RE: Deciding in public (Was: SVGWG SVG-in-HTML proposal)

Jeff Schiller wrote:
>a) I think most of the members of the WG believe the ability to mix
>existing languages on web pages (SVG, MathML) inside HTML is a good
>idea.  Though the form of how this will work for SVG is still being
>discussed

I would personally agree with that statement as-is, but I think some others would disagree, unless you rephrased as "I think most of the members of the WG believe the ability to mix the semantics of a couple of existing languages on web pages (specifically, SVG and MathML) inside HTML is a good idea.

>b) Some of the members of the WG want the potential to integrate
>future languages into web pages without having to add those elements
>to the HTML language (decentralized extensibility)

Yes.

>c) There is a way to both of the above right now with existing
>technologies (using XML serializations)

...which has a whole bunch of other implications, including the legacy problem.

>d) It appears that the IE Team wants HTML extensibility [1] but does
>not believe XHTML is the right approach [2] and has other ideas behind
>HTML extensibility [3]

[1] yes, I think it would be a good idea.
[2] I would not say "XHTML is the wrong approach," per se - I am recognizing the realities of both deploying XHTML, given that it is not supported in current versions of IE and the adoption rates put that several years out at best, and the reality of the obvious lack of belief from the other browser and content representatives that XHTML is a good way to deliver content.  I would say, in short, "XHTML might be a right way to deliver this some of the time; particularly in the future.  It's not really deployable in most situations today.
[3] I considered this a tweak to our current extensibility system in HTML - more a way to hack around missing functionality in IE8 by using components than an end unto itself.  I do not consider it to be a complete extensibility system, nor was it a major investment of our time, given that we already supported "XML" in HTML.

>Can we please understand the arguments from the IE team around why
>they believe XHTML is not a good approach?  At one time, Chris, you
>seemed in favour of the approach [4], yet it seems that we are no
>closer to having an IE implementation of it years later.

Re: no closer - I will stand by the choices we've made in prioritizing work for IE to do thus far.  If the IE team had infinite capacity to develop code in a single product cycle, it would have been done already.  Along with a bunch of other stuff.  :)

You are correct, we cannot definitively say why XHTML has not been successful on the Web.  However, I do believe that part of that lack of success is due to the less-forgiving XML syntax, and part of it is due to the degradation story (or lack thereof) in browsers and versions that don't support it.  (I don't want to turn this into a pro/con XML debate either.)  Part of its success in the future will be due to the important and focus it is lent by all of the major browsers.  Perhaps I am misreading the tea leaves; I don't see much interest in XHTML's future from the other browsers.  I do think XHTML would have a lot of positives as a basis; however, it does have a few negatives, and it would need to be a universal push if it were to be successful.

>I'm just trying to
>understand the reasons behind Microsoft's stance or plans.  Something
>that is notoriously difficult to obtain it seems ;)

Just ask.

-Chris

Received on Friday, 1 August 2008 21:39:23 UTC