you're referring to User-Agent info?
browsers would have to implement an authentication scheme that robots can't
spoof.
this is a great idea, but it has to be designed in a way that's fast. you
don't want this overhead to be much more than a second. even that is a
lot.
one approach is distributed authentication... basically you want a closed
set of network agents to authenticate themselves via some unique common
knowledge -- info which is unique to the closed set of agents and known to
each.
i have a book on this (Reasoning about knowledge; Fagin, Halpern, Moses,
Vardi; MIT Press 1995, ISBN 0-262-06162-7) but i haven't reached the end to
see if it includes a practical solution to this problem. the common
knowledge of a network of agents can evolve and thereby remain secure.
this is very powerful mechanism that number theory research is exploring
for a new kind of encryption as well as authentication.
-john
_________________________________________________
This messages was sent by the robots mailing list. To unsubscribe, send mail
to robots-request@webcrawler.com with the word "unsubscribe" in the body.
For more info see http://info.webcrawler.com/mak/projects/robots/robots.html