W3C home > Mailing lists > Public > ietf-http-wg-old@w3.org > May to August 1998

Re: Implications of introducing new scheme and port for exis

From: Albert Lunde <Albert-Lunde@nwu.edu>
Date: Tue, 2 Jun 1998 17:32:34 +0100 (BST)
Message-Id: <v03110700b199d6aef1c0@[]>
To: http-wg-request@cuckoo.hpl.hp.com
X-Mailing-List: <http-wg@cuckoo.hpl.hp.com> archive/latest/184
>     I would agree with Josh Cohen and David Morris that introducing a new
>     URL scheme will affect proxies and is a bad idea. I also agree with
>     David Morris that a new HTTP scheme should be avoided for the same
>     reasons.
>     Using a new default port is okay. The stuff about firewalls being
>     affected is true to the extent that they also had to be opened up for
>     the default ports used for other protocols - It really isn't a big
>     hassle reconfiguring them.
>     However, I side with David Morris in questioning the need distinguish
>     between IPP and HTTP. Is there one?

I'm not an expert in this area, but I'd speculate that the suggestion of
introducing a new URL scheme and a new port number, comes from
considerations of firewalls and security. If one is implementing a policy
of "deny-everything-not-explicitly-allowed", it might be regarded as a
"feature" that IPP would not go thru existing firewalls or proxies until
they were reconfigured.

I guess the reasonableness of this condition depends on the intended market
for IPP, and the default security policy it should be expected to have.

I'd argue that a new URL scheme is a reasonable way to introduce a
different default port; but again the price is that a lot of existing web
clients will give you dumb looks if handed a new URL scheme. But this may
not matter if HTTP is just seen as a transport for IPP.

    Albert Lunde                      Albert-Lunde@nwu.edu
Received on Friday, 5 June 1998 11:50:48 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 14:40:22 UTC