W3C home > Mailing lists > Public > public-html-comments@w3.org > May 2010

Re: Can accessing the device microphone and camera be added to HTML5?

From: Ian Hickson <ian@hixie.ch>
Date: Mon, 3 May 2010 22:43:01 +0000 (UTC)
To: bob quinn <rcq@sockets.com>
Cc: Eduard Pascual <herenvardo@gmail.com>, Fenton Travers <fenton_travers@yahoo.com>, Jock Murphy <jockm@stufflabs.com>, public-html-comments@w3.org
Message-ID: <Pine.LNX.4.64.1005032238280.8532@ps20323.dreamhostps.com>
On Mon, 3 May 2010, bob quinn wrote:
> At 09:35 PM 5/2/2010, Ian Hickson wrote:
> >Browsers don't generally claim to support a particular version, unless 
> >they're lying. For example, nobody has ever implemented all of HTML4 
> >correctly. I don't think that's been a problem.
> 
> Inconsistent implementations is the result of a lack of an identifiable 
> standard, and yes, it most certainly is a problem.

I strongly disagree that the inconsistent implementations are the 
result of the lack of versioning in specifications. HTML4 was very clearly 
versioned, and it wasn't consistently implemented. I think what causes 
inconsistent implementations is vague requirements in specs, which HTML4 
had many of, and which we've dramatically reduced in today's HTML spec.


> Speaking as both a web developer and implementor, trying to hit an 
> ill-defined target is just a headache.

You've never been able to target a specific version of HTML, so I don't 
think that not having specific versions changes this at all.


> >I'm more interested in having the spec be realistic and useful for 
> >implementors and authors than optimising it for making it easier to 
> >make claims of conformance.
> 
> If there is a difference between these two goals, then there is a 
> problem.

If the goal is to make it easier to make claims of conformance, the 
easiest way is to remove all conformance requirements. Clearly that isn't 
the same as making a spec useful.


> Implementors MUST work together to assure consistent behaviors and 
> identify the ambiguous and less-than-useful areas of the specification.  
> Only after hands-on, head-to-head, "bake-offs" to verify "realistic and 
> useful" status has been obtained should the spec be deemed "standard."

I agree. All I'm saying is that this should happen on a section-by-section 
basis rather than on the basis of the whole spec or specs.

-- 
Ian Hickson               U+1047E                )\._.,--....,'``.    fL
http://ln.hixie.ch/       U+263A                /,   _.. \   _\  ;`._ ,.
Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'
Received on Monday, 3 May 2010 22:43:37 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 1 June 2011 00:14:02 GMT