RE: Summary: D-G0017

> OK by me, but I have to admit that I don't understand the distinction that you are making.
 
Perhaps an example may help.
Standardizing protocols and formats in computing is like standardizing the
gauges and threads of the inlets and outlets of oil pumps.
Standardizing mechanisms in computing is not unlike telling manufacturers
that standard oil pumps are to be driven by electric motors only, which will
apparently stymie innovative efforts in developing alternative means to
drive oil pumps, like combustion engines that burn natural gas recaptured
in oil fields on-site.
 
[snip]
> My background is physics, which has a strong tradition of 
> using simple words when possible, so bear with me ...
 
Physicists are accused of many things.  "A strong tradition of using
simple words" is not of them. :-)  I don't think so, unless one considers
cryptic terms like "entropy" or "tactical thermal application" simple. 
I'm a believer that simplicity is good, but over-simplicity is not.  
(Wasn't it that your physics brethren Einstein who once said, "whatever
can be made simple should be made simple, but not simpler?" :-)
Anyway, simplicity has little to do with discerning "protocol" from
"mechanism" in our context.  The two words simply mean two
different things.
 
Cheers, 
 
Joe Hui
Exodus, a Cable & Wireless service
=======================================
 
 
 -----Original Message-----
From: Cutler, Roger (RogerCutler) [mailto:RogerCutler@chevrontexaco.com]
Sent: Wednesday, March 27, 2002 12:01 PM
To: Joseph Hui
Cc: www-ws-arch@w3.org
Subject: RE: Summary: D-G0017



OK by me, but I have to admit that I don't understand the distinction that you are making.
 
I have found the following definitions (extracting only the relevant parts), for the two words:

Protocol - A standard procedure for regulating data transmission between computers.
 
Mechanism - An instrument or a process, physical or mental, by which something is done or comes into being.

What is limiting about having a standard "process by which reliable messaging is done"?  Or is it just that "protocol" is more specifically a comp sci word?  I must admit to having a personal preference for using, when possible, simple words in their common meaning rather than technical terms, but I certainly would not insist on it.  Does "mechanism" have a specifically comp sci meaning of which I am unaware?  
 
My background is physics, which has a strong tradition of using simple words when possible, so bear with me ...
 
-----Original Message-----
From: Joseph Hui [mailto:jhui@digisle.net] 
Sent: Wednesday, March 27, 2002 1:35 PM
To: Cutler, Roger (RogerCutler)
Cc: www-ws-arch@w3.org
Subject: RE: Summary: D-G0017


Hi Roger,
 
> DAG0017
[snip]
> Measurements:
> Is there a standard mechanism for implementing reliable messaging?
                               ^^^^^^^^^^^^^
I believe the word "mechanism" should be replaced with "protocol."
It's a bad thing to standardize mechanisms.  It stymies creativity.
(Standardizing protocols and formats is a good thing, OTOH.)
 
Cheers,
 
Joe Hui
Exodus, a Cable & Wireless service
=======================================
 -----Original Message-----
From: Cutler, Roger (RogerCutler) [mailto:RogerCutler@chevrontexaco.com]
Sent: Wednesday, March 27, 2002 10:57 AM
To: www-ws-arch@w3.org
Subject: Summary: D-G0017



DAG0017 

The reference architecture should provide guidance for the development the web services infrastructure needed for implementing common business functions.

  
Measurements: 
Is there a standard mechanism for implementing reliable messaging? 
Transaction processing?  
Routing and intermediaries? 
Is there a standard way to uniquely identify and order messages involved with a web service?