"Cur nunc Latinam studeo": but why now?

So, my new spontaneous interest is to learn Latin.  But why?  What could I possibly gain from that?  Not much, I guess, but then again, what do I really gain from creating algorithms for generating mazes on n-sided polygon tessellations?  The answer is simple.  I have long enjoyed, and have been very good at, studying and learning languages, but... I don't enjoy talking to people.  Latin is a dead language.  Thus it is the obvious choice.

An overview of Russell's paradox

This will be the first in a set of two articles about Russell's paradox.  Russell's paradox, discovered in 1901 by mathematitian Bertrand Russell,  is an antinomy that illustrates that certain attempted formalizations of naive set theory lead to contradiction.  Most simply, it is defined as:

Soaring pantless through the forest: an experience in social nudism

The mission was simple.  There were 4 bananas in the kitchen across the park that required retrieval.  Two minutes later, I found myself completely naked, dashing through cold rain in the dark, carrying a floral-print umbrella, transporting four bananas on whose skins someone had written their name in red marker.  In the distance another gentleman stood under his umbrella, pantless. An overwhelming thought came into my head... "My god, if I were to describe this event to someone, some context would certainly be necessary lest I be perceived as completely, totally, insane."

Maze generation on n-sided polygon grids

I wanted to see if a maze could be generated on a grid that is not made of rectangles, but instead of various interlaced polygons.  In two dimensions, of course, triangles and hexagons are the only geometries that make sense.

Recipe: recursive self-referential structure in rails

Here is a solution for displaying nested data in a tree-like structure n-levels deep.  Regardless of how deep the nesting is to go, only one flat table is needed. Suppose, for example, there is a table of "groups," where a group can belong to any other group, nested as deeply as is wanted.  The desired output would need to clearly show the relationships, for example: