- IsTomWell, instead of one agent (the user) to phish there's two (both the user and the browser agent) and you only need to convince one.
- Probably there are less people excited to make another CRUD app than to write embedded code.
- There were surprising results in computational complexity before. We can't prove that P /= PSPACE which should be much easier to prove and yet people are stuck on trying to prove P /= NP. Personally I don't think that P = NP and if it were it would be very surprising, but "very surprising" in not "out of the question".
- I feel there is more than one type of mental fatigue. Some of them can be forced through (e.g. the emotional kind that happens when you have to do something you don't want to do) and some can't (e.g. not having enough sleep for prolonged time).
- Ah the famous AI efficiency and productivity boost.
- So why don't AMD and Valve release ICan'tBeliveit'sNotHDMI2.1 drivers?
- > It works via model building and next token prediction, which is not interpolative.
I'm not particularly well-versed in LLMs, but isn't there a step in there somewhere (latent space?) where you effectively interpolate in some high-dimensional space?
- That kind of sounds like they should be put in jail to stop this.
- > Try talking about biological operations without invoking “function”.
If you had a strong vendetta against mistaking map for territory, you could very well talk in terms of past survival and statistics. It's just not necessary for regular biological talk. It becomes relevant only when you start going to the boundaries.
> Why does striking a match against a matchbox consistently produce fire?
Because you wouldn't call these objects a "match" and a "matchbox" otherwise.
- > What's the motivation or goal? And why that goal is pursued? What is the force behind seeking of that goal?
There's no force and there's no goal. These things happen because every moment is a direct consequence of the previous one.
- Before 7th harmonic all you have is octaves, fifths and major thirds. If you want to stick to making other pitches out of stacked fifths and major thirds you'll end up with other compromises.
- 7/4 ratio should be simple, but it'll sound out of tune (over 30 cents) in a normal context. Many BP intervals are just as simple and they'll sound very out of tune to people unused to them.
- Seventh and eleventh harmonics are way off. If ratios were the fundamental thing Bohlen–Pierce would be about as pleasant as 12TET to people.
- > allowing multiple keys to be played without retuning the instrument
That's another simplification, real pianos are not tuned to 12TET, but use stretch tuning (which can be over 30cents off in the lower range).
I'd rather argue that people like what they're used to and so "people like pure ratios" is seemingly only true if "pure ratios" are not really pure. And that's ignoring a lot of the music that doesn't have roots in Europe.
- Typical person on the internet will be really used to 12 tone equal temperament, for just tuning most intervals are 10+ cents off even if you allow pretty large ratios: https://en.wikipedia.org/wiki/12_equal_temperament#/media/Fi...
- Many just intonation chords (where ratios are actually integers, not approximations of them) will sound out of tune to most people.
- > Humans everywhere seem wired to favour simple integer-ratio rhythms
That's what people write in the sheet music, but reality is more complicated than that. Notably in swung rhythms ratios are blurry (and dependent on BPM) and specific performers in band will play different ratios at the same time (e.g. drummer will play straighter, soloist will swing more).
- > As someone who has lived in several countries, I do not believe every developed nation achieves better healthcare outcomes than the United States.
Is that true for a median-wage earning person?
- I think OP means that it has come a long way from the simple mental model of µops being a direct execution of operations and with all the register renamings and so on
- I'd hypothesise that people doing small web kind of blogs just don't find GenAI interesting. They're doing things instead of trying to make LLMs do things.
