W3C home > Mailing lists > Public > public-xml-processing-model-wg@w3.org > March 2008

Uncle! New alternate draft

From: Norman Walsh <ndw@nwalsh.com>
Date: Thu, 13 Mar 2008 16:09:17 -0400
To: public-xml-processing-model-wg@w3.org
Message-ID: <m2skyus7wy.fsf@nwalsh.com>

I have tried (and tried, and tried) to write a draft which describes
our decisions with respect to options and scoping. I simply cannot do
it. I'm not saying it can't be done, but I have failed utterly. There
are simply too many related, but not quite identical, concepts
floating around under the rubric "p:option". Every attempt to
disentangle the concepts and describe them produced sentences that
would only invite ridicule.

However, we *must* make progress, so I took a step back and decided to
try something else.

If the problem is that we've got too many concepts under a single name,
let's fix that problem by ... given them different names!

So there's a new draft at


In this draft, I have:

- Split the functionality of p:option
  - Using p:option only for declaring options on p:declare-step
  - Using p:with-option to specify values for options in atomic steps
  - Added p:variable for holding computed values

- Renamed p:parameter to p:with-param for consistency with p:with-option

- Renamed c:parameter and c:parameter-set to c:param and c:param-set,
  for consistency

I think this is an improvement. I *really* think we should remove the
value attribute from p:with-option and p:with-param. It wouldn't
remove any functionality, it would remove a needless difference from
XSLT, and it would simplify the spec.

Comments most welcome.

                                        Be seeing you,

Norman Walsh <ndw@nwalsh.com> | Puritanism--The haunting fear that
http://nwalsh.com/            | someone, somewhere may be happy.--H.L.
                              | Mencken

Received on Thursday, 13 March 2008 20:09:57 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:32:45 UTC