Google released an interesting study about the content of over a billion webpages in their database. It’s nicely seperated into categories of markup, and identifies the most used HTML tags and attributes. I found the comments on the most frequent mistakes very interesting. Obviously many tags are used incorrectly for presentational (instead of semantic) purposes, and many pages are brimming with deprecated elements. A surprising number of pages had specialized tags, and in some cases, there’s no known online documentation that tells which program generated them or for what purpose. It’s a fun read.