My objection is recorded in the following two W3C email archive posts:
http://lists.w3.org/Archives/Public/public-html-comments/2010Jun/0005.html
http://lists.w3.org/Archives/Public/public-html/2010Jun/0277.html
That the WhatWG no longer seemingly has disparaging text towards the W3C in a
document that is then referenced in the W3C HTML5 ignores the greater harm:
- There are differences in the two documents, and the differences are
significant, and the differences are confusing and misleading to the web
community
- The web community is becoming increasingly confused about which document is the
"official" HTML document
- The editor has shown himself to act in a manner counter to the wishes of the
W3C HTML WG, and does so within the WhatWG document, which is then referenced
in an official capacity from the W3C document
There is no need for the W3C document to reference the WhatWG document, other
than acknowledgment, as is given other documents in other organizations. There
is nothing in the charter, neither is there a technical or other need, to
reference the licenses of both organizations. Doing so leaves the web community
confused as to which license has precedence.
I am not a lawyer, but there appears to be some potentially problematic issues surrounding the legality of
the WhatWG "license". There is no legal entity known as the WhatWG, and the companies that
supposedly "own" the copyright for the WhatWG documents--Apple, Opera, and Mozilla--have not, as
far as we know from the WhatWG site, entered into any legally binding agreement that would provide the legal
basis for such any agreement or license. This may not be an issue with the W3C, since the W3C patent policy
and other agreements cover the W3C documents, but the W3C's continued deference to the WhatWG within the
HTML5 specification could imply to others that the WhatWG has some form of legal standing that it does not
have. The W3C's very credibility is potentially being compromised by a continued reference to the WhatWG
"license".
There is no need to tell people to address concerns and issues to both the W3C
and the WhatWG. Such action, again, confuses the web community, as well as
splitting these concerns across two organizations. Too often, discussions in
the WhatWG email list appear as already completed edits in the W3C document,
without any prior discussion within the HTML WG email list. Then the only
recourse is to begin the discussion yet again in the HTML WG, but this time
through bugs and issues, all of which can take months, even years, to resolve.
Encouraging people to address concerns with the W3C document in the WhatWG
email lists is counter-productive.
The W3C HTML5 specification should be wholly contained within the W3C source
control system, without a reference to the WhatWG source control system. The
only source control that matters to the W3C is the W3C's system; therefore this
should be the only source control system linked with the HTML5 specification.
This is particularly important when we again note that there are significant
and far reaching differences between the W3C HTML5 specification in the W3C
source control system, and the document being maintained in the WhatWG source
control system.
The WhatWG document calls itself some form of "future" versionless HTML, which is going
to create significant confusion about the roles of the W3C and the WhatWG as regards the future of
HTML. This is particularly problematic, because there has been no discussion that there will be a
future "versionless" HTML, and it gives an impression that the W3C is ceding the future
of HTML to the WhatWG. This appearance is given credence when there is a link to the WhatWG
document in the very beginning of the W3C HTML5 document.
Many times those items that the W3C HTML working group has disagreed with end
up as part of this supposed future, versionless edition of HTML. They have been
items that have been examined, and found wanting by the HTML WG. Therefore,
they should not appear in any possible version of HTML, period. To do
otherwise, undermines the credibility of the W3C, as well as the HTML WG. There
is little that we can do about what the WhatWG group does with its documents,
but we do not need to give the effort additional credibility by linked to the
effort.
The HTML5 specification should not refer to a WhatWG FAQ in order to explain
_anything_ about the W3C document. Again, this undermines the authority of the
W3C, and adversely impacts on the credibility of the W3C, and the W3C HTML
Working Group.
Recently some of these same issues regarding links to WhatWG being included in
another organization's documents arose in regards to the BiDirectional or
Server-Initiated HTTP Working Group at the Internet Engineering Task Force
(IETF). The co-chairs of that group made three statements that are pertinent to
my formal objection[1]:
"Ian Hickson has agreed to continue to be the editor for IETF WG WebSocket Protocol
draft following the WG rules. We requested consensus to adopt the
draft-hixie-thewebsocketprotocol-76 as WG draft. The draft-hixie-thewebsocketprotocol has
been an individual draft in that the author can put whatever they want in it. However as
it is now a WG document, the editor need to put in the draft not what he wants but what
the WG wants. The current version (00) is to be considered as a starting point and the
working group need to figured out what changes are need such that it reflect something we
can all live with."
