but a big list of all the resources in
A set of pointers to a lot of resources might be of some use - but the
things to which it points are the REAL value. And if all those are are
pointers to pointers to pointers back to pointers, then the user is frustrated
by the loops.
--Larry W. Virden INET: lvirden@cas.org Personal: 674 Falls Place, Reynoldsburg, OH 43068-1614
[
Usenet Hypertext FAQ Archive |
Search Mail Archive |
Authors |
Usenet
]
[
1993 |
1994 |
1995 |
1996 |
1997
]
© Copyright The Landfield Group, 1997
All rights reserved