This morning, I think we'll look at some code. As a start, I have a question about Room. Brief LLM-AI digression.

Ten thousand years before the events of Dune, mankind fought the Butlerian Jihad against the tyranny of machine intelligence. We need a Humans Union.

I did a little off-article cleaning of Cell. We'll look at that. And I have an idea which may be good, or then again, not. Weirdly, things go better than expected.

Yesterday's closing idea looks good to me. Let me know what you think, if you care to. We find that we're not quite finished. A serious lesson is relearned. LLM-AI diversion at end.

With our new path deviation operation working, it's time to look at the code and see if it's asking for anything. We end with a seriously questionable choice.

We need the validity check in PathHelper. And I see how to generalize it. Which should be first? Updated: found a test.

Let's see if we can make PathHelper less bad, perhaps even make it better.

We still get straight paths sometimes. I have an idea ...

We can make most of our paths wander, except for the straight left-right or up-down ones. I've had an idea. Is it a good one? I decide to use a prior idea, in an interesting spike.

When last we met, we had just discovered a nice way to make a wiggly path. Since then, I've improved the code a bit. We'll look at that and then see what we see.

Our minimal paths are too straight for most purposes. Is there a nice way to make them adjustably not so good? Is 'adjustably` a word? It is now. A nice discovery arises.

Yesterday at session end, I tried to hack in a minimal-path idea. Failed. Today let's try to fail better.

It is certainly no surprise that the day after something first works, there is plenty of room for improvement. We'll be looking for some of those things this morning. We learn from a small step.

Let's finish up the path between suites and see what it looks like. I suspect we'll want to try to tune it.

Nothing deep here, I mean a path in the dungeon, not a way of life.

We are working toward making dungeon description easier. I think of it as creating a little language. Today, adding randomness, if my luck holds. It holds, and we belay the whole plan.

Yes, the "AI Genie" seems actually to be useful. But it is also seductive, selectively harmful, and in important ways harmful to us all. Executive, manager, consultant, developer, ordinary person: Resist!

Last night was Tuesday night, i.e. Friday Geeks Night Out, and an inappropriate rant has got me thinking.

Well, dungeon CREATION. We'll start with a little refactoring. I remember why I recommend testing bug fixes.

Making a dungeon seems generally to require looping and even conditionals. Do we want special support for those?