Yep. I wrote bard, a Lisp that combines features of Dylan, Common Lisp, and pure-functional languages purely as a long-running experiment in what I could do to

Yep. I wrote bard, a Lisp that combines features of Dylan, Common Lisp, and pure-functional languages purely as a long-running experiment in what I could do to make programming more fun for myself. I subsequently used two versions of it in client projects (to prototype some solutions and to compile them to data used by the delivered product; nothing delivered to the clients depended on my weird Lisp dialect). I wrote a tool called nym to help generate names for characters and other things in games and stories. It's enough fun in itself that I sometimes just play with it. I wrote another tool called model-namegen to generate names with stricter constraints for science-fiction stories. Those names are mapped uniquely and reversibly to 64-bit integers used in the story to generate names for machine intelligences. I also wrote a prototype for a 3D multiplayer game called The Fabric set in the world of those stories, and wrote code to generate clusters of colored cubes for use as avatars representing the AIs. Each character's unique ID is mapped to a unique name and also to a unique 8x8x8 configuration of lighted cubes. An example name is Ixion Eleven Chrysotile, the name of the autonomous robot who is the narrator of my Fabric stories. Following is a link to a very short movie of the avatar for the character Miriam Five Bittersweet Earth: https://evins.net/downloads/miriam-5-bittersweet-earth-2-2.mp4 https://evins.net/downloads/miriam-5-bittersweet-earth-2-2.m... I wrote a couple of versions of a library for Common Lisp to make certain functional programming idioms I like more convenient for me in Lisp. They're called folio and folio2. I wrote a library called taps that similarly makes use of Richard Waters' SERIES package for Common Lisp a bit more convenient for me. Actually, taps started as part of folio and I ended up wanting to use it on its own often enough that I broke it out separately. I've written several little tools for calculating and keeping track of information that's useful for friends-and-family gameplay. I've written several other libraries and applications purely for the purpose of exploring some idea or other that interests me. For example, beasties was a pond-life simulation that I used as a playground for recreational genetic programming. Panels was an experiment in user interface that used a constraint solver for laying out windows and widgets, and an append-only log of all UI events together with a query engine and a knowledgebase. Panels could display graphic representations of recent user activity and retroactively define mappings from sequences of events to user-defined handlers. It operated on the principle that everything detectable that a user did potentially had some purpose, even if we don't know what it is, so we should remember all the user's actions and offer them the opportunity to define what they meant. I've also written a few little things primarily for the purpose of helping friends and family with this or that task. I've spent a lot of time over the years on things like these. As you would expect, they've taken varying amounts of time and I don't really keep track of how much. I use some of them (e.g. taps and other pieces of folio) almost every day, and others intermittently. A few of them (such as beasties and panels) I haven't really touched in years now.