From: Jon Biddell [mailto:[EMAIL PROTECTED]]
>
> >I think you could reasonable argue that the above clause
> does not cover NAT.
> >Note that no computer behind the NAT box is actually
> 'connected' to the
> >service - rather the NAT machine is relaying requests of the
> machines not
> >connected.
>
> Ah... That argument is so thin it's anorexic...:-)
I know, but consider the following situations:
1) A single machine connected to the ADSL running a web browser.
2) A single machine connected to the ADSL. Dumb terminal connected to
machine on a serial port running lynx to browse the web.
3) A single multihomed machine connected to the ADSL. Dumb XTerm connected
via private ethernet running netscape on first machine to browse the web.
4) A single multihomed machine connected to the ADSL. Second machine
running Netscape on first machine via X to browse the web.
5) A single multihomed machine connected to the ADSL and running squid.
Second machine running Netscape but proxied via squid to browse the web.
6) A single multihomed machine connected to the ADSL and running NAT.
Second machine running Netscape but proxied via NAT to browse the web.
7) A single multihomed machine connected to the ADSL and running NAT. Many
machines running Netscape but proxied via NAT to browse the web.
Where do you draw the line? It can be easily argued that each step is no
different to the one above it as it is *always* the first machine doing the
connecting to the net.
Failing that you claim that the Linux box is there for firewalling and the
AUP clearly states that you are responsible for the security of the client.
I'd say it is grossly irresponsible to connect a Windows machine directly to
the net. ;-)
John Wiltshire
--
SLUG - Sydney Linux User Group Mailing List - http://slug.org.au/
More Info: http://slug.org.au/lists/listinfo/slug
