Re: Decision about Host?

> Fine, call a program residing on a firewall that accepts HTTP requests
> from the outside world, does *whatever*, including maybe even caching
> results, and forwards those packets to other servers inside a network
> a "gateway".

Yeah, that's what it is.

> I'll take everything you said at face value except:
> 
 >> Now, what happens behind the curtains (between the gateway and the
 >> origin server) is none of our business.  There is no need for that
 >> communication to even be HTTP.
> 
> While it seems *possible* to take that position, it just doesn't seem
> even slightly *useful*, because in practice what will be running inside
> the network will be other HTTP servers, not XYZ servers using some
> proprietary protocol.

Information hiding is not useful?  In practice, a gateway is just as
likely to be a front-end for WAIS, Gopher, or a CORBA domain as it is
to be an HTTP server behind a firewall.  The gateway is responsible for
the protocol on the other side, whatever that protocol may be, and it
simply isn't useful to invent situations that do not exist.

> Now maybe the multiple port thing is such a remote case in this
> instance that it shouldn't be supported, or should be required to be
> entirely handled by the "gateway". Fine. I didn't notice your message
> declaring this was a closed topic.  I had mistakenly thought that working
> through the implications was the purpose of having a discussion.

Working through the implications of a problem that doesn't exist and
is not even within the purview of this WG is a total waste of our time.
If you want to explore nonexistant design issues within the conceptual
probability of future applications of HTTP, please do so on the
<www-talk@w3.org> mailing list.  I read both, as do most WWW developers,
and they can tell you that the problem doesn't exist without delaying
the WG's progress on known problems and without provoking yet another
response from me when I'd really prefer to get some work done on the
HTTP/1.0 and 1.1 specifications instead!

All discussion on the topic of Host is closed.  You may continue discussion
of it anyway, if you wish, but there is no guarantee that I will read it.
As far as I can tell, the only working definition of "closed" is that the
editor can get away with putting a mail filter (real or imaginary) on a
topic and still be doing his/her job.

 ...Roy T. Fielding
    Department of Information & Computer Science    (fielding@ics.uci.edu)
    University of California, Irvine, CA 92717-3425    fax:+1(714)824-4056
    http://www.ics.uci.edu/~fielding/

Received on Saturday, 7 October 1995 18:40:08 UTC