httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Dean Gaudet <>
Subject Re: Funny stuff
Date Mon, 06 Mar 2000 20:28:16 GMT
one solution is to build a distributed search engine which takes into
account your personal tastes and can't easily be tainted by corporate

for example, your search profile is based on the email you read, and the
web pages you visit (through a proxy which collects the stats).  your
profile is augmented by those of people you trust -- the trust model is
part of your profile.

the "distributed" part is something akin to how email/dns work.  for
example, email/dns would never work if there were a single site for either
of them.  they're distributed systems in which no piece has knowledge of
the entire system.

so suppose my trust network includes all of the folks who've ever posted
to new-httpd.  then a search on my trust network could find pages which
all of you have visited -- which, for example, would include a lot of
stuff on java, which is a topic i haven't done much searching/reading on.

when i find that my results are being tainted i make a negative adjustment
to my trust model.

pretty high view of a concept, but i think it's achievable given today's
technology -- it'd just take a few years development.


On Mon, 6 Mar 2000, David Reid wrote:

> Doesn't anyone else feel that this is getting just a bit out of hand???
> ----- Original Message ----- 
> From: <>
> To: <>; <>
> Sent: Monday, March 06, 2000 6:40 PM
> Subject: Re: Funny stuff
> > 
> > In a message dated 00-03-06 12:55:19 EST, you write:
> > 
> > > Does Apple or MS prevent
> > > people from using apple.gid or microsoft.gif or windows.gif
> > > as NAMES of GIFs?? :) :)
> > 
> > Not currently... but they could if they wanted to.
> > 
> > Looks like the law is coming down on the side of those
> > who consider any document or file 'sitting on the web' to
> > be covered under the 'published materials' laws. If the
> > name is accessible to the public ( such as the name of
> > an entity on a Web page ) then is considered 'published'.
> > 
> > 'Deep drilldowns' and file/page names that are NOT mentioned
> > in any 'public' html document still seem to remain, in the
> > courts eyes, 'unpublished' material.
> > 
> > This whole thing about 'search engines' finding 'other peoples
> > pages' is going to get nastier as the days go by. The debate 
> > is being fueled by recent findings that when Joe/Jane Q. Surfer is
> > looking for something he/she won't even ask for the second page
> > of a search result. If it isn't on the first page he/she gives up.
> > 
> > I have already heard that some search engines are going to start
> > charging a fee to 'prioritize' the hits which means if you have
> > the bucks you can pay them to 'bubble' all your page hits
> > to the top. Top of the list goes to the highest bidder.
> > 
> > Kevin Kiley
> > CTO, Remote Communications, Inc.
> >
> > 

View raw message