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

Re: 1.1.1 How do the WHATWG and W3C specifications differ?

From: Ian Hickson <ian@hixie.ch>
Date: Thu, 8 Sep 2011 17:40:18 +0000 (UTC)
To: "Jens O. Meiert" <jens@meiert.com>
cc: Julian Reschke <julian.reschke@gmx.de>, public-html@w3.org
Message-ID: <Pine.LNX.4.64.1109081720120.7754@ps20323.dreamhostps.com>

(Please avoid cross-posting to the WHATWG list and other lists, as it 
causes thread fragmentation when people who are only on one list reply.)

On Wed, 7 Sep 2011, Jens O. Meiert wrote:
> >
> > Please clarify -- (a) the decisions do not make sense or (b) not 
> > applying them doesn't make sense?
> 
> My main concern are the number of differences between the WHATWG and the 
> W3C version, hence the question whether we’re on it at all to improve 
> this.

Over the past few months the actual number of differences has gone down, 
not up, so the trend is good. I'm not aware of any coordinated effort to 
try to reduce the conflicts, though. If there are any differences where 
you think the W3C version is technically superior, I encourage you to 
present the technical arguments in favour of the change on the WHATWG 
list. For differences where you think the WHATWG version is technically 
superior, you should presumably bring the issue up here (e.g. by filing a 
bug, possibly after discussing it with the chairs).

As far as I'm aware, most of the changes are either editorial issues 
driven by different opinions of what makes a good spec (e.g. the W3C copy 
omits a non-normative paragraph suggesting that browsers can apply image 
analysis heuristics to help users), or normative issues where the W3C 
decision is, as far as I'm aware, simply technically wrong or inferior 
(e.g. there was a decision relating to the term "fallback content" where 
IMHO the decision is based on an incorrect understanding of the term in 
context). The differences in the W3C HTML5 spec vs the WHATWG spec are 
always minor enough that there's not been much point me making a fuss over 
them here; it just means the W3C's spec is slightly less technically 
solid, without it seriously affecting implementations or interop. (The 
chairs are aware of a case involving another spec where the difference was 
not minor and where there therefore was a fuss caused.)

Note that the W3C HTML WG's charter says that the W3C is to pursue 
convergence with the WHATWG. I'm not sure how this is supposed to manifest 
itself, or whether it has manifested itself at all.

-- 
Ian Hickson               U+1047E                )\._.,--....,'``.    fL
http://ln.hixie.ch/       U+263A                /,   _.. \   _\  ;`._ ,.
Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'
Received on Thursday, 8 September 2011 17:42:36 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 9 May 2012 00:17:38 GMT