Warning:
This wiki has been archived and is now read-only.

ChangeProposals/ChangeStatus

From HTML WG Wiki
Jump to: navigation, search

Change to Status Section

Written by Steve Faulkner (based on the text by Shelley Powers).

Summary

Including text and links that refer to a different specification being developed under a different process and policies, as if they were the same specification, is not accurate, useful or appropriate. So remove the innacurate text and links from the status section of the specification.

Rationale

Remove the WhatWG references in the Status section.

The WhatWG document that is linked in the Status section is not comparable to the document a person will find in the W3C. Any person who does not understand the relationship between HTML5 and HTML 'living standard' may follow a link to the WhatWG document, assuming both documents are the same. They are not.

The WHATWG specification does not contain links to submit feedback on the W3C HTML5 specification.

The WhatWG document doesn't have the same title as the W3C document. Rather than reference "HTML5", which the reviewer might pardonably expect, he or she will find a document labeled "Web Applications 1.0". The version number is given as "Living Standard", which is not a term that the W3C has originated, nor is it a term that is used by the W3C standards community. This, of course, will generate confusion, especially because when one accesses a link labeled "editor's draft" one should be able to expect to at least access a version of HTML5, not something completely different.

A counter-proposal for this issue has made the following statement:

"WHATWG's contributions to the development of HTML5 are substantial, to say the least. 
Acknowledging its existence, briefly explaining its 
relationship to the W3C and linking to several useful resources hosted by the WHATWG
is the least that the W3C could to. With the WHATWG having dropped the "5" from HTML5, the risk of confusion is lower than ever. 
More importantly,  lacking evidence of actual confusion, acting pre-emptively because there might be confusion is a waste of everyone's time."

While there is no doubt thte WHATWG has played a large part in the development of HTML5, the place to aknowledge this is in the Acknowledegements section of the HTML5 specification. The acknowledgement to the WHATWG can be as verbose and effusive as is befitting their contribution. Refer to the details section for draft text.


As to the removal of the "5" and the "living standard", rather than clarify issues, this only adds to the confusion. "Living standard" is not defined in the W3C, so there's no reference between the two documents based on this term. To all intents and purposes, the WhatWG document that is linked looks like a completely different document. In many significant ways, it is. And while it is true that the WHATWG specification is no longer called HTML5, the introduction of the specification claims that the specification is HTML5:

"1.1 Is this HTML5?

In short: Yes."


As does the 'web developer' version:

"HTML5
A technical specification for Web developers"


But neither of these documents are are strict subset or superset of the W3C HTML5 specification. The W3C HTML5 specification and the WHATWG Living Standard are different specifications being developed under different methodologies and policies, the W3C and the WHATWG specification have diverged, so it is no longer sensible to claim that they are the same or direct users to the WHATWG document as if they were the same.

Having said that, there is no reason why at some point in the future the 2 documents may converge and in that case it would again be reasonable to have the links.

The WHATWG document omits references and text related to accessibility that have been agreed for inclusion by the W3C HTML working group via working group process. Prefrerring instead to mischaracterise the decisions of the working group:

"* The W3C version omits a paragraph of implementation advice because of a working group decision from June 2010.
* The W3C version includes a paragraph of advice redundant with the ARIA specifications because of a working group decision from March 2011.
* The W3C version gives incomplete advice regarding the alt attribute and instead references other documents on the matter because of 
a working group  decision from March 2011.
* The W3C version includes a link to an incomplete document that contradict this specification because of a working group decision from Februray 2011.
* The W3C version has different prose regarding the use of tables for layout purposes because of a working group decision from March 2011. 
In contrast, this specification unambiguously disallows the use of table elements for layout purposes.
* The W3C version requires authors who are writing HTML e-mails with images to people they know can see the images (e.g. a wife sending her 
husband photos of their child) to nonetheless include textual alternatives to those images, because of a working group decision from April 2011.
* The W3C version does not state that the img element's alt attribute is its fallback content, because of a working group decision from April 2011.
* The W3C version is missing a clause that requires conformance checkers to discourage cargo-cult accessibility authoring because of a working 
group chair decision from May 2011."

The WHATWG document contains substantive content that is not in HTML5:

  • Canvas 2D Graphics Context
  • Microdata
  • Microdata vocabularies
  • Cross-document messaging (also known as Communications)
  • Channel messaging (also known as Communications)


A last concern with pointing people to the WhatWG web site, especially when it comes to recommending people use the WHATWG for problem resolution, is that feedback does not get automatically or manually get processed by the HTML WG and is not covered by working group prcedures, as required for all last call comments.

Details

  • Remove the reference to the WhatWG subversion server. There is a reference to the W3C CVS server, which is sufficient. In addition, the material at the WhatWG subversion server differs from the material at the W3C server. Such differences generate confusion.
  • Remove the commit-watchers-list for the WhatWG. We already have way for people to follow commits in the W3C space. In addition, the WhatWG work is not identical to the W3C's work, and commits in the WhatWG space may generate confusion about what is in the W3C HTML5 spec.
  • Correct the reference labeled "CVSWeb interface with side-by-side diffs". It seems to be pointing to a directory, rather than a specific document with side by side differences.
  • Remove the reference to the annotated differences document at html5.org, which is not tracking changes to the W3C HTML5 specification,it tracks chnages to the WHATWG document.
  • Remove the reference to subversion access to WhatWG documents. Not pertinent, not useful for those accessing the W3C documents.
  • Remove the paragraph mentioning work is also being done at the WhatWG. This is a spec, not a marketing brochure. People don't need to have discussion about the "convergence" between the two groups embedded in a tech spec.
  • If desired, add an acknowledgement to the WhatWG in the acknowledgement section.

A copy of the status section from revision 1.4839 has been provided with the edits marked using INS and DEL elements.

Addition to the Acknowledgements section

"The HTML working group thanks the WHATWG contributors for providing an invaluable and continuing contribution to the development of HTML5. Visit 
the WHATWG web site for information about its activities."

Impact

Positive Effects

  • The changes will result in users only having pointers that refer directly to HTML5 as developed by the HTML WG process, and ensure that feedback and reviews are specific to the W3C document.

Negative Effects

Some time to make the edits.

Conformance Classes Changes

None

Risks

None


References