zug_zug

A monad is a weird/unnecessary concept for a type that wraps another type (and makes certain guarantees). Like you can have a List<Int> so List is a monad type. There's also an "option" type in many languages (which is just a list of size 0 or 1), which is a monad.

Some people call futures monads.

But the real takeaway is that people pretend they want you to understand all this type stuff, but they don't, they want to make up new complicated words for obvious ideas that you already understand, so they can feel smart. Once you understand it they aren't "smarter" anymore, so they make sure to explain every weird term they made up with 5 other weird terms they made up and insisting that all this abstract theory is relevant to everybody when it's not.

Every chart shows a trend down from peak but still way up from 2019. Seems like what I’ll call a tee-up article - very fact-light with a provocative title so people can vent their preexisting frustrations after skimming it.

I like the connection between Goodheart's law and overfitting. However these examples are a reach:

--- Goal: Healthy population Proxy: Access to nutrient-rich food Strong version of Goodhart's law leads to: Obesity epidemic

I'm not sure I believe this one. Exactly who's target is "access to nutrient-rich food" and how would removing that target fix the US obesity epidemic? Is "nutrient-rich" a euphemism for high-calorie? My understanding is that there are plenty of places with high-nutrient food but different norms and much better health (e.g. Japan).

We can and do measure population health across (including obesity), this isn't a proxy for an unmeasurable thing.

--- Goal: Leaders that act in the best interests of the population Proxy: Leaders that have the most support in the population Strong version of Goodhart's law leads to: Leaders whose expertise and passions center narrowly around manipulating public opinion at the expense of social outcomes

Is this really a case of "overfitting from too much data"? Or is this just a case of "some things are hard to predict?" Or even, "it's hard to give politicians incentives." It'd be interesting if we gave presidents huge prizes if the country was better 20 years after they left office.

--- Goal: An informed, thoughtful, and involved populace Proxy: The ease with which people can share and find ideas Strong version of Goodhart's law leads to: Filter bubbles, conspiracy theories, parasitic memes, escalated tribalism

Is "the goal" really a thoughtful populace? Because every individual's goal is pleasure, and the companies goals are selling ads. So I don't know who's working on that goal.

That's such a weird attitude.

It's so weird to say "unproven" when you're talking about a 1000 chemicals, because statistically the odds that they are all harmless seems negligible.

Also strange unfair exercise you're setting up -- that we can't be upset about chemicals if we aren't exercising?

How about instead -- the FDA finds a short list of proven safe food-grade wrappers, and we all pay the extra 1 cent per product to use those? I really cannot understand your mentality at all here.

Yeah, important distinction. But also I think hopefully there's some understanding that random things aren't good.

My attitude used to be "who cares if there's a bit of random chemicals in your body, what's the worst that can happen?" but then I learned that tons of random chemicals happen to interact with our biology, like Xenoestrogens which are a whole class of chemicals (some in food packaging, like soda cans) that fit into estrogen receptors in the body and act like estrogens.

I feel like maybe you're intentionally misunderstanding the point I'm making, because your example kind of illustrates my point:

In that case you'd probably find nearly 100% agreement on steak over feces, or salad over feces, or fish over feces. You'd have complete agreement on any alternative to feces.

But if you read this thread, it's actually full of people actually saying "I was forced to use [alternative tool] and I prefer Jira". There is no tool that has 100% support over Jira.

In fact I'd be surprised if there's a single alternantive to Jira that even 40% of people agree is better.

Meh, I get people hate Jira, but every time I ask people "What would you rather do instead" there's absolutely no agreement on what's a better solution.

I think it's imoprtant to recognize that if a whole crowd of people say "We hate X" but everybody in that crowd hates eachothers proposals even more, it just creates a facade of agreement where there is none.

Well I would never use a script like this at a real job outside of local testing.

What you'd want on stage is probably using opentelemetry, which I believe has auto-instrumentation for all network calls. Then you have the data forever and it's in a public, shared platform that is permissioned and everybody knows, and will still exist in 10 years.

I've always used a proxy, like charles proxy, for this exact purpose.

A neutral middle-man that gives exact timing/response data.

This song is amazing, it's like midieaval FTL. The game's good too.

Because interoperability is key to competition...?

Imagine a universe where different hardware doesn't interoperate (e.g. androids can't call iphones, Macs can't use google.com, ChatGPT is only accessible to Comcast subscribers) I hope we can all agree it's a nightmare scenario. Sometimes the free market gets it right without intervention, sadly sometimes it does not.

Great, I'm sure he's equally excited if we have a future where we're tapping all billionaires phonecalls with their accountants to make sure they don't form a fake shell corporations to dodge taxes, or try to bribe their way out of a DUI ticket, or anything like that.

5 paper straws is probably better for the environment than one plastic straw.

If you can do this repeatedly, say 15 times, and record the numbers each time (say just put them in a spreadsheet), you can actually pretty easily run a "statistical test" to prove your theory at 95% confidence (which is what research journals often do).

