eru

11 ms·

421.
by eru 2mo ago
Well, at least afterwards the book is scanned and preserved digitally in their archives of training data. If the book was just rotting away in some forgotten bookstore, it would more likely be unceremoniously disposed off in the future with
422.
by eru 2mo ago
What you are suggesting also takes away a choice.
423.
by eru 2mo ago
Most old books aren't worth protecting, and the publishing industry destroys oodles of them as waste that no one wants.
424.
by eru 2mo ago
And the solution to the water issues can be found in any introductory textbook on the subject: a water price. > People keep pointing out, and it doesn't seem to sink in, that model trainers only have use for a single copy of a book.
425.
by eru 2mo ago
> Is the corpus of human knowledge useful for high quality AI training now essentially frozen in time? No. They also use lots of other methods to get training data.
426.
by eru 2mo ago
> They're destroying them so their competitors can't use them. That's pretty silly. My competitor can't ride my bike either, and I didn't have to destroy the bike for that to be true.
427.
by eru 2mo ago
My personal wastebook at home is so rare, it's unique. That doesn't mean it needs preservation.
428.
by eru 2mo ago
Why doesn't the competition undercut them, if it's such a no-brainer? It's not like Bose has a monopoly on headphones.
429.
by eru 2mo ago
You are probably right. I blame the Matrix movies for this: they made matrix sound cool for a while, and then oversaturation lead to it being less cool.
430.
by eru 2mo ago
Your preferences are not necessarily shared by other people. What's next, banning pineapple on pizza? > Because it's destroying the planet creating mountains of ewaste. Don't buy stuff, then? Or in general, ask your legis
431.
by eru 2mo ago
You can create something like verbraucher-rechte.de just fine.
432.
by eru 2mo ago
> That seems obviously stupid, right? Yes. > And the sort of thing that ought to be illegal? No, why? As long as you know ahead of time what you are getting, seems fine to me. People are also allowed to buy impractical shoes in stup
433.
by eru 2mo ago
> Tons of anorexya/bulimia issues would dissappear once people discovered that, well... almost no one it's a supermodel at all, and even that-supposedly-all-hot-girl for sure has some sort of issue with some spot or whatever in
434.
by eru 2mo ago
Well, apparently telling them that you have hold-out data (for eg a perforance optimisation challenge) seems to make them overfit less. So your idea might work.
435.
by eru 2mo ago
I'm still disappointed the ML people talk of tensors, but I barely ever even see them use a (proper) tensor product. It's all just matrix multiplication at most.
436.
by eru 2mo ago
'Some public spaces' is underselling the prevalence of FKK a bit.
437.
by eru 2mo ago
https://news.ycombinator.com/item?id=49371471
438.
by eru 2mo ago
Your problem only occurs when you try to naively transport equality of real numbers into equality of floating point numbers. https://www.netlib.org/fp/dtoa.c is how eg CPython parses literals like 0.3 into floating poi
439.
by eru 2mo ago
Huh, what? Floating point numbers have a standard, you know. They aren't non-deterministic YOLO numbers. By default, the compiler has to stick to what the standard requires, and can't just say add arbitrary imprecision. Your eps
440.
by eru 2mo ago
Thanks for putting in the effort. You might also https://www.redblobgames.com/ and learn a lot from how they do their presentations.
441.
by eru 2mo ago
Well, they don't produce the same result in floating point math, I'm afraid. So you'd need to teach your compiler about what your formulas mean and what context you are using them in. (Ie are you actually doing geometry, or
442.
by eru 2mo ago
Agreed. Though sometimes it's useful to keep track of 'fake' units like for angles, to make something like dimensional analysis work for you. But that's more for analysis of your code / formulas than when you actua
443.
by eru 2mo ago
Mostly agreed. However you can solve the problem of having x% of disks corrupted via error correcting codes. It's what Google already does for their own internal storage: when you own millions of hard disks, some of them will inevita
444.
by eru 2mo ago
In absolute terms, yes. In relative terms, no.
445.
by eru 2mo ago
Haskell has a pretty good story about how to use different dialects in the front end of the language, as long as the backend is still the same. Well, when I say Haskell, I mean ghc. Alas, C++ still mostly does its libraries via conceptual
446.
by eru 2mo ago
See https://en.wikipedia.org/wiki/The_Pinnacle@Duxton
447.
by eru 2mo ago
People already live in different altitudes, ie with vastly different amounts of rock between themselves and the bits of the earth that produce her magnetic field. So colour me skeptical of this claim as well. And while I like going barefoot
448.
by eru 2mo ago
Thanks for looking up the data. > Working class typically rely solely on wages and have little to no wealth. So any purchasing power decrease is often referred to as a hidden tax on wages. That's only true if wages are fixed. They
449.
by eru 2mo ago
> For Swedes it must be weird to be able to find a product named after yourself and your place! Plenty of people named Billy in the English speaking world, too.
450.
by eru 2mo ago
A (big) decrease in employment would be sufficient to tank the minimum wage, but it ain't necessary. I'd say, concentrating on employment is a bit of a red herring. A worker's compensation is a whole basket of pay, working c

More ›