Why is there separate properties for each side of a border, but not for the outline property which is mostly the same thing but it doesn't affect layout? (sigh).

WTB outline-top-width, outline-top-color, outline-top-style, etc.

While writing HTML5, the W3C group seems to have broken compatibility with HTML 4 again. No longer can you specify colspan='0' on a td tag to let it span the rest of the row. (sigh) I guess it wasn't all that important since only firefox ever implemented it correctly in the first place. Opera did support rowspan='0' as well, but that is no longer valid in HTML 5 either.

Firefox misinterprets the CSS3's Editors draft and ignores any CSS properties (like transition) where an angle or time specification is given as simply 0. All other browsers accept 0 for the time specification, and the CSS 3 specification says that it is to be backwards compatible with the CSS 2.1 specification that clearly states that the unit of measure is optional when the unit is 0. Bad mozilla, bad!

Apparently, the optimization framework (Also known as Bundling & Minification) has a nasty bug in it when doing bundling of javascript. When given a set of files, it first does a simple concatenation of the files together before doing a minification on the result. At first glance that seems to be a fairly obvious thing to do, well, except that if you have a whole line comment as your last line in a javascript file that doesn't end with a CR/LF that is. Unfortunately, that is exactly what some of the more popular minification tools currently do. It'll concatenate the first line of the next javascript file onto the comment line of the last file, and well, things go badly from there.

Supposedly there is a fix coming, but not in time for our latest release. Yay for having to delete all our pre-min'ed files, and recreating them without the debug comment that is typically there to help debug in chrome. Oh well, at least there is a workaround.

Apparently Chrome has a bug when dealing with padding on elements with display:table. They've had one for a while, and it APPEARS they hacked it to always assume subtract the padding from the element's width, which is fine except for when you are in box-sizing:border, they still subtract it. Fail. Chrome breaking standards based layouts. Yes, it's been reported, "fixed" but broken in new and interesting ways -- for over a year.

As part of doing due diligence in creating websites that are accessible to the handicap, being able to use semantic markup is paramount. For forms, this means grouping logical sets of form elements, and wrapping them in a fieldset tag. As part of one of my most recent web designs, we used the fieldset to encapsulate groups of common tags that were collapsible. Well, that was until we found out that we had to bastardize our markup because firefox's implementation of the fieldset tag doesn't allow you to have content that is larger than the visible area that is. Yup, firefoxes implementation acts like it has overflow:visible stuck on, no matter what overflow setting you give it. Boo firefox for forcing us to wrap our pretty fieldsets in divs just because of your broken implementation.