W3C WG Goals (Was Re: validating against the standard W3C)

[AndrewWatt2000@aol.com]

In a message dated 11/12/01 22:06:17 GMT Standard Time, michael.h.kay@ntlworld.com writes:


>  This is the clearest spec of XPath I've seen, and much easier to read
> > than the verbose XPath spec. Then again, maybe XPath lends itself to
> > this kind of specification. (denotational semantics)
>
> It's easy to read once you understand the language it's written in! Which is
> the point I was making: formal specification limits the number of people you
> can communicate with, which in the end can become counter-productive if it
> means that users and implementors and teachers and authors start to turn
> elsewhere for their explanations. I've actually worked on projects where the
> coders were making things up as they went along because they couldn't
> understand the specification.


Michael,

I agree. Precision/formal description is often inimical to good communication.

Inevitably some or many of those who are placed in the role of interpreting a formalism will not be fully conversant with it and, if those are tools implementors, then we find different behaviour of tools such as in validators for XSD Schema (W3C XML Schema).

I would be interested to know, assuming it isn't giving away state secrets, if W3C Working Groups have a quasi-formal goal for the Recommendations they produce. At least some recent specifications seem to aim for a goal of "It is our goal to produce as terse and formal a specification as is possible" whereas in my view a goal of "It is our goal for the XXX Recommendation to precisely define the technology while communicating effectively to implementors and developers." would be significantly more desirable and relevant.

Surely it is impossible to "lead the Web to its full potential" if communication is attempted only with a tiny minority of those with an interest in leading the Web to that potential?

Andrew Watt