both, but i reiterate that i did not mean to imply anything negative! (in general i think being weird is good and people should be more weird)

Ack, I didn't mean lacking substance! I don't mean anything negative at all, I just think it's interesting that such a small pool of people can (and do?) develop such a novel vernacular.

Ha, I'm not sure how I came up with that word. This says it's not well recognized but also somehow captures exactly my intended meaning.

@norvid-studies.bsky.social @gracekind.net

I think Bayesianism is mostly immune from this problem? (Odd, because I don't think it's *designed* to be immune to that issue...)

Don't choose! Just run for as long as you feel like running, then walk until you feel like running again. The dirty secret is that if you do this for a year or two, you'll find yourself running a few km with almost no effort.

Metacritic ranks them this way (only 78 for Hard Eight!?)

There Will Be Blood: 93
Licorice Pizza: 90
Phantom Thread: 90
Inherent Vice: 81
The Master: 86
Boogie Nights: 86
Punch-Drunk Love: 78
Hard Eight: 78
Magnolia: 77

What I find funny is I acknowledged that USUALLY people are biased to think everyone is more similar to us than they really are. But I claimed this was an exception.

I guess it isn't!?

That said, many people seem quite happy to dispute all three of the fundamental premises. I claimed that most people would only dispute #2, but it seems that was empirically incorrect!

Yeah I might not have worded that in the best way. I really only think that you should be *concerned*, i.e. that bad outcomes are possible. However, by stressing the word "obviously" I think I might be giving a different impression.

Hmmmmmmm

This is amazing. Might be worth somehow constraining the analysis to be shorter? Seems like it's currently describing the image so thoroughly that it's able to reproduce it without trying to guess the intent

This is an excellent conspiracy theory, but what's the safety angle?

they should all support full grammars (like llama.cpp) not just JSON

I'm mildly skeptical (but only mildly). On the one hand, they do have a plausible mechanism and the data seems good. On the other hand, polyps are increasing among all young people and they have no control group of similar non-ultramarathoners, so it's possible they just have very high sensitivity.

Scream all you want, all I've been doing is asking you to clarify what you mean when you say an analogy is invalid. (And also: Did I claim that property is true? I never used the word analogy.)

You have (still) failed to even identify a claim that you think I made that you think is correct. All I see is you pointing to some fuzzy concept of an analogy being "invalid". You realize that's not well-defined, yes?