I find the people who end up with spaghetti code did so because they didn’t translate their normal processes over.

Being completely methodical about development really helps. obra/superpowers, for example, gets close but I think it overindexes on testing and doesn’t go far enough with design document templates, planning, code style guides, code reviews, and more.

Being methodical about it takes more time, but prevents a good bit of the tech debt.

Planning modes help, but they are similarly not methodical enough.

That works until you make a plan/tests/etc, set the thing loose, and then when it has trouble it decides "actually the pragmatic thing would be [diverge from the plan/change the tests/etc]" and goes off the rails. I'm so frustrated by these things right now.

I have honestly not had that problem much. Being specific, concise, and strong with your prompts helps out a lot.

It would be interesting to make that sort of thing generic. A law couldn’t legalize something without backporting the legalization.

I think people rarely reject even if it fails a slump test though

The sklearn to me is (and mirrors) the insane amount of engineering that exists/existed to bring Jupyter notebooks to something more prod-worthy and reproducible. There’s always going to be re-engineering of these things, you don’t need to use the same tools for all use cases

Hmm not quite what I meant. Sklearn has it's place in every ML toolbox, I'll use it to experiment and train my model. However for deploying it, I can e.g. just grab the weights of the model and run it with numpy in production without needing the heavy dependencies that sklearn adds.

In amsterdam, few people wear modern/synthetic rain coats as well. Just riding around in the rain with what I assume must be waxed duck out something

I’d always hoped something like this could take advantage of FPGAs directly

FPGAs won't rebuild fast enough for it to matter vs software simulation I'd wager. Even FPGA-in-CPU has been a dream for decades and there you have more time for some workloads, still never was commercially viable for general computing.

There was research a few years back that tried doing something like this with an FPGA, and they found that their algorithm actually exploited defects in the particular chip (not the model, the actual single specific chip) they were using to use electrical interference for computation that shouldn't have worked on paper. They could not reproduce their design on another FPGA of the same model from the same lot.

11MPG/7MPG if towing. My Uncle was a traveling pipefitter and had the ‘87. The two tanks were _just_ big enough to make it from Phoenix to Yuma after a miscalculation of where the next gas station would be.

I drive a '91 GMC half-ton, with a 350 v8. I consistently get 14MPG around town and 16MPG on the highway.

I don't drive enough to justify the expense of something more efficient, and it was my grandfather's. I'm content to just keep it in good condition and drive it until it's either unrepairable or one of my kids wants it.

I’ve had a few different specialty brands of milk and there can be a difference, but I think that has more to do with the cows (and their diet) than the process. Jersey cow milk is probably more different than raw milk than pasteurized is from raw milk.

I am enjoying the RePPIT framework from Mihail Eric. I think it’s a better formalization of developing without resulting to personas.

The last year it has really gone down hill — hard. Reporting is mediocre, photojournalism is forgettable, and the opinion section is absolute garbage.

That's not just WaPo. That's the newspaper business. The business model for newspapers just doesn't work anymore, and they've all been trying to come to terms with it since Craigslist launched in 1995.