W3C home > Mailing lists > Public > www-tag@w3.org > February 2005

Re: "Protocol" (was Re: [XMLVersioning-41] Scope

From: Stuart Williams <skw@hp.com>
Date: Wed, 23 Feb 2005 16:47:09 +0000
Message-ID: <421CB38D.6090403@hp.com>
To: Mark Baker <distobj@acm.org>
Cc: David Orchard <dorchard@bea.com>, Bjoern Hoehrmann <derhoermi@gmx.net>, www-tag@w3.org

FWIW I particularly like G.J. Holzmann's [1] description of the 5 
elements of a protocol [2]

  1. The *service* provided a protocol
  2. The *assumptions* about the environment in which a protocol is 
executed.
  3. The *vocabulary* of messages used to implement the protocol.
  4. The *encoding* (format) of each message in the vocabulary.
  5. The *procedure rules* guarding the consistency of message exchanges.

Relevance to this discussion is that formats are *part* of a protocol 
definition; sequencing rules are *part* of a protocol definition.

The other point I'd make is that when we go shopping on the Web... the 
"application protocol" is the one define by the set of actions that the 
web site (the links and buttons presenting the user with the action 
choices that they have) lets us take and (in a Restful sense) consists 
of a constraint sequence of resource manipulations... some of which have 
consequences - such as goods arriving in the mail and money being 
debited from credit accounts - or... there is a protocol going on here 
that is going on at at layer above HTTP... and maybe its protocols all 
the way up :-).

Regards

Stuart
--
[1] http://spinroot.com/gerard/
[2] "Design and Validation of Computer Protocols"; Gerard J. Holzmann; 
Chapter 2 Section 2.2 [ISBN 0-13-539925-4]

Mark Baker wrote:

>On Tue, Feb 22, 2005 at 09:30:01AM -0800, David Orchard wrote:
>  
>
>>I'm somewhat confused by what you are calling a "protocol".  I think you are calling XML 1.x, XPath, XPointer, XQuery, etc. protocols.  I tend to think of the Web in terms that the AWWW talks about, that is formats and interactions, where interactions=protocols.  The split between formats and protocols seems to be W3C + OASIS doing formats and IETF doing protocols.  Some ietf examples are tcp, ip, udp, smtp, ftp, http, sip.  
>>    
>>
>
>I think the word "protocol" would make for an interesting subject of
>an etymological study.  Today, unfortunately, the word has different
>meanings in different circles.  The IETF, for example, has long referred
>to data formats as "protocols".  I looked this up a couple of years ago,
>and found RFC 1026[1] as an early (1987) example;
>
>"This RFC suggest a proposed protocol for the Internet community, and
> requests discussion and suggestions for improvements.  Distribution
> of this memo is unlimited."
>
>Even very recent data formats[2] are still referred to that way;
>
>"This document specifies an Internet standards track protocol for the
> Internet community, and requests discussion and suggestions for
> improvements."
>
>But I agree with you that in other circles, including more recently at
>the W3C, "protocol" has largely been reserved for the sub-data-format
>layers.  Moreover, in Web services circles, the word is used to refer
>only to more lower layer agreements, specifically sub-application layer.
>
>It's an unfortunate disconnect, as IMHO, it's largely responsible for
>much of Web architecture (and the architecture of other existing
>Internet based applications) not being given its due in recent
>distributed systems efforts such as the Grid and Web services (viz a
>viz "protocol independence").
>
>*shrug*
>
> [1] http://www.ietf.org/rfc/rfc1026.txt
> [2] http://www.ietf.org/rfc/rfc4010.txt
>
>Mark.
>
>  
>
Received on Wednesday, 23 February 2005 16:51:07 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 26 April 2012 12:47:32 GMT