How about.... email?
Why not send an email to the client's account. When the email arrives, it
starts up a viewer on the selected information.
The thing you have to ask yourself is what happens if 1) you're not logged
in, 2) you're not running a viewer, 3) your machine is turned off, .... when
this new document arrives.
Why not just extend your email viewer to handle external references that can
include URLs in addition to anon-ftp and all that stuff?
>
> Using the Shepards model -- suppose I'm researching a-b-hydrolithium
> as a cure for aging. I'd want to put a script into the servers for a
> number of of journals to inform me of any document that is added that
> refers to a-b-hydrolithium or any document that had a citation to Dr.
> Foo's groundbreaking paper on the use of lithium "A" compounds.
But what's the mechanism for telling you? I think that's what the question
is to be answered. -- Darren