vesterde

Still trying to get a Play Store page with stats onto my AI Impact page, but if their numbers are anything like Apple's (and they're likely much higher), then it's a tough life for them right now:

https://vester.si/ai-impact/appstore/

I made this bodily oddities page and posted it last week. The response has been great! I now have a very odd dataset on my hands of what people's bodies sometimes do.

What stood out from the submissions from readers was how many different things make people sneeze.

At the end of this post is a list of user-submitted reasons, and a few I added, I would love to see which apply to you, and you can submit more.

And thank you to everyone that gave feedback and suggestions, and voted on oddities. Keep it coming.

And to nip it in the bud, yes AI helped (is helping) with this.

To address all of the peer comments at the same time: I don't know.

There are places where men dancing is gay-ish and that deters them. That's an easy answer, but it's not enough because very progressive countries don't do much better on this, by my experience.

I would agree learning to follow in the very beginning can be easier, so that sounds like maybe men don't remain because their job is harder, but I wouldn't agree with this explanation because I observe fewer men showing up for beginner classes, when they don't know their job might is harder (whether or not it actually is).

Maybe men are more asocial on average? I'd be willing to believe that, especially in these times.

A comment above yours comments that this isn't true in Turkey, Buenos Aires, and I would add Paris to this list from experience, and then there's the entire continent of Africa we haven't mentioned and I can't really comment on, but what those places have in common is a cultural presence of dance, often a deep connection to dance where families dance together, maybe mom dances when she's cooking, dance isn't something you pick up as an exotic hobby, it's just that thing everyone does, and maybe you decide to do it a bit more than others.

Every culture with a lot of cultural dance presence would have a different view on this, I suspect.

If you're male (or want to dance as a leader), there are universally not enough of you to go around in partner dancing (with occasional exceptions), so any partner dance you pick you should be well received if you aren't a creep and just do the work to improve. Some of these dances are easier to get into than others. So your area might have salsa classes. Salsa is nice because it's easy to get into, but you can keep getting better at it forever. Bachata is similar but it's a bit more intimate in my view, so depends what you're after.

Some places just have "social dancing" classes, which is a different thing each week, but they might require you bring a partner, so look into it.

If you're from the US, or less commonly outside of the US, line dancing is a non-partner dance, so a group dance. One of my best friends does it and he loves it. Again don't need to be good to start.

I did not expect my weekend and week to be spent on the first page of HN, let me tell you.

I agree, and actually I put that asterisk there all that time ago when I wrote that article intending to put a footnote at the end, and clearly forgot. I meant to say that yes certainly people don't learn in the same way, but the thing I'm averse to is someone saying "I learn by seeing", and leaving it at that. It's simplistic, and I've found that individuals can also be wrong about their own opinion if their learning style.

The way I read some of the literature is that these may more often map to preferences, so the individual may prefer to see things, when learning, but it might not actually lead to more learning happening.

In dance I would yield that some people can repeat a movement just by seeing it, and others can't, but again I would say that's a learning style, I'd say there's a capacity inside them that is less strong/developed, translating movement seen to movement done. And it can be improved.

Several, but as the other comment pointed out, Urban Kiz is my main focus, as well as Tarraxo and Zouk.

That makes sense, thanks!

I've heard these fever dreams referred to as big-small as well. I believe it's the same phenomenon. For me it was pillars from high up above, where they're all the same width, but one is also much bigger at the same time. Hard to explain... Just an odd feeling.

And, anecdotally, I have gotten a similar feeling reaching deep enough meditation.

Generating those has been its own little project inside the project. Many iterations to go.

You've been great with clicking yes/no on the oddities, and providing more examples! I've already added some more, and refined some texts. I'll give it some more attention after I'm home again later today.

Thanks to everyone!

I started off having an nsfw tag, but it wasn't too populated, and decided for now to not include it. Might revisit.

Ye ye, she is frustrated it's taking so long to learn it. It's so valuable for divers.

I agree, the images will need iteration.

Oh great point this will go on the oddity once I add it!

I can't do it either... Needs a good song or scene or situation to trigger it...

My friend who does free diving would kill to be able to do it! She can't learn it.

Ah great! I'll put it on my list to read up on. Thanks

It's there! :)

Not that I know of. But some people so far visiting the site have voted that they do no get hiccups, which is very odd.

Please point out the claudeisms, I want to clear them all.

V1 looked like complete Opus slop, but since I make many little sites all the time I can get it to undo a lot of them, one at a time. I just explicitly tell it the aesthetic to avoid per-element (like the famous left border), and eventually it looks okish.

Edit: Ah and for text, Standard Technical English is a good instruction to give it, but I also told it to be dry and technical, and no embelish or do storytelling. And some manual editing out of cringe stuff still after that.

You're right, I moved it

To either my credit or against it, wikipedia provides a lot of the raw data. However that makes you feel.

I've been making this list for years as markdown files with a few sentences and links to wikipedia (or xkcd for his visual field comic), but the site is indeed made with the help of AI.

I keep being surprised at both which feelings are universal, as well as which aren't. I think I know the one your'e talking about but from being kicked in the balls... For jumping I might have experienced it playing video games and falling further than I expected.

So it seems, it just happens to include everyone I've ever talked to about it

Lovely! I mean odd… I will look into it, thanks!

I'm sorry this seems terribly HN of me to not answer your question and go on a tangent, but consider a retreat. If you want to give it a serious try, it's night and day compared to doing it at home for 20 minutes a day, especially in the beginning when nothing makes sense and it's hard to know which direction progress even is.

The first time I did a Vipassana retreat many years ago it really showed me that "oh, right, the mind can get into a veeery different state if given the right environment".

Some might say it's an artificial environment you can't replicate at home, and I agree, but by getting much further in such a dense time, it provided a better anchor for me to know what I'm even trying to achieve in my 20 minutes at home.

But it can be very intense, it's possible it would be too much for someone that isn't in an appropriate mindspace.

I'm a very unserious meditator, but when I sit down I at least know when I'm doing "right concentration" or not, so it's possible to have a direction.

I'm speaking very vaguely, but, you know... it comes with the territory.

And I know not all of these are going up because of AI. I have some controls, and some maybe don't deserve to be there at all.

I while back I made a page that showed the impact of AI just on GitHub stats (PRs, commits, etc...) which I liked because it showed that they are getting absolutely hammered.

Since then, I've wanted to see more graphs that are seemingly going vertical because of AI. But also going vertical downward (StackOverflow) because of AI.

So I found some primary data sources and packed them tight onto this page: https://vester.si/ai-impact/

Now includes data from lots of software things like GitHub, npm, pypi, but then slightly further from that with arxiv, crossref, domain stats (.ai tld as an example), and even US Jobs (BLS has a nice API).

I'm still working on some cleanup with another one or two datasources, but it would be interesting to hear if you know of some other numbers to put on here.

But please no paperclip factory stats.

Gemini models return a thinking signature that you, I think, must send back when invoking further, so they seem to keep them?