Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[silverspawn]

Nikki Haley has reached a polling high. She's now only about 50 percentage points behind Trump. Ez.

[silverspawn]

Online chess has a huge cheating problem

[silverspawn]

honey has virtually infinite shelf life

[Awaclus]

In other news, I just found out that there's a 3rd party desktop program for ChatGPT which is based on the OpenAI API keys, which is nice for two reasons:

1) It's way more convenient than using a browser for it
2) You pay the API request rates instead of having to pay a monthly subscription fee to access GPT4, which could actually cost more money if you use it a lot, but I (and probably most people) don't use it anywhere near that much.

Does it know about current events? (The monthly version has a knowledge cutoff on January 2022.)

It has a cutoff on September 2021.

Right. So I said it definitely doesn't because it's not the fault of the test. If the two groups have different means, it seems like the test is working, and if the one group is small, well that's not really something the test can help. So Cohen's d seems more fair

The problem isn't that one group is small, the problem is that there is a ShiTton of people who have one of the attributes (the relevant MBTI type) and not the other (the profession). If there was a person in the world with two attributes that nobody else has (e.g. the only smoker is also the only person with lung cancer), that would be a strong correlation between the two attributes even though the group is small.

And it is actually the fault of the test that it picks up a ton of (for the purposes of identifying professions) false positives. Like, let's say that the aforementioned only person with lung cancer in the world also happens to be INTP, as smokers are somewhat more likely to actually be IRL. In this hypothetical case, it would certainly be true that 100% of the people with lung cancer are INTP — a remarkable statistic — but it would be insane to use the MBTI test to identify lung cancer. That's effectively what you're trying to do here.

[silverspawn]

And it is actually the fault of the test that it picks up a ton of (for the purposes of identifying professions) false positives. Like, let's say that the aforementioned only person with lung cancer in the world also happens to be INTP, as smokers are somewhat more likely to actually be IRL. In this hypothetical case, it would certainly be true that 100% of the people with lung cancer are INTP — a remarkable statistic — but it would be insane to use the MBTI test to identify lung cancer. That's effectively what you're trying to do here.

I really don't view it this way -- specifically, I wouldn't describe this as false positives. I think the test measures psychological traits, and those psychological traits correlate imperfectly with profession, and that's where the false positives come in. They're not false positives in what the test is measuring.

[Awaclus]

And it is actually the fault of the test that it picks up a ton of (for the purposes of identifying professions) false positives. Like, let's say that the aforementioned only person with lung cancer in the world also happens to be INTP, as smokers are somewhat more likely to actually be IRL. In this hypothetical case, it would certainly be true that 100% of the people with lung cancer are INTP — a remarkable statistic — but it would be insane to use the MBTI test to identify lung cancer. That's effectively what you're trying to do here.

I really don't view it this way -- specifically, I wouldn't describe this as false positives. I think the test measures psychological traits, and those psychological traits correlate imperfectly with profession, and that's where the false positives come in. They're not false positives in what the test is measuring.

Well, the test also measures the psychological traits inaccurately (which is hard to measure, but obviously true). Anyway, I guess you could say that it's not the test's fault that it usually fails to predict people's professions, but it is a bit of a weird framing IMO — if you need to predict someone's profession, then clearly MBTI is not the right tool for the job, regardless of whose fault it is. You wouldn't use MBTI to identify lung cancer just because it isn't the test's fault that it can't actually identify lung cancer.

Separately, the Myers–Briggs Company markets the test as though it could do a lot of things it actually can't, which is definitely the test's fault.

[silverspawn]

Well advertising is often nonsense, and I'm certainly not defending that.

[silverspawn]

Alright I'm again once again again at 1111188888880000000 again. Again. Also when you repeat a word like again again and again and then again, it starts to feel very again. Pronk.

KIMBA YOU'RE BAD

[silverspawn]

The fact that there are people who think you shouldn't place Oxford commas is pretty triggering to me. I don't really care that one in 100 sentences is ambiguous; even in perfectly clear sentences, it's just such an aesthetic violation. It's gross, ugly and wrong.

[silverspawn]

That itself is enough reason to prefer English over German.

[silverspawn]

It's gross, ugly and wrong.

and it reads "[gross] {pause} [ugly and wrong]" which MAKES NO SENSE

[silverspawn]

I should just ban chess to the last 4 hours of every day. Why is that so difficult?

[silverspawn]

In fact that's what I'm gonna do from now on! Until, say, the end of the week, then I'll reevaluate

[silverspawn]

The best argument against the Sam-Harrisian idea that people genuinely derive their values from scripture is that you don't see any pro-choice people making religious arguments. The bible simply isn't anti abortion, and this fact doesn't seem to make any difference.

[silverspawn]

In fact, the idea that the bible is anti abortion seems like one of the most impressive myths of our time. I think something about the idea that all life is sacred just hacks into people's aesthetics so directly that it compensates for the utter lack of scriptural support. And it's genuinely an exception; it's not like other religious ideas have no connection to what the book says.

[heron]

I don't understand why it would be interesting if myers-briggs predicted being into esports or whatever.
You could make a personality test predict this kind of thing very well by asking questions like "are you into esports" on the test.
Obviously the myers-briggs test doesn't do exactly that, but it's not that far off in spirit?
I'm not impressed that the test divides people into sixteen groups and people within the groups have some similarities between them.

[silverspawn]

I tend to find personality tests impressive to the degree that they can conclude things that aren't obviously related to the stuff they ask. The questions in MBTI tend to not obviously be relevant for most professions iirc

[Awaclus]

I tend to find personality tests impressive to the degree that they can conclude things that aren't obviously related to the stuff they ask. The questions in MBTI tend to not obviously be relevant for most professions iirc

The I/E questions are pretty relevant for a lot of professions.

By the way, the last time I took a personality test was 12 days ago (it was the ClearerThinking one) and just now I took the 16personalities one to remind myself what the questions are like in a test that is exclusively designed for MBTI results. ClearerThinking gave me 72% E, 99% N, 84% T, 98% P (these percentages are what % of users have a lower score or a score in the other direction, not about the maximum possible scores; I believe the 16personalities percentages are out of the maximum possible scores instead of being a comparison to other users).

My 16personalities result just now: 76% I, 76% N, 72% T, 72% P

I did have a couple of moderately difficult social interactions in the last few days so I guess that affected my result, and the test results aren't directly comparable because they're different tests and they report the results differently, but even then an almost 50pp swing in the most legit axis just because of something like that is ridiculous, and in practice people do probably use various different tests because the real one is proprietary.

[silverspawn]

I just opened my eth wallet and then suddenly remembered that I had a dream that eth went down to almost 500

[silverspawn]

In fact that's what I'm gonna do from now on! Until, say, the end of the week, then I'll reevaluate

I failed T_T

[silverspawn]

I'll try another policy. I can play chess whenever, but if I play poorly in a game, I have to take a 90 minutes break. Result doesn't matter.

It really is astonishingly difficult to get my chess playing under control. I usually don't struggle with addictive behavior, at least not nearly as much.

[Awaclus]

Why is competitive gaming so great that even your subconscious recognizes it as the most reasonable thing to be doing?

[silverspawn]

Every time GPT-4 gets updated I get this rush of existential anxiety. That said, this update is much less scary than the one a few months ago where it attempted to mimick sequential reasoning

[silverspawn]

I suppose there was no way they wouldn't do this eventually. In retrospect that was true for the aforementioned update as well.

[silverspawn]

Although right now it's using it for a question that I think it would easily be able to answer without searching the internet first. That kind of makes it less useful as an alternative to google just because of the time.