RE: That routing/redirection thing: mapping (?)

[Gary Tomlinson <garyt@entera.com>]

The term mapping works well.  I assume the term is then "Request Mapping"?

Gary

-----Original Message-----
From: owner-cdn@ops.ietf.org [mailto:owner-cdn@ops.ietf.org]On Behalf Of
Mark Day
Sent: Thursday, November 09, 2000 12:11 PM
To: cdn@ops.ietf.org
Subject: That routing/redirection thing: mapping (?)


Here's what I learned so far from the discussion about terms:

1. If we use the terms "routing" or "forwarding," we should ensure that they
are in line with the way those terms are used for IP, i.e. that routing is
the exchange of information to build tables; forwarding is sending things
along in accord with those tables.  However, I'm not keen to use either term
because of the potential for confusion with the IP functions.

2. Some people would like to go back to redirection, but others disagree.

3. Some other terms have been proposed: content routing, request navigation,
request forwarding, rendezvous, petitioning, guidance.  Some are better,
some are worse, none seems like a clear winner.

However, in the course of yesterday's panel at iBAND, I noticed that Avi
Freedman of Akamai was using the term "mapping" for this problem.

Here are my test phrases redone with "mapping":

a. "the system maps the request to the surrogate"
b. "a mapper decides which surrogate to use"
c. "mapping is different from distribution"

I think this works relatively well, but I want to check for any concerns
within the group before I do a wholesale replacement in the model document.

--Mark