Thank you for resurfacing this, it has been my daily commute read for years, it was great!

"Any sufficiently complicated C or Fortran program contains an ad hoc, informally-specified, bug-ridden, slow implementation of half of Common Lisp."

-- Greenspun's tenth rule

He had some lack of conviction to scope it so narrowly.

With what credit card provider?

I've done it multiple times when a vendor wasn't behaving fairly and it always went through.

I don't recommend doing it to a vendor you plan to have business with again in the future as they might ban you (eg food delivery apps)

This is another problem with charge back, EBay is screwing you? Do you want to be black listed from eBay all of your life or lose that $100? Can you dare to chargeback Apple or Google?

It’s too risky tbh.

Evolution didn't spend millions of years so we could sit 8 hours a day and slouch through the rest.

It just an anchor. If it were 50k would you say the same down to 25k? And if so how many more times would it apply?

The interesting thing is that it was manageable solo (in many ways it's _more_ manageable solo+AIs than with coworkers+(their)AIs), and in such a short amount of time.

Original RSL library is 36k LoC. And this is C++. Rust should be like 50% smaller, that is, 18k LoC. This library is so big that I bet the author has no idea if it works or not. 1300 test generated by AI say nothing about actual quality.

In the end it is just a lot of unmaintainable code quickly generated by AI.

This is uncharitable, but makes a prediction. I imagine you'd bet the author won't be successfully using this, at MS/Uber or wherever they are, in a year time?

Rust makes no promise of being terser than C++, and RSL does less than this considering the optimization.

Also it's only 45/50k LOC so not so very from the 36k LOC.

Yes, I would bet it won't go anywhere.

The blog post mentioned the project is 130k LoC multiple times. Where 45/50k LoC comes from?

>Rust makes no promise of being terser than C++

True, but Rust has no header files, this alone is a great LoC saver.

50k LOC wouled be the rust code without tests.

But it's not apples to apples because they seem to have done much more performance work though, this is far from code golfing.

RSL’s 36k LoC includes tests and should be compared with 130k LoC, not 50.

Having 90k LoC of tests for 50k LoC codebase also a problem. At least in my experience LLM generate too many tests. It does not evolve test suite but throws more code into it as development happens. Unless I aggressively refactor tests I quickly end up with a test suite that I don’t understand. Then LLM modifies tests to “make code work” and I have no idea if this is a legit edit or LLM cheats. I wonder if the same thing is happening or about to happen with this codebase.

Has Rust code generally been found shorter than C++ in practice? I don't see an obvious reason for it.

I see no reason for Rust to be shorter than C++,. when using latest standards.

the interesting thing is how fast it becomes unmanagable.

Also that, I suspect that's correlated to how practical is to have multiple people (with their agents) iterating on it.

It might not be intended for the public.

Similar to your printer fingerprinting: https://en.wikipedia.org/wiki/Printer_tracking_dots

how can that possibly work while supporting offline backup & restore?

The compute power needed use to be of the order of 5s per password try. So it effectively mitigate brute force back them, you need a absurd compute power to crack them.

Moore law did its thing, now you can do it with a lot less computer power.

> Moore law did its thing, now you can do it with a lot less computer power.

s/power/time/ maybe? Or on second thought: so energy-efficicient that it actually uses less power in the same-or-shorter time… which brings me back to "less compute power".

The standard biology formula for heritability is h = Var(genetic) / Var(phenotypic)

Which I bet is very useful for some kind of technical work, but it's amusingly confusing to lay people.

The author goes on to critique its misuses but the textbook example to make clear "heritability" is not as obvious as it sounds is that by this definition human bipedalism heritability is near zero because there's near zero variance.

It's at odds with "heritable or not", the interesting question to us peasants. If there's a disease that gives you spots, we want to know if you can get it from your ma and pa. We don't want to know that nearly everybody hides the spots with makeup.

Then there's the matter of whether there's just a small population with the genes for it, and whether it's polygenic, or mitochondrial, or otherwise non-mendelian, and all that gets factored into this heritability value along with cultural things like the use of concealer and the probability of having your face torn off by a bear. It kind of reminds me of inflation, as a useful measure.

You can see why this is so frustrating for laypersons, but the point isn't that you can't use a colloquial meaning of the word when shooting the shit with your friends or whatever. It's just important to keep the rigorous definition separated from the informal definition when citing the literature, or you end up in weird places.

thinking about how to communicate it in a clear way. “the control knobs of what actually makes us different from one another” — you don’t expect one of your kids to be quadripedal. otoh this doesn’t really capture the precise notion either.

You should try all of them, then update your opinion about your information sources accordingly.

exe.dev landing page is sublime. The call to action is "ssh exe.dev" and you can bet it works.

I love the line on the landing page with a link back to hn:

> That must be worst website ever made.

the level of confidence (this is a second time founder after all) to put that on their website gives me confidence that they can make this work