RE: IE8 incompatibility issues (was: Re: Issue: IE 8 adds new DOM Properties for ARIA -- not compatible with other impls)

Chris, 

Innovation is fine - but innovating behind the scenes in an open 
environment where authors want their web content to run on multiple 
operating systems and browsers results in incompatibility. The end result 
is a lot of bloated coat with a lot of if-the-elses or suffer from 
incompatibility. Speaking from a company that wants to deliver IT as I 
described, I can also that this drives up the cost of development though 
you think you are being innovative. After all, your intent with IE8 was to 
deliver a standards mode which should allow for much more portability 
across browsers and reduces the amount of JavaScript through support of 
more advanced CSS features.

So, I don't believe Microsoft was trying to do anything wrong but I would 
suggest that you might want to think about being more open with the other 
browser manufacturers in this space going forward. As a deliverer of IT 
and a Microsoft customer I would very much appreciate it. 

Rich


Rich Schwerdtfeger
Distinguished Engineer, SWG Accessibility Architect/Strategist
Chair, IBM Accessibility Architecture Review  Board
blog: http://www.ibm.com/developerworks/blogs/page/schwer




Chris Wilson <Chris.Wilson@microsoft.com> 
03/14/2008 01:04 PM

To
Aaron M Leventhal/Cambridge/IBM@IBMUS
cc
Anne van Kesteren <annevk@opera.com>, Charles McCathieNevile 
<chaals@opera.com>, Cullen Sauls <cullens@microsoft.com>, Dave Pawson 
<dave.pawson@gmail.com>, Jon Gunderson <jongund@uiuc.edu>, Marc Silbey 
<marcsil@windows.microsoft.com>, David Poehlman <poehlman1@comcast.net>, 
Richard Schwerdtfeger/Austin/IBM@IBMUS, Simon Pieters <simonp@opera.com>, 
"w3c-wai-pf@w3.org" <w3c-wai-pf@w3.org>, "www-archive@w3.org" 
<www-archive@w3.org>
Subject
RE: IE8 incompatibility issues (was: Re: Issue: IE 8 adds new DOM 
Properties for ARIA -- not compatible with other impls)






Actually, for HTML 4.01 it’s not really a problem, because the HTML DOM 
details all the HTML attributes.  J
 
As previously stated, I want to get to an interoperable point here. 
However, I would point out that “If it's not in a speci do not support it 
or you break everyone else.”  would prevent all innovation outside 
Recommendation status.

 
From: Aaron M Leventhal [mailto:aleventh@us.ibm.com] 
Sent: Friday, March 14, 2008 10:55 AM
To: Chris Wilson
Cc: Anne van Kesteren; Charles McCathieNevile; Cullen Sauls; Dave Pawson; 
Jon Gunderson; Marc Silbey; David Poehlman; Richard Schwerdtfeger; Simon 
Pieters; w3c-wai-pf@w3.org; www-archive@w3.org
Subject: RE: IE8 incompatibility issues (was: Re: Issue: IE 8 adds new DOM 
Properties for ARIA -- not compatible with other impls)
 

I think we could say it's useful but the problem is it's not spec'd. So 
authors that use it will get broken content everywhere but IE. 

This isn't an ARIA-specific issue. Since it's arguably useful you could 
try to get attribute mirroring into the relevant specs. If it's not in a 
speci do not support it or you break everyone else. 

- Aaron 




Chris Wilson <Chris.Wilson@microsoft.com> 
03/14/2008 01:47 PM 


To
Simon Pieters <simonp@opera.com>, Marc Silbey 
<marcsil@windows.microsoft.com>, Anne van Kesteren <annevk@opera.com>, 
Dave Pawson <dave.pawson@gmail.com>, "w3c-wai-pf@w3.org" 
<w3c-wai-pf@w3.org> 
cc
Cullen Sauls <cullens@microsoft.com>, Jon Gunderson <jongund@uiuc.edu>, 
Aaron M Leventhal/Cambridge/IBM@IBMUS, Charles McCathieNevile 
<chaals@opera.com>, David Poehlman <poehlman1@comcast.net>, 
"www-archive@w3.org" <www-archive@w3.org>, Richard 
Schwerdtfeger/Austin/IBM@IBMUS 
Subject
RE: IE8 incompatibility issues (was: Re: Issue: IE 8 adds new DOM 
Properties for ARIA -- not compatible with other impls)
 








Simon Pieters [mailto:simonp@opera.com] wrote:
>I understand that IE works this way internally, but this behavior -- that
>all attributes are reflected by DOM attributes and that any DOM 
attributes
>(or JS properties) on elements also turn into real attributes -- is not
>backed up by any DOM spec, and Opera, Safari and Firefox don't do this. 
In
>those browsers, unknown attributes are only accessible with
>getAttribute(), and saying elm.foobar = 'x' just creates a JS property
>"foobar" without adding/changing the "foobar" attribute on the element.

IIRC, this does not necessarily happen with unknown attributes - only with 
known attributes.  If it's a known attribute, it gets reflected into the 
DOM with camelCasing.  If it's an unknown/unrecognized attribute, it is 
only accessible via getAttribute().

-Chris