This has a lot of good stuff in it, but I think it's hurt by including stuff where you just have to make an assumption, E.g. I can interpret the following: "

This has a lot of good stuff in it, but I think it's hurt by including stuff where you just have to make an assumption, E.g. I can interpret the following: "Some\tText" in at least three different non-ridiculous ways, even assuming the encoding is known. It's worth keeping this in mind, but intertwining it with other tractable assumptions feels like it hides some of the super-useful nuggets contained within. That said, I'll admit the dividing line is fairly indistinct, plus it's still a great list.