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

RE: Call for implementation ideas

From: Dailey, David P. <david.dailey@sru.edu>
Date: Wed, 18 May 2011 16:46:29 -0400
To: Karl Dubost <karld@opera.com>, James Graham <jgraham@opera.com>
CC: HTMLwg WG <public-html@w3.org>
Message-ID: <C64F09DF6833C44782B27844765560BC178B44D562@MSFEXCH01.srunet.sruad.edu>
I agree strongly with Karl on at least one point here (jumping in without having the history of the discussion in mind): so often the cost of having a million developers waste 10 hours apiece is not weighed against the cost of 100 hours of time to implementers, in determining how specs will evolve. The first item on my wish list for HTML6 is to revamp the nature and structure of the conversation starting very much from the ground up. 

regards
David
________________________________________
From: public-html-request@w3.org [public-html-request@w3.org] On Behalf Of Karl Dubost [karld@opera.com]
Sent: Wednesday, May 18, 2011 3:03 PM
To: James Graham
Cc: HTMLwg WG
Subject: Re: Call for implementation ideas

Le 17 mai 2011 à 04:41, James Graham a écrit :
> I believe the CSS WG considered 2. to be "two UAs that interoperably implement all features".

Product A, B and C (3 columns)
for a defined set of features in a logical functional group

feat1  X _ X
feat2  X X X
feat3  X X _
feat4  _ X X

There is no interoperability between the different products even with a double implementation of each features. This might in some circumstances completely make the full logical group not usable.

Having the full set interoperable add indeed difficulties, but is interesting for Web developers who are at the end of the food chain and have to deal with incompatibilities. There is a strong cost associated with the lack of interoperability.

We might be able to leverage all of that with what has already been done in terms of testing. Sites such as whenCanIUse helps to figure out what seems relatively well implemented (even if lacking information in the details). Somehow I have the feeling that HTML5 is impossible to test without a large collaborative effort from the Web community.

Maybe there is a need for NotUsableYet site identifying what's broken. The inverse photo of whenCanIUse

--
Karl Dubost - http://dev.opera.com/
Developer Relations & Tools, Opera Software
Received on Wednesday, 18 May 2011 20:51:38 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 9 May 2012 00:17:32 GMT