The 2022 Mant Prize is a monetary prize to encourage creative innovation in the exploration of reflexive and introspective interpreters and processes in the context of building intelligent, adaptive, flexible systems. One great inspiration for the prize is the work from the late 1970s and early 1980s by Jon Doyle and others at the MIT Artificial Intelligence lab.
(Please direct your close attention to the bolded words in the sentence above--an ideal entry should combine all these elements!)
I (William E. Byrd) announced the Mant Prize--and the associated QuIPS Prize--at the end of my 2021 Strange Loop keynote, Strange Dreams of Stranger Loops.
The Mant Prize is $1,000 USD, to be paid in a manner convenient and/or hilarious for both me and the recipient. I will announce the winner at the 2022 Strange Loop conference, which will be held Sept 22--Sept 24, 2022.
The deadline for entries is August 22nd, 2022, exactly one month before Strange Loop 2022.
Please email your entry to me at webyrd@gmail.com
with the subject heading Mant Prize
. You may submit a link to a public code repository, a public video, a public website or interactive tutorial, etc. However, whatever you submit (1) must be public and (2) must not have been publicly shared before. No recycling old blog posts, talks, or published papers! (After all, the point of the prize is to encourage playful innovation!)
Anyone may enter. I will gracefully refrain from entering myself, however. If you win and happen to be a 5-year-old Cat living in another Country/Planet/Galaxy/Universe who wants to remain anonymous, or whatever, getting the prize money to you may prove difficult. I'll do my best to work out something, though. Of course, it would be more fun to hand you the prize in person at Strange Loop 2022, modulo pandemics. We'll try to work that out as well.
The winner will be decided solely by me! If none of the entries Spark Joy in me, shock me, give me Nerd Chills or a sense of playful delight, or produce some other Visceral Reaction, or if all of the entries are Boring or Mundane, no prize will be rewarded, and I will instead blow the money on unbelievably delicious Japanese candy to console myself! However, I put a premium on experiencing Nerd Chills, and would much prefer to give away the prize, so please submit an entry of Face-Melting awesomeness and fun.
As I described in my Strange Loop keynote, I am especially interested in entries inspired by the reflexive interpreter work by Jon Doyle, or other introspective systems that can reason about themselves, and about other code or processes, for greater flexibility, and to avoid getting "stuck" doing the same thing, the same way.
Here are a few related resources that you might find helpful in understanding what the Mant Prize is about, and might also stimulate your brain:
My 2021 Strange Loop keynote, Strange Dreams of Stranger Loops, and the miniKanren code from my keynote.
Jon Doyle's 1978 PhD thesis proposal, "Reflexive Interpreters", which is the clearest explanation of the ideas that I hope will be explored in the Mant prize entries.
Jon Dolye's Why I Don't Play the Piano, which is a response to the piano analogy in Question 30 on page 8 of Brian Cantwell Smith's How is a Knowledge Representation System Like a Piano? A Questionnaire on Knowledge Representation Reseach. Doyle's response is a short but clear explanation of the ideas that I hope will be explored in the Mant prize entries.
Jon Doyle's 1980 PhD dissertation, 'A Model for Deliberation, Action, and Introspection'.
Jon Doyle's A Truth Maintenance System (MIT A.I. Memo 521, June 1979), in which he discusses the notion of arguing TMSes, dialectical arguments, modeling the belief's of others, etc.
Jon Doyle's publications page.
Building Problem Solvers by Kenneth D. Forbus and Johan de Kleer, which shows how to implement various types of Truth Maintenance Systems.
LISP 1.5 Programmer's Manual, with meta-circular interpreter on bottom of page 13.
Guy Lewis Steele Jr. and Gerald Jay Sussman. 'The Art of the Interpreter or, the Modularity Complex (Parts Zero, One, and Two)'. MIT AI Memo 453, May 1978.
Abstraction, Re-presentation, and Reflection, and Interpretation of Experience and Piaget's Approach. by Ernst von Glaserfeld, SRRI-209, 1989.
Interview with Jon Doyle by Heather Knight, 2006.
Interviews with other MIT AI lab researchers, including Minsky, Sussman, McDermott, etc.
Structure and Interpretation of Computer Programs (SICP), by Abelson, Sussman, and Sussman.
Software Design for Flexibility by Chris Hanson and Gerald Jay Sussman. Open Access HTML version.
Lisp in Small Pieces by Christian Queinnec, translated by Kathleen Callaway. Code from the book.
Essentials of Programming Languages, Third Edition (EOPL) by Friedman and Wand. Code from the book.
The Original 'Lambda Papers' by Guy L. Steele and Gerald Jay Sussman.
Relevant books by Douglas Hofstadter:
Douglas Hofstadter's talk The Nature of Categories and Concepts in which he tells the "Danny at the Grand Canyon" story. (Stanford Symbolic Systems Distinguished Speaker Lecture, Thursday, March 6, 2013)