W3C home > Mailing lists > Public > www-ws-desc@w3.org > March 2005

Re: Why do we have a component model?

From: Sanjiva Weerawarana <sanjiva@watson.ibm.com>
Date: Thu, 10 Mar 2005 09:18:22 +0600
Message-ID: <02f701c52520$926ac860$26a53109@LANKABOOK>
To: <paul.downey@bt.com>, <ryman@ca.ibm.com>
Cc: <Anish.Karmarkar@oracle.com>, <bparsia@isr.umd.edu>, <jean-jacques.moreau@crf.canon.fr>, <umit.yalcinalp@sap.com>, <www-ws-desc@w3.org>

<paul.downey@bt.com> writes:
> +1  the Z notation isn't so much another layer of abstraction, but rather
> a validation tool for the spec itself.

Sure. However, since when do specs include validation tools in
the specs themselves? 

> When Z raises an issue, it's either the consistency of the 
> abstraction or unclear 
> text that is at fault. i've far more confidence in Z than the
> models we're all carrying 
> in our heads, certainly the one in mine is very broken.

Well, you've apparently decided that Arthur's interpretation when
he wrote the Z is correct whereas what the editors intended (and 
maybe even the WG at least understood as the intent) was incorrect. 
On what basis are you assuming that Arthur's interpretation is right?

Sure, the spec needs to be tighter. Show me a spec that doesn't
need further tightening and I'll owe you a pricey dinner.

Don't misunderstand me- I do understand and appreciate the value
of a formal notation and its implied opportunity for specificity.
First of all, this has nothing to do with the subject of this
conversation, and second, I do not agree that having this notation
helps further the understanding of the specification. The simple
reality is that very few (computer) people in the world can handle
this level of mathematical abstraction. 

> now it might have been nicer to use OWL, given we're the W3C, and avoid

Yeah sure, yet another widely understood and accepted notation. Maybe
you have math and semweb background, but I certainly don't (actually
I do have a math background but its been a few years) and do not assume
that our customers (implementors) do either. In fact I work with some
of the folks implementing Axis2 for Apache and they barf at this 
formalism.

I'm shocked to read "use Z" and "use OWL" from the guy who presented
a simplicity focused versioning plea at the tech plenary!! Are you
the same Paul Downey?

> the difficulties of the Z chars with browsers and pub tools, and maybe 
> move the Z to a separate document (problematic), but complaining
> about spec issues raised by a QA technique? come on!

If you want a QA technique then let's put it somewhere else. The Z
QA technique is a problem of its own now because that has introduced
yet another *normative* interpretation of the spec. 

What are the spec issues? Let's fix those without beating around
the bush.

Sanjiva.
Received on Thursday, 10 March 2005 03:33:40 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:58:35 GMT