XML Editor - Download a Free Trial >
See What's New >
Buy Now >

[Home] [By Thread] [By Date] [Recent Entries]

  • From: James Clark <jjc@j...>
  • To: Henri Sivonen <hsivonen@i...>
  • Date: Thu, 16 Dec 2010 14:50:36 +0700
On Thu, Dec 16, 2010 at 2:15 PM, Henri Sivonen <hsivonen@i...> wrote:
On Dec 15, 2010, at 23:10, James Clark wrote:

> > (a) Allowing something like <br></br>
>
> There's legacy content that looks "wrong" if that doesn't render as two line breaks.
>
> > (b) Allowing something like <a/>
>
> There's legacy content that looks "wrong" if that's treated as an empty element.
>
> And this legacy content starts with "<!DOCTYPE html>"?

We want to minimize variation between modes. More modes leads to more code and testing. Unfortunately, there's one mode-dependent parsing behavior:
http://hsivonen.iki.fi/last-html-quirk/

So the heart of the issue is what is the relative priority as between

  (a) reducing variation between modes
  (b) having standards mode behave in a reasonable way

HTML5 has given absolute priority to (a).  I think there is scope for reasonable people (especially those not associated with browser vendors) to disagree with that priority.

Your link was very interesting.  Back in April 2009, both Gecko and WebKit ignored </br> in standards mode.  Now, thanks to HTML5, they interpret it as a break.

Is it now the case that in a modern HTML5-compliant browser the only effect of putting <!DOCTYPE html> at the beginning is the last quirk you mention (<table> ending a <p>)?

James

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index]

XML Editor - Download a 15 Day Free Trial Now >
See What's New in Stylus Studio >
Buy Stylus Studio - XML Editor - Now >
Site Map | Privacy Policy | Terms of Use | Trademarks
Free Stylus Studio XML Training:
W3C Member