Re: [xml-dev] To namespace or not to Namespace ....

[Andrew Welch <andrew.j.welch@gmail.com>]

On 9 April 2010 10:37, Michael Kay <mike@saxonica.com> wrote:
>> XQuery really messed this up big time... such a simple thing,
>> and such a huge amount of hassle caused by it.  I know I'm
>> not the only one to be amazed when you first realise that is
>> what is happening.
>
> Yes, it's definitely a usability trap: but then, so many things with
> namespaces are.
>
> My recollection of the history, at a high level of abstraction, is that I
> was constantly pointing out namespace problems to the XQuery Working Group,
> and proposing new features to deal with the problems, and the WG
> collectively was getting more and more irritated by the complexity that each
> new feature added to the language; so it got to the point where I decided it
> would be prudent to let the WG hear the problems from users rather than from
> me.
>
> It all goes back to the fact that the namespace facility as designed back in
> 1999 was a really bad piece of design work. I wish I could find the email
> that I wrote at the time saying "it doesn't matter that it's bad: it's an
> optional feature, and it's so bad that no-one will use it." But I think that
> not using namespaces isn't really an option today: they are too ingrained in
> the ecosystem.

:)

In XSLT generating XHTML from XML in no namespace is straightforward,
doing the same in XQuery is a pain... it's not really a design fault
of namespaces, just a real oddity in XQuery...




-- 
Andrew Welch
http://andrewjwelch.com
Kernow: http://kernowforsaxon.sf.net/