W3C home > Mailing lists > Public > w3c-dist-auth@w3.org > October to December 2005

Re: Status of Bugzilla Bug 10, Round-tripping various information in properties

From: Geoffrey M Clemm <geoffrey.clemm@us.ibm.com>
Date: Thu, 22 Dec 2005 13:25:30 -0500
To: " webdav" <w3c-dist-auth@w3.org>
Message-ID: <OFB673ADA2.6F0914A4-ON852570DF.00647D6B-852570DF.00654386@us.ibm.com>
There were several postings (most recently, one a couple of weeks ago)
that explained why escaping the XML as #PCDATA has a variety of problems.
For example:
So until a non-problematic solution is proposed, I stick by my
preference for a recommendation that servers preserve namespaces
(namespace preserving parsers are not that hard to come by, so I
don't see this as an unreasonable recommendation).


Wilfredo Sánchez Vega <wsanchez@apple.com> wrote on 12/22/2005 01:12:23 

>    For what it's worth, I talked to Greg Stein about this at 
> ApacheCon, and his position is that namespaces will not be preserved 
> by mod_dav, insofar as they are not semantically meaningful to a 
> namespace-aware XML parser.  The presence of other specifications 
> which add semantic meaning to then doesn't change that.
>    This is basically the position I was taking.
>    I (and I believe Greg) don't think this belongs in the spec, even 
> as a SHOULD.  An explicit recommendation that clients sent properties 
> as #PCDATA still makes a lot more sense to me.
>    I have yet to hear a reason why serializing the property as 
> #PCDATA isn't an acceptable solution for whose who wish to avoid 
> forcing the server parse (and therefore interpret) the property value.
>    -wsv
> On Dec 22, 2005, at 8:24 AM, Geoffrey M Clemm wrote:
> > (1) I support adding an explicit recommendation that servers
> > preserve namespace prefixes in property values.  I'm neutral
> > on whether we use the term "SHOULD" in that recommendation.
Received on Thursday, 22 December 2005 18:25:48 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:01:34 UTC