I recently completed a failed font experiment, and, although I don't know if it's of any interest to anyone, I thought I'd share it anyway. If my wasted time fuels someone else's creative endeavors, then it wasn't entirely wasted.
The concept was weighting glyphs in a font based on their frequency of use in a given text canon. So, for example, my plan was to have a Shakespeare version of the font, a Brontë version of the font, etc. The less frequently a character is used in the canon, the darker the character would be in the font.
In any case, a more detailed writeup that includes python scripts and examples can be read here.