I think most of these eponymous laws sound good, but don't pan out in the real world. In my experience, only about one in 10 seems to hold true.

As stated by the law. 90% of everything, including such laws, is crap.

Can’t tell if driest sense of humor ever or worlds most literal person.

Looks like humor to me.

wow, it even works on Hacker News comments!

That's the joke. You just restated the joke.

I'm posting a comment here so that the odds for better comments elsewhere may improve.

I'm the creator of the 1000 skeptics theorem:

You take infinite skeptics, give them all typewriters, poof! All the great works debunked.

I get the humor, but I do feel as if there's some converse to Sturgeon's Law in certain areas, in that for some things, 90% of everything is more than sufficient for most people or situations.

Athletics is maybe a good example. A bunch of caveats aside, most products being marketed to the broadest markets will more than suffice, and expensive alternatives have only marginal actual utility for a miniscule subset of those markets. There have been studies showing, for example, that the actual gains for use of super running shoes are very small in absolute terms, and generally only apply to elite runners; typical runners, even good but not elite runners, may not see any gains, and may even lose time from them. What gains they might have are not going to make a difference to their times in the grand scheme of things. I think similar arguments could be made for a lot of cycling products etc.

Maybe they're different issues but sometimes I feel like Sturgeon's Law applies well, and sometimes the opposite is true.

When I'm buying something in a new category that I don't know much about, I'll go and buy something that's above the bottom tier. Often that's good enough for me, and other times I'll learn there's something lacking and find one that satisfies that. In certain areas I'll simply splurge because I can justify it from the amount of use or value it adds to my enjoyment (audio gear) or work.

I see what you did there :-p

A modern incarnation of that is to say the quality of something has "a long-tailed distribution." Calling things Pareto distributed is similar, along with the more flippant, "pa-ree-to, pa-reh-to" where instead of being thought terminating clichés they're more like depersonalizing clichés.

99.9 if you include everything that most of us don’t see but is available somewhere

Only 90%?

Take for example elo ranking in video games

Top10% or top5% may sound good

But the gap between top10% and top1% (let alone top 0.1%) is so huge that you could say that those ppl barely understand this game

Same when top 1% is matched against 0.1%

You're using too broad a definition of "crap".

The top ten percent is not actual feces. It's just varying degrees of mediocre.

Is this really the case?

A top 1% ranked Dota player is about Divine 5, the only higher bracket being Immortal and its associated ranking individual. I think you're really stretching the idiom if you're implying Divine 5 players (most of whom have 3000+ hours) "barely understand this game." Obviously a team of immortals shouldn't ever lose to a team of divines, but I don't think it would have the appearance of a competition of total newbies vs seasoned pros.

There is actually a different rule at play: People ranked below you barely understand the game; people ranked above you are nerds and no-lifers.

Yeah this is the real advantage of AI. Optimize that 90% -> 99.9999%!

This is brought up every time percentiles are brought up on Hacker News.

Based on this logic, do you think the law should be that 99.9% of everything is crap?

It's just the good ol' power law at play.

90% are crap. 9% are so-so. 0.9% are ok. 0.09% are good, 0.009% are very good, ...

0.0009% are actual super-geniuses.

For the lack of better term. ( If anyone could introduce me to one, surely there is a term for it ) If we use the Normal distribution or empirical rule, we would get a top end number of ~83.9%, 97.5%, 99.6% of things are crap depending on your definition of crap.

And if your taste just sit between the 83.9 and 97.5 number. It turns out to be just about 90%.

Nowadays, with AI producing so much (crap) he'd say it was more like five nines of crap (99.999% is crap).

So by increasing the amount of crap by ten times, we get ten times more noncrap as well. What a deal

Also, 50% of everything is below average.

Pareto law. Not new.

I actually have a different version to this: 9X% of everything is crap, where X ranges from 0 to 9.999. That variation though is what differentiates a category and a time, since either 1 in 10, in 100 or 1 in 1000 can be a great find.

This is the reason why movies or music seemed better in other times, because X was close to 0 then and now it's 9.

I tried different note taking apps for years and you really have to try them to realize most of them are crap. I don't understand how people use flat notes like Obsidian, so I now rule those out. I want a hierarchical system. The last one I tried seemed very promising, it was actively worked on, looked nice, featureful, supported markdown. But after trying it for a weekend, I realized it was losing my notes! Not deleting my notes behind my back is kinda the most important feature for me in a note taking app.

Anyway, I'm back to using Zim Wiki event though it doesn't support markdown because, at least for my brain and usecase (which is being able to find business logic and workarounds to problems when I need it), it seems to be the only one that's not complete crap.

Obsidian has nested folders. Using tags to organise flatly is a choice. I organise with a hierarchy just fine - what are you looking for

Emacs includes a built-in mode named Org. Org-mode offers both hierarchy and tags. Org does many things, but it's especially accomplished at notetaking.

Elisp authors have written package extensions (aka "plug-ins") for Org-mode that provide even more powerful and specialized note-taking. Among the most popular today are org-roam (built on the zettlekasten method) and denote.el.

I used org-mode for a while, but I just don't like emacs to be honest. Especially on Windows (which I'm forced to use at work)... my experience it the plugins are just not written with Windows in mind, doing stuff like hard-coding link logic with unix syntax.

Also, at least for my use case which is to find random notes I took about stuff 2 years before and that's hidden God knows wheres, I find the way Zim does it to be superior because it allows fuzzy searching on a note basis.

For example: you can't remember how to kill and recreate a new default-vm to fix docker-machine, you search "default vm", and it will find all the files that have both "default" and "vm" in them, even if they're not on the same line.

Nested folders aren't really the same thing. I want the notes themselves to have subnotes.

The way Zim implements that is it creates a file for your note, then creates a folder that contains all the images you put in that note and any subnote.

You could technically do the same in Obsidian, but I don't want to micromanage the file/folder structure.