> >
> > Change proposal:
> >
> > Devices commonly known as "reverse proxies" and "(origin) server
> > accelerators" are both more properly defined as a type of
> > ^^^^^^^^^^^^
> > surrogates acting as caching http->http gateways to replicate
> > ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
> > the content.
> > ^^^^^^^^^^^
>
> Depending on its policy and implementation, a mirror can be
> indistinguishable from a "reverse proxy" or "origin accellerator" so I
> believe mirrors should be included as a type of surrogate. If they are not,
> then the work I am most interested in will not be covered.
>
> Micah Beck
>
I second on that.
My other feeling though is, that we are too much tempted to
cover mirrors and other (?) proxies with a single abstract notion
and thus risk oversimplifying the mirrors issues. Remember the earlier
discussion on how replication is different to mirroring ? (no, don't want
to go there again). Just a comment.
However I agree that we should not leave mirrors out, and perhaps
consider them surrogates. Would it complicate anything ?
--w
This archive was generated by hypermail 2b29 : Thu Nov 18 2004 - 11:21:29 MST