You obviously don't need to justify your observations to anybody else, but if you did find a conclusive result it would make a good blog post.

[P.S. chat gpt could help with the statistical test part]

Yeah I guess my hot-take is that exaggerating summary articles maybe shouldn't exist at all, or not be shared on HN. What I've found is that, and this article is a great example, the summarizing article is more likely to lead to a less accurate picture than no article at all.

"Puberty hasn't changed" is not what this article concludes. And rightly so, because that's much too broad a statement do draw from taking 13 paleolithic skeletons.

I really don't like these articles that don't necessarily link to (or even have the same title as) the actual research.

I'd be curious if anybody is collecting distribution of the ages of puberty across country and across time. I don't think you can simply look at mean because there can be outliers in both directions.

So for what it's worth here's my experience -- generally not super impressed with supplements, tried cod-liver oil, didn't notice much effect, but after starting to make homemade salmon sashimi from costco salmon (Yes it's a thing, youtube it), I started noticing reduced anxiety and better concentration on days when I did so.

I would eat about 2/3rds of a pound of raw salmon and notice this effect (which would be like $60 or more at a sushi restaurant). I've done this at least 20 times now and never had any issues.

Generally I don't mention the mental effects of the salmon to people because I feel like they wouldn't take it very seriously, but since you're asking and it's on-topic, I thought I'd share.

It's interesting to read this study... the degree of complexity of the analysis reminds me of a overcomplicated codebase.

Would love to see completely independent analysis before trusting such a large codebase is bug-free.

> what does the evidence say? Afaict, not much.

That's only because we have no way to gather evidence, because almost everybody alive has microplastic exposure, so we don't have a control group.

It's just the nature of poorly understood psychology and the long human lifespan that most scientific research wouldn't be able to establish causality on a whole host of negative things (e.g. look how long it took to even get agreement that football concussions were detrimental in the long-term).

I cannot for the life of me understand why SRE of all roles would be the one to attempt to use agents for. IMO it's one of the last roles that it would apply, long after core development.

I mean is the AI going to read your sourcecode, read all your slack messages for context, login to all your observability tools, run repeated queries, come up with a hypothesis, test it against prod? Then run a blameles retrospective, institute new logging, modify the relevant processes with PRs, and create new alerts to proactively catch the problem?

As an aside - this is garbage attempt at an article, kinda saying nothing.

You're not explicitly saying to dumb it down, but I'd be curious if this advice would scale to genuinely deep content.

The cynical hypothesis is that people don't want to think about what you wrote, and if you put enough pretty pictures and big fonts that even people who didn't pay much attention will have a good time.

I'd be curious if perhaps a seminal blog post (e.g. one of PG's) could be rewritten in this style without sounding reductive and oversimplified.

I agree.

I do remember a study that people often think label their more popular friends as their "best" friends, but if you go ask THOSE friends, they label THEIR even more popular friends as their "BEST" friends. It's often asymmetrical.

Though tbh going too far down these rabbitholes usually isn't healthy/productive imo.

This feels like such a speculative non-article, trying to read really deep into a shallow interview.

All companies except the most successful companies in the world maybe.

Seems like a cool premise. Though I guess people building things always want to convince you they are worth-it (sort of a conflict-of-interest), would like to read an unbiased 7-day migration to this.

This is fascinating. So it looks to me like the proposals is that our brain has a map from 6 -> "multiples of 6" and 8 -> "multiples of 8" and it triggers both nodes, and that our brain then finds the MOST stimulated node (i.e. numbers that are multiples of both 6 and 8).

Because 24 is a multiple fo 6 and 8 it's causing interference.

This seems believable, curious if there's hard data on this (that indeed 6x8 is the hardest and other numbers with shared multiples) but not curious enough to look it up.

I was looking for the comment that simply said "This looks right, good work!" and since I couldn't find one, let it be me:

This looks right. Good work!

I agree about the bullshit meter. 20 years ago it was people claiming they could hear the difference if they had a headphone cable with gold-plating.

> it's just if there's no objective measure of how it's "nicer"

I think with all of these topics there is a fundamental test: make the person do a blind test and see if they can see the difference.

Sometimes people can (somebody challenged me to taste-test filtered water vs tap, a coworker could identify bottled coke vs normal) and others can't (a friend couldn't identify FLAC file vs normal).

If the person is afraid to do the test, that's a good indicator to me that they might already know it's more of a hype thing.

I'm sure there's a systematic solution to that.

If that's the problem, maybe require a $10 submission fee on each writing (or enough to pay for a basic sanity on writing quality). You can even optionally refund it for cases of obvious merit.

I read that article, but this article cleverly circumvents the challenges that article raised.

By not needing measure the microbiome the testing pains are avoided. And though mouse models aren't human, they are effectively blinded.

Seems like a very worthy finding that adds to the legitimacy of microbiome's significance.