I have blogged about miniKanren in general and specifically Adderall various times. While I still love the concept and idea behind the language, I have concluded that I’m not good enough with the language to actually write anything really productive. Thus, I’m going to set it aside and not use it for parts of the AI routines for pyherc. I still have the Reasoned Schemer on my desk and I plan to keep on learning the language, but I’m going to do it in separate projects and not as part of pyherc.
This of course means that now I have to come up with something that is:
I don’t know yet what it’ll be, but I’m confident that I’ll find something. Might even be something funny like macros that build code and reason about output based on miniKanren code.