W3C home > Mailing lists > Public > www-tag@w3.org > October 2003

RE: Proposed restatement of syntax-based interoperability principle ( was RE: Action item on syntax-based interoperability)

From: Olivier Fehr <Olivier.Fehr@ofehr.com>
Date: Mon, 27 Oct 2003 20:37:02 +0100
Message-ID: <5E4B8DD68C8C384BA827D267700C4A072E47@elmer.ofehr.com>
To: "Bullard, Claude L (Len)" <clbullar@ingr.com>, "Bill de hOra" <dehora@eircom.net>, <www-tag@w3.org>



-----Original Message-----
From: Bullard, Claude L (Len) [mailto:clbullar@ingr.com] 
Sent: lundi, 27. octobre 2003 20:18
To: Olivier Fehr; Bill de hOra; www-tag@w3.org
Subject: RE: Proposed restatement of syntax-based interoperability
principle ( was RE: Action item on syntax-based interoperability)

And I am tempted to reply, which of the big words did you 
not understand?

A standard that provides an abstract model for the 
interoperating **system** (eg, X3D) is sufficient to enable 
two different implementors to create two different
implementations for which a third author can write 
a document in any of the authorized encodings and expect 
it to operate with either of those implementations to some 
degree of rendering or behavioral fidelity.  Interoperation, 
on the other hand, is at the level of the command verbs 
of two instances of the two implementations communicating 
to achieve mutually agreeable goals, behaving in mutually 
recogniable ways.  Data is portable.  Systems interoperate.
[Olivier says] Yes, if data can be 'understood' by different systems,
they can interoperate. If I take XML as an example, then 'syntax' just
means 'well formed', which in turn means that two systems must agree on
what is well formed.

Without a solid API, that will break down.  Syntax 
can do nothing to prevent that.  It just saves one some 
headaches of learning multiples.  Syntax is easy; the 
object model is hard.
[Olivier says] Not sure I have the same understanding of API. Example:
System A calculates something -> output A which is serialized and sent
over the wire according to some rules -> system B de-serializes that
data and uses and API to manipulate that 'object'. So the API tells you
how to manipulate date once you have gotten it. Of course, if the date
is not 'syntactically correct' i.e. well-formed, the treatment via the
API may fail.
 

Syntax is a payload.  The network is the wire.  The 
syntax is bits on the wire.   One can have a very large 
very noisy network and nothing getting done.  Interacting 
is not interoperating.
[Olivier says] Well, noise on the wire is not information, correct.
However, I wouldn't call Syntax noise.
  

HTTP is the fundamental basis of interoperation of the 
real web.  Not the Internet mind you, but what common 
parlance considers 'the web'.
[Olivier says] Agreed. The Internet is much more than just the web.

Syntax is not fundamental to interoperation.  Syntax 
eases the burdens of interoperating systems and 
reduces the costs.  It is not necessary. It is convenient. 
Lots of systems are interoperating on the web without 
a common syntax.
[Olivier says] That may be because the humans provide the 'glue'? If
your goal is to automate this thing without human intervention, then
syntax may be indispensable.
Cheers
Oliver

len

-----Original Message-----
From: Olivier Fehr [mailto:Olivier.Fehr@ofehr.com]

I am tempted to ask, and your point is?
Received on Monday, 27 October 2003 14:37:05 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 26 April 2012 12:47:22 GMT