2008-07-12 02:26 UTC Minor Acid3 update – take two
As with everything with Acid3, the update I did earlier today ended up being more complicated than expected. It turns out the SVG working group got their errata wrong (instead of changing the strange requirement to a sensible one, they changed it to another strange requirement, and I doubt that's what they intended it to be).
At the same time, the CSS working group contacted me and asked for a change because apparently media queries rules for handling bogus keywords are in flux, and despite the interoperability we have today because of Acid3, they still want to change the behaviour to an arguably more sensible one. ☺
I've commented out the relevant parts of the test (you can see the commented out bits — search for "COMMENTED OUT" in the source), so that browsers and users using Acid3 as a vague indicator of compliance aren't subject to the prevailing winds on these minor issues. (WebKit and Opera builds should be back to 100/100.)
When Acid3 was written, I only tested specs that had been marked as "done" (in CR) for three years or more. Apparently specs take more than three years to bake! Oh well.
When we do Acid4 (probably around the time we have at least three major browsers shipping Acid3-passing browsers), I think we'll have to focus on testing fewer, more critical things. Acid3 tests a lot of critical stuff, but also checks a lot of less important stuff at the same time, and it's in those areas that we've had the most problems with specs changing under us.
In the meantime, if you want a weekend project, try printing out the Acid tests in various browsers and seeing if you can track down the bugs! There might be some legitimate differences due to the different media type, but in theory most bits of the tests should render fine. In my brief testing, the browsers seem to do much worse than I expected.