Re: Annotations, sets, & servers, and redirects

> From roscheis@Xingu.Stanford.EDU Wed Oct 18 21:25 PDT 1995
>   >Yes, indeed, the design of my annotation proxy prototype had the
>   >concept of organizing the annotation set into a sorted N-ary tree
>   >of HTML documents.  The N-ary tree could reside entirely on one
>   >HTTP server or it could be structured to span a multitude of servers
>   >to distribute load.  Since the annotation set was just a bunch of
>   >HTML documents, they could easily be cached by caching proxies.
> So, if an annotation set spans more than one server, how does access
> control work then.  Consider the case of adding members to access
> control groups.  OK, let's design an authorization update protocol.
> All security issues considered ?  Really secure ?  Is it worth the
> overhead in protocol complexity ?  Are there other solutions to the
> hotspot problem to attain the same effect ?  (e.g. replicating sets
> and splitting up members over sites once they subscribe--as a
> pragmatic one, etc.)

1) The points mentioned above are quite valid.

2) I admitted in my original message that I had not actually implemented
   the N-ary tree stuff.

3) I will be perfectly happy if we come up with a solution that requires
   that all annotation sets reside on one machine with one server.