[whatwg] Using the HTML5 DOCTYPE as a new quirksmode switch

Tue Mar 13 12:13:22 PDT 2007

On Sat, 10 Mar 2007 14:41:15 +0100, Mihai Sucan <mihai.sucan at gmail.com>  
wrote:

> So, with each new version of IE (released every 5-10 years), we will have
> a new DOCTYPE, and a new rendering mode?

No. Internet Explorer hasn't implemented (at least that I am aware of) any  
proprietary or broken stuff in their browser for over 7 years. What they  
have done is improving on what they have and implementing what they didn't  
have (like *:hover) according to the standards. As long as:

- They continue this path,
- Have a steady and open development process (the IE blog has done  
wonders, Shelley Powers will probably do wonders too),
- Are active in the standardization processes (Chris Wilson is the chair  
of the W3C HTML WG),
- Make sure that what they implement is according to the standard,
- Make sure that if they do something wrong and implement it in a  
non-standard compliant way, they fix it (fast!)

IE.next won't need a new DOCTYPE. The only reason we have "standards mode"  
and "quirks mode" is because the development cycle for IE was extremely  
slow, they continued inventing proprietary stuff, they implemented stuff  
that didn't comply with the standards, they had a completely closed  
development group and process and they weren't very active in the  
standardization processes.

Seeing how much have changed since last time, it's at least room for hope.  
And I'm willing to give the IE team the benefit of the doubt and let them  
use <!DOCTYPE html> as a CSS2Compat switch that will let them do  
"everything" right.

> Instead of using this DOCTYPE switch, I was even thinking of conditional  
> comments, DOM document property, etc. Yet, other methods only add  
> complications. If Microsoft considers adding a new rendering mode as a  
> must, such that it will not break many sites, then this DOCTYPE is an  
> elegant solution. History will repeat itself, no matter how elegant the  
> solution might be.

Time will tell. I sincerely hope they can and will get the record straight  
this time around.

> Probably I don't really like this proposal very much only because it's a  
> solution for *this* very moment, forgetting the fact that rendering  
> bugs, and sites that rely on the bugs, will exist forever (a constant  
> problem).

With a faster development cycle, the IE team can implement bug fixes  
before developers get accustomed the bugs and start catering for them.

-- 
Asbjørn Ulsberg     -=|=-    http://virtuelvis.com/quark/
«He's a loathsome offensive brute, yet I can't look away»