W3C home > Mailing lists > Public > public-html@w3.org > February 2011

Re: ISSUE-140 CPP — no conformance versions

From: Karl Dubost <karld@opera.com>
Date: Mon, 7 Feb 2011 13:10:58 -0500
Message-Id: <2659C2B4-6C6B-4C60-A82A-D2ADD2E52D4B@opera.com>
Cc: "HTML WG" <public-html@w3.org>
To: Anne van Kesteren <annevk@opera.com>
Summary of this mail:
    The decision/proposal seems premature as we have not 
    really defined a useful set of conformance

Le 6 févr. 2011 à 08:19, Anne van Kesteren a écrit :
> Summary: Conformance to HTML should not have versions.

The proposal is a bit too short with being able to 
"Identify who and/or what will implement the specification." [1]

* Browsers: 
  There is no common behaviors when processing 
  documents. There are at least two modes strict/quirks 
  and some have different code paths such as IE. But 
  basically a conformance version doesn't have a real 
  impact on the implementation itself.

* Validators: 
  It might be valuable to be able to conform 
  to a profile (this profile carrying a version 
  number - but let's put that aside for a bit) 

* Harvesting bots:
  A stable conformance version seems to be pointless.

* Web business contracts
  In terms of business, a version is often required by 
  web business contracts. It is a way to define a baseline 
  of common understanding of what will be used for 
  creating the Web site. It has positive and negative 
  impacts. The positive impact is to define a way to 
  assess the end of the contract by having checked the 
  code conforming to a required version. The negative 
  impact comes from the people not understanding that we 
  could achieve better quality by pushing further the 
  envelop. Example: A client will impose XHTML 1.0 Valid 
  in a RFP and then will not allow the new form features.

* Authoring tools:
  might be useful when you have to deliver a specific 
  profile, then again it really depends on what we are 
  talking about.

* Lint, repairing tools:
  What do I repair? What does that mean to repair?

The issue is that we do not know yet what it means to "conform to HTML"

"Conformance to HTML" and version seems to be the wrong association. If we look at the specification which is at http://www.w3.org/TR/html5/ It is a big pile of many technologies. Defining a conformance to this document will not achieve anything good. It will leave out plenty of technologies and put it things which will be difficult to define.

The first question before having a version for a "conformance to HTML" is to define what are the things, we want to identify as conformant. 

* DOM?
* Markup (aka HTML, the syntax or/and the semantics?)
* HTTP?
* APIs?
* etc etc


[1]: http://www.w3.org/TR/qaframe-spec/#implement-principle

-- 
Karl Dubost - http://dev.opera.com/
Developer Relations & Tools, Opera Software
Received on Monday, 7 February 2011 18:11:35 UTC

This archive was generated by hypermail 2.3.1 : Monday, 29 September 2014 09:39:22 UTC