Re: [xml-dev] XHTML 2 Working Group won't be renewed?

[Kurt Cagle <kurt.cagle@gmail.com>]

Dave,

I think there's no question about control, sadly. I can see some disadvantages with predefining namespaces by default, chief among them being things like determining whether <html:input> vs. <xforms:input> was in force in a given document for the <input> element. My suspicion is that the answer from the WHATWG side was that XForms is not considered to be a a part of HTML 5, therefore, no ambiguity arises.

I'd say that what we have here is a schism. The HTML side doesn't like namespaces in part because they add to the complexity of JavaScript code (and JavaScript code doesn't have an intrinsic notion of namespaces, something which would have been rectified with ES4 before ES4 was wiped out as part of Harmony). The presumption throughout on the HTML side is that as a language it is near perfect, and only needs a few minor tweaks before it is completely self-sufficient. For those of us who DO actually work with XHTML content with mixed namespaces, this is a horrendous assumption of course, but this also reduces HTML to yet another XML dialect, so yes, the control issue is paramount.

The question, though, is what to do about it. There are times that I feel that the W3C has become a harried parent facing a couple of spoiled petulant teenagers, and rather than laying down some discipline the parents give in to their every whim ... and then are surprised when they wreck the family car. That namespaces need to be fixed is a known problem, and should have been resolved several years ago, but to give in to a perfectly reasonable suggest to expect well-formedness simply because a few programmers are afraid of a little extra typing seems to me to be just lazy.

Kurt Cagle
Managing Editor
http://xmlToday.org

On Wed, Jul 8, 2009 at 2:43 PM, David Orchard <orchard@pacificspirit.com> wrote:

One of TBL's suggestions that I thought was excellent was that the
text/html mime would define a bunch of namespaces such as xmlns:svg,
:mathml, :aria, saving the author the ns decl steps.  That didn't even
get close to traction when we (myself and Henry representing the TAG)
pitched it to henri and anne because the author could still define
their own.   Remember, aria uses aria- to prefix their names so they
have the same # of characters and are forced to use a "-" separator
rather than a ":".   That's what convinced me it wasn't about
authoring, it was about control.

Also, to a certain extent, it's about whether you think the space of
names is growing substantially or not.  Roughly the "looking
backwards" vs "looking forwards" camps.  I'm in the camp that the
space of names can and would grow substantially if html5 supported it,
and the other camp points to how few namespaces have really emerged as
needing support in the browser.  Of course I then respond that the
only reason they could get support in the browser was because html did
allow new names and html5 defines an unknown name as an error!  And so
it goes.

Cheers,
Dave

On Wed, Jul 8, 2009 at 1:54 PM, Michael Kay<mike@saxonica.com> wrote:

>  > There's supposed to be an extensibility workshop in September at one of
> the F2Fs where namespaces in general will be hashed out - I plan to be
> monitoring that one carefully, as I suspect that there will be a move to
> "fix" namespaces in a way that will have long term negative repercussions
> for the XML community.
>
> Let's approach this positively. XML namespaces are a pretty awful piece of
> design. Perhaps this is an opportunity to revisit the requirement and do
> something a bit more elegant.
>
>
> Regards,
>
> Michael Kay
> http://www.saxonica.com/
> http://twitter.com/michaelhkay
>
>
>
>