IPP Mail Archive: Re: IPP> IPP, independently of HTTP, Requires Chunking ? Was

Re: IPP> IPP, independently of HTTP, Requires Chunking ? Was

Harry Lewis (harryl@us.ibm.com)
Wed, 15 Apr 1998 14:13:08 -0400

I don't know how to draw the picture "extracting-foot-from-mouth" in AS=
CII
text...

In reviewing the IPP minutes, I realize the source of the "chinking"
statement... it was my own raw notes of the Portland meeting which I se=
nt to
Carl-Uno!

This was an attempt, on my part, to capture the list of Paul Moore's co=
ncerns
with IPP as we were discussing them in the "Server-to-Device" session. =
Pardon
the typo... but I may also have inserted confusion regarding the whole =
topic by
poorly capturing the initial concern. I tried to find Paul's original l=
ist and
get his language, but I can't fine it.

In general, we have an IPP model which is DEFINITELY transport and prot=
ocol
independent. Then we have an IPP protocol document which is really made=
up of
two parts, an ENCODING and a TRANSPORT. I BELIEVE the ENCODING is ALSO =
meant to
be entirely transport independent. The published transport mapping is O=
BVIOUSLY
transport dependent, because it represents a CHOICE to map to a particu=
lar
transport. It should not, however, give the impression that the encodin=
g is
somehow bound to that transport - ONLY! I believe the goal, all along, =
was to
anticipate mappings to additional transports. We just had to pick one t=
o get
the job done and we picked the one we thought would best scope the Inte=
rnet at
this point in time.

So, again, I'm not sure what Paul's original concern was but it had som=
ething
to do with his impression that IPP is not transport neutral. My guess i=
s that
Paul was trying to state his desire to have a printing protocol which r=
uns over
networks, local ports etc.

Harry Lewis - IBM Printing Systems

ipp-owner@pwg.org on 04/15/98 10:39:59 AM
Please respond to ipp-owner@pwg.org
To: ipp@pwg.org
cc:
Subject: Re: IPP> IPP, independently of HTTP, Requires Chunking ? Was

I think whoever wrote

> Encoding is HTTP independent except for chinking.

should elaborate on what they meant.

Harry Lewis - IBM Printing Systems

=