eob

Bit of a side question to the crowd:

If anyone wants to commission an AI novella, holler out to me. (For free -- this is a hobby)

I've been trying to build an automated book publisher for about a year now. While the books aren't yet "human good", the slop factor is getting fairly low. Part of that is because I've stuck to formats easier for AI to avoid the uncanney vally in: recipe books, historical re-tellings, etc.

With the speed of vibe coding.. I'd expect the "Reverse Engineer Anything" folks to come out with a "Man in the Middle Anything" app by lunch

A lot of us here are VC minded hackers *because* we spent years of our lives working minimum wage jobs in academia and then eventually realized we preferred to build!

Rest in peace Bob. One of the great early tech writers. The feeling of his writing -- the blend of fact and opinion and sense of humor, and more than anything the sense the really cared -- always felt to me like one of the seeds that developed into the blogosphere.

Simon I hope from this day hence, your bio always includes:

"Simon Willison, among other things, is an advocate for the inclusion of pelican geometry in LLM training datasets."

I think because you don't know /which/ developer you're going to get.

One interesting aspect of LLMs is that each one, weights frozen, can be thought of as a single developer whose work you have already evaluated.

The cost of finding, evaluating, and negotiating with a new human is tremenous.

These comments are so filled with cool projects and visualizations I’ll throw my own out there:

Trying to generate non-slop, print-worthy recipe books for different diaspora communities.

www.robotbookclub.com

I’m pretty sure the recipes have surpassed slop. The photos are pretty close. The layouts, intros, and “chef photos” need a lot of work though.

This is how I write down recipes for myself! There’s got to be some cognitive archetype that works better with this style encoding.

Do you have any suspicion about what is different between the backends?

That's an absolutely bonkers statistic: it would mean spurious differences in hosting container overwhelm the performance differences between models.

My guess it's just the emergent behavior that results when a company doesn't provide developers time to fix bugs.

If their week is already booked full just trying to keep up with the roadmap deadlines, a bug ticket feels like being tossed a 25lb weight when you're drowning.

You could say: "but have pride in your work!"

But if your company only values shipping, not fixing, that attitude doesn't make it through the first performance review.

You think like a person who’s debugged large systems failures before :). That feels very plausible.