Re: Decision about Host?

>Jeffrey Mogul writes:
> Balint Nagy Endre writes:
>     Jeffrey Mogul writes:
>     > How about
>     > 	(1) clients SHOULD transmit the FQDN
>     > 	(2) the HTTP 1.x protocol DOES NOT SUPPORT server hosts with
>     > 	semantically different bindings to multiple FQDNs
>     > 	if any pair of those FQDNs share a common prefix.
> 
>     Oops. Most ISPs want to have
>     www.isp.net
>     www.customer1.com
>     www.customer2.com
>     ....
>     www.customerN.com
>     on a single host, because their customers want the illusion of their
>     own web server.
> 
>     I'm against this idea categorically.
> 
> Good point about the current naming schemes.  But then what does it
> mean if a client sends:
> 
> 	Host: www
> 
> To me, this is an error, and the server can report it as such.
> What else could it possibly do?
> 
> So perhaps I would modify
> > 	(3) server administrators SHOULD NOT configure servers
> > 	in violation of rule (2).
> to be
> 	(3) servers MUST return an error [to be specified] if
> 	a client sends a semantically ambiguous prefix in a
> 	Host: header.
300 Multipe choices would be fine.
Andrew.
>