Re: First Draft of W3C version of URL Spec

The following is about HTML in particular

Yoav wrote:

>Let me start by stating that I'm not particularly attached to either the
>WHATWG or the W3C.
>I care for the Web, and this issue, which seems to be spinning out of
>control, puts it at risk.

While I am attached to the W3C, I also care about the web and in particular
how web technologies are implemented to support accessibility and how
authoring advice and requirements of HTML are defined to support
accessibility.

My experience, since 2007, with pursuing this has resulted in interactions
with many of the people involved with HTML in both W3C and WHATWG in mailing
lists, bug reports and IRC etc. Personal experience has lead me to conclude
I can best achieve the objectives of my work within the W3C framework as it
has an organizational structure that recognizes non browser stakeholders.
This structure allows non browser vendor participants to actively influence
the defining of HTML apart from the mainstream browser implementation
specific aspects.

I realized that if I wanted to participate in getting a particular
mainstream feature implemented (such as a new element), pursuing this
involved interacting with the WHATWG mailing list directly as that is where
most browser implementers respond to technical input on such subjects. This
did not mean working solely with one organization or the other, it just
meant reaching out on the WHATWG mailing list for feedback on proposals
(implementer buy in).

I have also realized that getting the accessibility layer aspects of HTML
features implemented does not involve the WHATWG (as HTML of any flavor does
a poor job of defining it), most of this activity occurs through discussions
at the W3C and/or directly with the accessibility engineers working on the
various browsers.

I happen to think that what the HTML specification says in regards to how
authors should/must/may use HTML is an important part of the specification
of HTML. 
WHATWG by its own admission is concentrated on defining the mainstream
implementation aspects of HTML for browser implementers. Which is fine, but
it does not follow that because it's the best venue to do this, it's also
the best venue to define everything to do with HTML. I suggest that the
mainstream browser implementation centricity of WHATWG influencers makes it
a sub optimal place to define how HTML is to be authored to best serve end
users and leaves accessibility layer implementation knowingly under defined.


I suggest that in regards to HTML, a path forward is for the WHATWG to serve
it core constituency by confining its definition of HTML to mainstream
browser implementation and that the W3C *continue* work on other aspects of
HTML and extensions to HTML. This would involve both parties making
compromises and coming to an agreement on roles and responsibilities. This
is effectively what has been happening since Ian stopped being the editor of
HTML at the W3C,  the 20% of differences[1] between the W3C and WHATWG
versions of HTML are 99%+ differences in how authors should/must/may use
HTML + accessibility layer implementation details (in wai-aria section
3.2.7).

I suggest that, if needed, starting from scratch on the non implementation
aspects of HTML's definition would be a daunting task, to comply with Ian
(WHATWGs) desired, but not formalised, no copying policy (like the whatwg
had to do back in the day to comply with w3c's policy). It would be a good
thing.

As I have publicly stated before HTML specification fragmentation is an
issue because neither the whatwg nor W3C has the support of all the various
stakeholders and control/ownership over the HTML language and it's not
something that can be claimed/mandated or argued from authority, the
difference being (I think) is that W3C recognizes the legitimacy of non
browser stakeholders. Until such times that there can be some agreement on
this, fragmentation will continue.

PS: I don't know how Microsoft as a browser implementer fits into the
WHATWG/W3C equation, but that's not something for me to work out.

[1] http://lists.w3.org/Archives/Public/public-w3process/2014Sep/0010.html

Steve Faulkner
TPG Distinguished Accessibility Engineer
-
Co-editor HTML 5.1

Received on Tuesday, 2 September 2014 06:34:17 UTC