Favorite Files

No favorite files added yet

Recent Posts

1. Comment - Latest Firefox beta passes Acid2 test, IE8 claims to pass also

   (Dec 21, 2007 - 3:38 AM)

   "Internet Explorer build 7.0.5730.11 at least apparently knows something has gone wrong, which is why the background has lit up red."
   
   oh boy, nothing could be more far from the truth. Please do your homework before writings about topics like this.
   
   The acid test combines lots and lots of tests that check for things like correct errorhandling and advanced formatting for both HTML and CSS. If a browser fails some aspects of the test, then the layout breaks down and some other parts of the page, that are _engineered_ *into the test page* will show up in broken browsers. Thus, the red background is an indicator of how *broken* IE7 is. The leaking red background is just some - usually invisible - formatting that the ACID2 page authors added to the page when putting together their test page, so that something red and screwy would become visible if a browser is broken in certain respects. Firefox2 is oviously *much* closer to getting Acid2 right than IE7.
   
   Also, (as others have pointed out), nightlies of Firefox3 seem to have been passing Acid2 for a year or so, which can be verified because these nightlies can and could be downloaded publicly.
   
   OTOH, IE8 is a private, in-development version, that nobody can get their hands on and about which nobody knows the overall state of maturity except that it seems to have been hacked forward to a state where the acid test page shows up (which means something because ACID2 really is a pretty nasty browser stress-test that encompasses some features which are simply not there in IE7, but it's far from saying whether IE8 is really mature, stable and overall reasonably standards-compliant).
   
   And concerning the official ACID2 test page. YES - the PAGE is broken. One of the aspects it tests is handling of 404 errors by browsers. It refers to nonexisting pages to test if browsers handle missing resources correctly.
   But for some stupid reason, their server seems to have been misconfigured to return the HTTP code 200 even if the requested resource is missing. This means that browsers gets the impression that the missing resource does actually exist and thus this misconfiguration is breaking the test even for browsers with *correct* error handling. This is not a deficiency in Firefox3 betas or Opera or Safari or any other browser which *can* render the test page. Those browsers can render the test page if it is served as intended (and others have pointed out the URL for a working mirror of the acid2 test page).