ietf-openproxy
[Top] [All Lists]

RE: AW: Using XML in OCP transport

2003-05-08 12:24:21
Alex,

good points, I agree with you.

So far it seems that the debate will go on.
I wonder how we can come to a closure?

Abbie


-----Original Message-----
From: Alex Rousskov [mailto:rousskov(_at_)measurement-factory(_dot_)com] 
Sent: Thursday, May 08, 2003 2:41 PM
To: OPES Group
Subject: Re: AW: Using XML in OCP transport




On Thu, 8 May 2003, jfcm wrote:

At 18:23 08/05/03, Alex Rousskov wrote:
I am sure there are other points. These biased lists are 
based on two 
points of views: OCP author and OCP/ICAP implementor.

I would like we clarify which is which (author, implemtator 
to try to 
best decide and document);

Both authors and implementors/coders care about the scope 
(item #1) and XML worries.

"less/more OCP work for us" is mostly for authors; however, 
Marshall and others may argue that this group is doomed to 
produce inferior specs and, thus, the less work we do 
ourselves (the more we reuse), the better for everybody 
involved, including coders.

Implementation complexity is mostly for implementors/coders 
but it also reflects authors ability to write simple protocols.

Migration motivation and complexity is mostly for those who 
advocate/market the new protocol or sell/install products 
based on the new protocol.

IETF pummeling is mostly for us, the authors. However, the 
same factor can be viewed as "reuse of existing technology".  
The latter is important to implementors.

I do not really see a product description for an implementator.

By "implementor", I meant those who will code or program OCP 
specification. What those people need is a protocol 
specification. We already have most of it for ICAP/1.1 case. 
We probably have enough for the other two cases; see the 
current OCP Core draft and the BEEP specification, and I also 
posted some examples.

By "author" I meant us, the working group.


I made a draft in HTML. I can keep doing it if I receive additions 
from others. http://jefsey.com/ocp.htm

You can make it a table to ease comparison (something I could 
not do in ASCII in the original e-mail): "OCP flavors" as 
horizontal heading, "decision factors" as vertical heading, 
and things like none/low/average/high as table cells.

Alex.