RE: XPath 1.0 or 2.0

I'm not a fan of predicating the XML processing model spec
on XPath 2.0.  Allowing use of XPath 2.0 is fine, but requiring
support of it for XML processing, nope, bad idea.

paul

> -----Original Message-----
> From: public-xml-processing-model-wg-request@w3.org 
> [mailto:public-xml-processing-model-wg-request@w3.org] On 
> Behalf Of Norman Walsh
> Sent: Friday, 2006 May 12 11:58
> To: public-xml-processing-model-wg@w3.org
> Subject: Re: XPath 1.0 or 2.0
> 
> / Jeni Tennison <jeni@jenitennison.com> was heard to say:
> | I'm afraid that I've grown so used to having XPath 2.0 that 
> I would be
> | loath to be without it. It's not the schema-awareness or the type
> | checking, it's the support for conditionals, regular 
> expressions etc.
> 
> By my reconning, the implementation bar for XPath 1.0 is about ankle
> high and the bar for XPath 2.0 is about waist high, so I'm a little
> concerned about forcing all implementors to support XPath 2.0.
> 
> Of the folks that support the idea of using XPath 2.0 at the language
> level, which of the following do you prefer:
> 
> 1. The language (conditionals and other standard components that
>    expose an XPath expression) uses XPath 2.0.
> 
> 2. The language allows pipeline authors to choose XPath 1 or XPath 2.
> 
> 3. The language allows implementors to choose XPath 1 or XPath 2.
> 
> Only choice 1 guarantees interoperability.
> 
> I feel pretty strongly that pipeline documents that don't use anything
> but the standard components should be completely interoperable.
> 
> On that basis, I'd prefer a binary choice: XPath 1.0 or XPath 2.0
> rather than some non-interoperable middle ground.
> 
> Having said that, my intuition is that the vast majority of the use
> cases for XPath in conditionals and peepholing, etc. will amount to an
> element test, possibily with some ancestry, and occasionally with a
> few attribute tests. All entirely achievable with XPath 1.0.
> 
> So my preference, at the moment, is for XPath 1.0. But I'll vote
> "concur" if it helps us get to consensus.
> 
>                                         Be seeing you,
>                                           norm
> 
> -- 
> Norman Walsh
> XML Standards Architect
> Sun Microsystems, Inc.
>

Received on Sunday, 14 May 2006 11:27:57 UTC