Post Archive
› January 31, 2006
Internet Explorer 7 Beta 2 Preview
It's finally public. Read more about it on the ieblog: Please test your sites with IE7 and Frequently Asked Questions for the IE7 Beta 2 Preview with ample linkage from there to more resources.
One of the things they have done is to fix the "* html"-bug. Since this bug was frequently used to work around other bugs in the rendering, many people relied on it as a CSS filter. When discussing the changes, one of the things I've heard many many developers say is that Microsoft shouldn't fix the bug without fixing all the bugs it's used to work around.
Now, I wonder how far these developers have thought. First of all, a lot of these bugs have been addressed, so just because the bug is fixed doesn't mean you won't have to change your css. More likely than not, if the selector bug had remained but not the rendering bugs it's used to work around, then we'd have millions of currently perfectly working pages broken because the rendering issues fixed would have to be filtered for ie7 and not only ie6 and earlier. By removing the bug, they assure that practically all pages that currently use it will not need any changes that might affect ie6 and earlier at all, and hopefully most css that was served to modern browsers is now supported in ie7, making for fewer workarounds needed. Instead of reworking all css using it, you now just need to add the code needed for ie7.
Comments
1. February 1, 2006 04:20 AM
2. February 1, 2006 04:45 AM
victor Posted…
I'm not sure how what you're saying is different of what those many many developers say: the css selector bug should only be fixed once the other bugs are fixed, and not before, right? And still, people will still use IE5 and IE6 for a long time, so the style sheets will need to include the workarounds for a long time to come, anyway
3. February 1, 2006 06:54 AM
liorean Posted…
Vicotor: No. What I'm saying is that they are doing the right thing by fixing the selector bug when they have started fixing the rendering bugs, and that it would be a considerably worse choice to not fix the selector bug just because they have not fixed all the rendering bugs it's used to circumvent.
The reason is that if they fix the selector bug but still have some rendering bugs (and probably some added ones) then current pages will have to be modified to account for just one browser - ie7. But if they on the other hand don't fix the selector bug but fix some of the rendering bugs they are breaking the code that is used to correct ie6 and earlier rendering bugs. This means that all code that uses the bug would have to be rewritten for ie7. By fixing the bug they are effectively making the bug a better filter for ie6 and earlier (which are dead and thus stable).
4. February 1, 2006 11:19 AM
Andreas Posted…
It's nice to see improved CSS support in the IE7 beta2, but with it, also a number of bugs were introduced. The most annoying one I encountered (and one that will affect quite a number of CSS based sites) is the width miscalculation of an absolute positioned element with padding. Test case, screenshot on Flickr. Let's hope they fix it in the next release.
5. February 1, 2006 12:06 PM
Brady J. Frey Posted…
I agree to avoid hacks -- but until they catch up with the rest of the browser world, I'm not going out of my way to mop up their mess. They've had more than enough time to fix ALL their bugs.
6. February 2, 2006 12:18 PM
Andrew Massey Posted…
One thing I noticed instantly was that the pseudo class :hover is not being rendered when it's @imported. Hmmm, two steps forward...
Hayo Bethlehem Posted…
I'm seeing a lot of sites where the :hover link state doesn't work anymore in ie7b2. Anyone got a clue what that's about?