IPP Mail Archive: Re: IPP> regarding "ipp:" (I spoke too soon...)

Re: IPP> regarding "ipp:" (I spoke too soon...)

Keith Moore (moore@cs.utk.edu)
Thu, 02 Jul 1998 16:25:51 -0400

> >> Not so. Every IPP packet is a fully conformant HTTP packet. We are not
> >> inventing a new protocol in the scheme sense.
> >
> >That's not the way IESG sees it. IPP is chartered to develop a protocol.
>
> Yes, but the WG chose to use an approach in which IPP server applications
> could be implemented on existing HTTP web servers. If it is a new protocol,
> then we can't use existing deployed web servers, correct?

There's a long tradition in IETF of reusing and/or adapting existing
technology, so there's no problem with that per se. But there are
several problems with overloading a new protocol on top of existing
web *services*. And the idea that IPP should be able to tunnel
through firewalls "by default" - thus overriding local site policy -
does great harm to the overall Internet architecture.

> >HTTP is an application by itself. TCP/IP is not.
> >IPP is trying to layer one application on top of another.
>
> True. However, layering one application on another has been the experience
> in OSI and other layered architectures.

And one of the things we learned from OSI is that if you have too many
layers - especially layers that don't fit well together - the result
isn't very effective. Otherwise known as the 'leaning tower' effect.

Keith