The operative term in that paragraph, "something we can all live with", is one
that can apply to the W3C effort: the W3C HTML5 specification document needs to reflect
what all groups can live with, though not all groups may be happy with all aspects of the
document. Specifications typically consist of compromises that meet the greatest needs of
the greatest number of impacted communities.
To allow a link to a "shadow" HTML5 document that refutes these carefully, even
painfully, arrived at compromises, undermines the work of the W3C HTML WG. Not only does
this again impact on the credibility of the W3C and the HTML WG, it discourages future
cooperation and compromise, as other groups see their effort either disdained, or
disregarded. The existence of the shadow document, linked within the W3C document, makes
these groups more reluctant to compromise, triggering a cycle that slows the HTML WG
effort, and even potentially threatens the future of the W3C HTML WG effort.
There can never be convergence between the W3C and the WhatWG, as long as there exists an
"easy out" for a subset of members of the HTML WG. If there were only one official
document, both groups would have to work through their differences. There would no longer be an
implied threat of one group "picking up its marbles, and leaving". Again, the W3C can do
little about the WhatWG, but it can assert its control over the HTML5 specification.
"We are well aware that the draft has been discussed for before in the whatwg
community, however as it is now an IETF official document, we invite all the people
subscribed to the whatwg mailing list and that are interested in influencing the design
of the WebSocketProtocol to subscribe to the HyBi mailing list and to discuss all the
issues there. This is the only possibility for who care of the WebSocketProtocol to
influence it, in its way to become an IETF RFC. Directing people to send comment to the
whatwg mailing list simply is counter-productive."
It is no less counter-productive to send people to the WhatWG email lists in
regards to HTML5. That such discussion can occur is outside the control of the
W3C, true, but aiding and abetting such fragmented, and counter-productive,
discussions should not be integrated into the W3C HTML5 document.
"The only venue to discuss an IETF protocol, as already mentioned is the IETF
mailing list, and the only place where to find the official version of the protocol is
within the IETF repository. Any version outside the IETF repository has to be considered
unofficial and any change has to be considered yet to be agreed by the wg."
This is especially important: the W3C needs to assert, once and for all, that
the only official version of HTML5 is contained within the W3C. Removing the
links to the WhatWG HTML document, issues database, license, FAQ, email list,
and source control system would be an assertive, concise first step.
Thank you for your time, and your patience with my rather long email. If I need
to address this email directly to Tim Berners-Lee and Jeff Jaffe, in order to
meet the procedures in place for a formal objection, and to ensure that it is
properly addressed, please let me know.
Regards,
Shelley Powers
- Sam Ruby
[1] http://lists.w3.org/Archives/Public/public-html/2010Jun/0215.html
On 05/18/2010 02:39 PM, Sam Ruby wrote:
This is a Call for Consensus (CfC) to publish new revisions of the
following documents in accordance with the Working Group Heartbeat
requirements:
HTML 5: http://dev.w3.org/html5/spec/
HTML+RDFa: http://dev.w3.org/html5/rdfa/
HTML Microdata: http://dev.w3.org/html5/md/
HTML Canvas 2D Context: http://dev.w3.org/html5/2dcontext/
HTML: The Markup Language: http://dev.w3.org/html5/markup/
and to publish as a First Public Working Draft (FPWD) the following
documents:
HTML/XHTML Compatibility Authoring Guidelines:
http://dev.w3.org/html5/html-xhtml-author-guide/html-xhtml-authoring-guide.html
HTML5: Techniques for providing useful text alternatives:
http://dev.w3.org/html5/alt-techniques/
Additionally, and as we have done with previous republications of the
HTML5 draft, the chairs would like to ask Anne to update the HTML5 diffs
from HTML4 to reflect the latest changes.
http://dev.w3.org/html5/html4-differences/
Silence will be taken to mean there is no objection, but positive
responses are encouraged. If there are no objections by Wednesday,
May 26th, this resolution will carry.
Considerations to note:
- As a First Public Working Draft, this publication will trigger patent
policy review.
- As a Working Draft publication, the document does not need not be
complete, to meet all technical requirements, or to have consensus on
the contents.
- Objections may be made on specific documents but not on others.
Counting the differences document, there are effectively eight different
documents under consideration at this time.
- Sam Ruby
