==> From: Erik Selberg
> the FIND group's proposed Common Indexing Protocol (CIP) is a method
> by which servers can send index info to other servers. The goal is to
> be able for an end-user to qpericy a query and be routed to the
> appropriate set of servers that may answer that query. It's supposed
> to handle replication. However, it's probably not directly applicable
> to a notification mechanism (it's a server-server protocol, NOT a
> server-client protocol). It may be used in conjunction with another
> protocol, however.
It could be used instead of notification if for every web server there is a
simple index server that "speaks" CIP. The CIP model is based on the idea of
distributed index servers. These servers can forward compressed indexes of
their own index to other (more uplevel) servers. They can do this every time
that something has changed. (at least this is what I understood at the IETF
in LA a year ago). CIP doesn't actually define a hierarchy of index servers,
but it would need one to be really effective I think.
CIP is a good idea I think but I am not sure that providers of search engine
software like this.
Peter Jurg
SURFnet
_________________________________________________
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