"The Joy of Finite Context Modelling" Explains the idea behind Markov chain compression techniques. This is one topic that can't really be tailored to the John/Jane Doe's of the world. You should hopefully understand some things about entropy coding/arithmetic coding as well as Markov chains (or at least know what a Markov chain is) before you can understand this. It also includes a few links just in case.
"A Recursive Approach for Octree Traversal" A small extension from a number of papers I'd seen using octrees to accelerate raytracing. It's possible that someone has already done the same, but here I have it in a relatively readable format. I bring up a few special cases that are not really mentioned in a lot of academian papers. Also, some of the papers I listed as references have errata that aren't so easy to find on the web. It is largely based on recursing the slabs approach for intersection between a ray and an axis-aligned bounding box. By exploiting certain properties of a standard octree, we have a fast traversal algorithm that can generate a list of all terminal(leaf) nodes intersected by a ray.
"Edge-Directed Interpolation" My attempt at demiystifying a really cool method of image interpolation that can allow you to scale an image up to higher resolution with minimal loss of quality. It's based on the paper by Xin Li and Michael Orchard. In addition to explaining the basic algorithm, I also introduce a handful of extensions and tricks one can use. It's pretty short and to the point.