It is finally official; the unofficial Web Hypertext Applications Technology Working Group may now be indexed by Google and reached by people who are interested. The Working Group is working on specifications that are backwards compatible with Internet Explorer 6.0 instead of creating specifications that are not. Another problem of the specifications that are created by various working groups of the W3C is that they are inconsistent with each other. Especially SVG breaks parts of CSS specifications, which means that it will be impossible to implement subsets of SVG in a browser that already has support for those parts of CSS. (Mozilla anyone? Comments on SVG 1.2 from a Gecko developer.)
Both Opera and Mozilla are working on these specifications and I believe that Apple (Safari) makes part of it as well, since one of the members is David Hyatt and it is stated that members are representives of various browser manufacturers. It could be that David Haytt still works for the Mozilla project, but I don't think that is the case. It could be very interesting if those three browsers, Mozilla, Opera and Safari have interoperable implementations of these specifications and there are workaround known for Internet Explorer (note that workaround should be available, since it won't become part of the specification otherwise). This would make development so much easier. There are currently three specifications aimed at extending HTML, CSS and the DOM in a backwards compatible way.
The specification that is most evolved is Web Forms 2.0. (Extensions to the select element is partly my work!) The specification defines forms in a better way and addresses quite a few items relevant to todays real world web authors, like the extensions to the input element and the repeating form controls model which is absolutely great.
The other specifications aren't evolved yet and need some more writing, but the ideas are very interesting. I hope we can see implementations of this soon.
Especially SVG breaks parts of CSS specifications, which means that it will be impossible to implement subsets of SVG in a browser that already has support for those parts of CSS. (Mozilla anyone? Comments on SVG 1.2 from a Gecko developer.)
I don't think you see the point of SVG. SVG is a styling language not ment for structure (like XHTML). Using another styling language is just stupid in my opinion. And CSS is not really necessary for backwards-compatibility.
The question is, should SVG use CSS at all? One thing is sure, if it does use CSS, it shouldn't break with it.