Re: step must contain only a signature

> I think its a bit too late to add such a section to the existing draft spec.

Ok.  Perhaps a link to an external page that takes time to explain the
syntax?  Just an idea.

>a lot of the people involved in XProc are also close to other XML technologies
and
>sometimes assume that a link to relevant spec is all that is required

I followed the links around for XSLTMatchPattern, but I never found a clear,
concise definition of what it is.  Most of what I found was stuff like ³it¹s
a subset of XPath.²  Maybe I missed it.  In any case, I agree that more
definitions would be helpful.

James Garriss
http://garriss.blogspot.com




From: James Fuller <james.fuller.2007@gmail.com>
Date: Fri, 19 Sep 2008 07:39:04 +0200
To: James Garriss <james@garriss.org>
Cc: XProc Dev <xproc-dev@w3.org>
Subject: Re: step must contain only a signature

On Thu, Sep 18, 2008 at 9:49 PM, James Garriss <james@garriss.org> wrote:
>> Is there something we could do that would have made things clearer?
>

Hello James,

> Maybe.  What if you translated the declaration syntax for us newbies by
> using examples to show the different ways it can be expressed, then put it
> in an appendix?  Maybe something like this:

A spec should strive to be used as a instructional document for end
users, but it does has its limitations ... There are some precedents
in place for documents that show problems and their solution (such as
the XQuery use case document) ... but I think its a bit too late to
add such a section to the existing draft spec.

> or like this (if you want to be less verbose):
>
> <p:delete match="BookStore/Book/Title">
>   <p:input port="source"/>
>   <p:output port="result"/>
> </p:delete>
>
> or even like this (if you want to be really succinct):
>
> <p:delete match="BookStore/Book/Title"/>
>
> And I wouldn't mind seeing good definitions/explanations/examples for things
> like XSLTMatchPattern, IRI, QName, and XPathExpression in the glossary.

I think this is what the C Glossary section is for; a lot of the
people involved in XProc are also close to other XML technologies and
sometimes assume that a link to relevant spec is all that is required.
I think a few more terms could be added to the glossary.

hth, J