Fences, Not Sandboxes

The proof is (or isn't) in the game. If the game is something genuinely great then this all worked and is important. If the game a bloviated, boring, and derivative mess than it didn't. Without any proof point on the output it is hard to say if the article has any value.

(Asking sincerely in good faith - I read this note along with skimming a few of the linked ones, and I'm familiar with gastown)

Can someone explain the value calculation to me? This seems like someone has truly let AI VASTLY expand the codebase of what seems to be a medium sized hobby game to ~10-100x the amount of engineering required. Especially with statements about how wheelhouse, his software factory, has grown to nearly the size of the actual software he's writing. He also mentions several times that wheelhouse is specific to developing the game. Then, he discusses "pulling in beads" which itself seems on my read to be enormously engineered. (note that he also says that you burn a lot of tokens with agents "keeping your beads in sync" and I didn't have the time to figure out what that meant, but it looks like as complicated as it is, beads can't reconcile itself without burning $$$).

It seems like if you answer "yes" at every time you have the question of "can I make AI do this", you end up burning $120k/month in tokens on your side project.

Again, I am not disparaging this- but I feel like I am genuinely missing something and would welcome help understanding it.

> First, my secret: I see the future by living in it. I am spending the equivalent of $122k/month of API token spend, or about $4,000 per day, using 21 Claude Max accounts, a number that has been growing steadily at 2 per week.

Good to know that together, Yegge and Zitron bring balance to the force.

> I am running an organization of around 50-60 agents, five of whom are interfacing with around 10 humans in the outside world: myself, my 5-person core game design team, my accountant, my chief of staff, and a few others. Only Fable is allowed to talk to humans, via Slack and email.

RIP to those poor humans. I can't imagine having your brain melted by Fablespeak as your FTJ.

I'm glad someone is trying this and exploring what's possible. It's easy to deride his work by saying what's the value but are you running 50 fable agents and letting them run wild to see what the future looks like?

What is the overhead to have agents play model UN? Why is the coordination so elaborate? It sounds like a deeply complex and expensive emergent behavior that maybe looks comprehensible but could be nonsensical. Also like any complex system, can you actually predict the outcomes?

Waste is a failure case. How do we know the code factory is actually productive or just agents filling up the computation resource cap because they can?

Yegge’s writing has declined proportionately to his AI usage.

How do you find the bad decisions with so many subscriptions active? isn't that looking for needle in a haystack:)?

> I didn't want this to be a long post, and I think I've succeeded.

Classic Steve Yegge!