IDTMC 1

A weekly newsletter of rabbitholes

Apr 27, 2025

I said I would try to write about one thing every weekend, no matter the quality/length. This week is David Deutsch’s ‘fun criterion’. I also want to link to ideas/thoughts/photos I came across that week.

Fun Criterion

'The fun criterion' is a rule of thumb for how rational knowledge creating agent (i.e. an individual, a school, a civilisation) should decide what to do next :

Choose the option that would be fun in the deepest sense of the word

What does fun mean?

There are two ways of defining it :

Everyday 'fun' : shallow pleasure, distraction, a break from work
Deutsch's idea of fun : the felt alignment that happens when your explicit goals and implicit unconscious expectations pull in the same direction. It feels like play, because there is no coercion. It's discovery for the pure sake of discovery. It's tackling a problem, for the purity of solving that problem.

Pertinent ideas I took away

Coercion kills creativity. When something is fun, you are in a generative error correcting mindset. You happily notice flaws, problems and you keep on improving your ideas
It's not that is shouldn't feel frustrating, difficult and effortful. Effort can still be fun. Long chess games, writing a proof, training for a marathon are hard, but can be fun if they satisfy the criterion that the effort is chosen rather than forced. The criterion tolerates difficulty and sacrifice so long as they are embraced for one's own reason rather than imposed
Boredom is a red flag : it usually means some part of you is being dragged along without being convinced. It's not that you stop at boredom, but become curious at why you might be bored. Take a different approach until curiosity emerges.
From an epistemological stance - Deutsch says that good explanations are created by conjecture (guesses) and criticism (ability to update based on observations). Coercion suppresses criticism- knowledge and creativity grows fastest in non-coercive fun environments
Morally - people are an end in themselves. Any scheme that requires grinding misery to succeed is ethically suspect. The universe is full of wonder and mystery - to crush that spirit of curiosity and wonder through 'grinding or forcing' - it actually touches a deep emotional part of me. Especially in children - if I ever see a child denied the ability to follow that curiosity - I’m deeply saddened.
Do it! It emphasises action. You learn what is fun by engaging with something and giving your full attention to it. Fun is emergent - trial and error. Conjecture and error correction. What you should't do it necessarily stop when it feels 'uncomfortable', but instead follow your intellectual curiosity. I think this is all 'felt'.

I'll end with a quote from the GOAT:

“If you're not having fun, you're not learning. There's a pleasure in finding things out.” - Richard Feynman

David Deutsch with Lulie on the Fun Criterion

Things I've been enjoying or thinking about

Picking a direction and just walking whilst being open to everything that arises.
Sitting in a cafe with a cup of coffee and going down giant internet/book rabbit holes.
I downloaded a local LLM voice model to my laptop, so just speaking everything rather than typing.
I uploaded almost a decade worth of typed journal entries (not daily) into the new O3 ChatGPT model and have been using it to analyse patterns that I'm not aware of. The annoying thing is I have so many physical journal entries, probably more than digital. Need to figure out a way to OCR my handwriting into text. I think from now on I'm going to be typing everything. It’s crazy that you can now revisit your past self so easily (I personally never actually go back and read anything I write). Half of the time my digital 'journal entries' are very cryptic. I have no idea what this means for example…
Writing on my blog and treating it more as a scrapbook of ideas. Wrote some thoughts on uncertainty.
The danger in AI-assisted writing and using LLMs is that you outsource your cognitive labor, but to understand deeply, you need to process it yourself. You need to go through the hard task of asking difficult questions, grappling with concepts and ideas, and explaining it to other people( writing or conversation). Cognitive load theory says that it needs to feel difficult for proper learning to happen

Quotes

I was thinking of a beauty that can only exist because of a fragility. All that we see comes from disorder and entropy, briefly stable and magnificent to our eyes...
Emergent complexity briefly defeats the void
Ince, Robin. The Importance of Being Interested: Adventures in Scientific Curiosity