I'm really interested to see that Guile is moving away from call/cc to prompts. A discussion of motivations: https://wingolog.org/pub/qc-2012-delimited-continuations-slides.pdf [PDF]
miniblog.
Related Posts
I've realised that AGENTS.md / CLAUDE.md solve two distinct problems:
* It tells the LLM the conventions of the project, avoiding follow-up prompts.
* It precomputes a summary of the project, saving a bunch of searches at the start of a session.
I'm amazed to learn that there is now a marketplace for buying and selling good prompts to midjourney/ChatGPT etc!
You pay and you receive a prompt that generates good results. For example:
Amazing project that trained the GPT-2 language model on text adventure games, producing a game that allows arbitrary prompts!
Game: https://aidungeon.io/
About: https://pcc.cs.byu.edu/2019/11/21/ai-dungeon-2-creating-infinitely-generated-text-adventures-with-deep-learning-language-models/
(Ironic that a text game has large GPU requirements to run too.)