From CSS Discuss
OK gang, let's start ripping this to shreds... This page, that is, not IE8.
IE8 Readiness Toolkit
IE8 bug reporting
IE Beta Newsgroup - This newsgroup is the all-in-one place to discuss items about IE8 betas. Microsoft MVPs and IE team members will monitor this newsgroup.
IE8 beta, 'release candidate' builds and final release
Help | About Internet Explorer -> Version
|8.0b||5 Mar 2008|
|8.0b2||28 Aug 2008|
|8.0RC1||26 Jan 2009|
|8.0||19 Mar 2009|
- : 05 Mar 2008: "Why isn't IE8 passing the Acid 2 test?"
-  05 Mar 2008: "IE8 beta feedback"
-  05 Mar 2008: "Internet Explorer 8 Beta 1 for Developers Now Available"
-  CSS Compatibility and Internet Explorer
- Position Is Everything's list of Internet Explorer bugs  by "Big" John Gallant
- IE8 CSS bugs  by Bruno Fassino
- CSS-Class page on Internet Explorer bugs  by Alan Gresley
- Bugs in IE 8  by Gérard Talbot
- IE 8 bugs  by James Hopkins
- hasLayout.net's Internet Explorer CSS Bugs  by Zoffix Znet
- IE8 overflow and expanding box bugs  by Hilbrand Edskes
- Internet Explorer 8 report  by Mark "Tarquin" Wilton-Jones
- Internet Explorer 8 bugs  by "mvdleij"
- IE 8 betas bugs  by Philip Taylor
- IE8 bugs  by Simon Pieters
- IE 8 betas bugs  by Anton Prowse
- IE8 Standards
- IE7 Standards
Trigged using Doctype switching, HTTP headers and Meta tags.
CSS Filters and Hacks
- IE8 b1: The IE-Mac Band Pass Filter  is not interpreted as a comment in IE8 b1. Testcase: 
- IE8 b1: The Caio Hack Caio Hack triggers in standards mode in IE8 b1.
- IE8: Estelle's IE8 CSS Hack  is a combination of the star hack and targeting newer browsers with CSS3 to target IE8
- A small line-height < normal cuts the ascenders and descenders. Testcase: | 
- negative text-indent causes clipping of text in IE8b1. Testcase: 
- specifying a padding value (more than null) on html element has no effect. Testcase: 
List-item whitespace bug
- In lists there are still cases where extra space is displayed (not related to presence of white-space in the source.) Some test cases with floats inside list-items , 
The @import rule
- 3 Level Nested Import Statement Restriction IE 
Fuzzy Specificity Bug
The fuzzy specificity bug that affects IE7 and under and Opera 7.54 and under, documented here , still affects IE8 b1.
Basic Selector Support Tests
IE8 supports all CSS2.1 selectors .
|5.7||Adjacent Sibling Selectors||PASS|
|5.11.2||:link and :visited||PASS|
|5.11.3||:active, :hover, and :focus||PASS|
|5.12.3||:before and :after||PASS|
- updated testcases: pseudo classes and elements
- In IE8 b1 the background (color and image) does not extend under the border, as it should  (previously this depended on the hasLayout status of the element).
- background images get lost on multi line inline links (old problem) 
- Use of repeated background images of a particular size causes very slow page reflow. Testcase: 
- In IE8 b1 some display 'table' properties are broken. These affect at least the values 'table' and 'table-cell'. If an elements has either of these display values then other property values on the same element are duplicated like 'border-width', 'line-height', 'padding' into a phantom box. Testcases: , , 
- Overflow boxes which are floated become un-scrollable. Testcase: 
- In IE8 b1 the vertical-align property seems only partially implememented in table cells. And has problems in other cases as well. For example, these test cases with images  show that the values middle, text-top, text-bottom are not always correct. And these test cases with inline-boxes  show that the value baseline for an inline-box is not always correct.
position: relative / absolute / fixed
- In IE8 b1 an a.p. element with auto offset following a float has a wrong static position. This is an old problem. 
Do the following IE7 problems still persist?
- rtl context: bad rendering of relatively positioned inline box 
- Negative top value on position:relative is deducted from border-width of relative positioned ancestor 
- In IE8 b1 relatively positioned elements with z-index:auto correctly do not establish a new stacking context. This was wrong in IE7 . The same thing (no stacking context) seems now to happen with z-index:0 , which is not correct, but it's a minor problem.
float and clearing
- In IE8 b1 the easy clearing method works, thanks now to the support for generated content. But sometimes problems show up in boxes following the one with the 'content-generated clear' , . Those problems (extra spacing and wrong float positioning) sometimes vanish as the page is resized (or re-flown for any other reason), and come back at a reload.
- In IE8 b1 a float may escape from the top of its containing block, when there is a previous float cleared by the containing block .
- In IE8 b1 a simple br with clear:both doesn't clear floats. This has always happened in older version of IE (now that hasLayout cannot be used anymore to get float enclosing, one expects that all other clearing methods work.)