cribbles

For those keeping track, this is now (at least) the third wholly or partially AI-generated article on the theme of "taste" as enduring source of human comparative advantage to hit the HN front page.

Previously:

Taste in the age of AI and LLMs - https://news.ycombinator.com/item?id=47677241

You Had No Taste Before AI - https://news.ycombinator.com/item?id=45288551

Yes. I use and enthusiastically endorse Math Academy. It is far and away the best self-learning educational resource I have ever tried; leagues better than Khan Academy.

I'm a self-motivated adult learner, so I don't know what it's like for kids. Though the program was originally designed for them, so I suspect their experience would broadly be similar to mine.

As other commenters have mentioned, you need to be okay with grinding through problem sets with no videos or UI pizzazz -- maybe this doesn't work for everybody. I'd compare it to the difference between trying to learn a language through scattered YouTube videos and Duolingo versus tandem and grinding on a good Anki set.

NB: I'm taking it for the Math for ML track and am currently most of the way through the Math Foundations III course. So I can only comment on the lower level courses.

These "what ifs" are kinda funny because the origins of JSX can be traced back to Facebook's XHP[1], which took explicit inspiration from E4X[2], an early JS standard that looked and behaved similar to the library described here.

[1] https://engineering.fb.com/2010/02/09/developer-tools/xhp-a-...

[2] https://en.m.wikipedia.org/wiki/ECMAScript_for_XML

The EU backed a good principle with a bad implementation. That is, users should be allowed to deny tracking, and -- here's the bad part -- companies must enforce this logic with a visual element that does not need to have a uniform UI.

A good implementation would be something like requiring enforcement of the Do Not Track header. In the article, the author suggests that companies should do so voluntarily. I agree, but then isn't that kind of the point? The EU should have made that the law in the first place. Instead, we have cookie banners.

> the scene is moving towards Leipzig

Speaking as someone who's split ~half my time between Leipzig and Berlin for the last 5 years, this is not true.

Leipzig's club scene is an extension of its university population. It's younger, straighter, whiter, and about two orders of magnitude smaller. People visit the clubs while they're going to school there, then they graduate and move elsewhere. Often to Berlin.

Why does this illusion exist? Because Leipzig is about an hour away from Berlin by train. Berliners visit for a weekend and think "wow, it's like Berlin in the 90s! Still cheap! And look at all these cool young kids at these scrappy clubs -- so that's where the underground has gone!". Then they go back to Berlin and spread the word to credulous out-of-towners, who go on to repeat this truism to people who have never visited either city.

In reality, Berlin's club scene -- both "mainstream" and "underground" -- dwarfs that of any other city. Nothing short of an asteroid hit is likely to change that.

If we limit this claim to _just_ German manufacturing competitiveness (which is often synonymous with "European manufacturing" in the context of these discussions -- and I will assume this is true of your comment as well, given the nuclear energy remark), this really isn't true.

What changed my view on this was one of Adam Tooze's newsletters from last year.[1] The relevant points here are:

- Germany manufacturing is less gas-intensive than the global average.

- Energy costs only constitute a small and decreasing share of total industrial costs -- about 5.8% for the German manufacturing industry as a whole, and 3% for leading export sectors (namely the auto industry).

- Most German manufacturing -- and this is true of European manufacturing more broadly -- is primarily in high-margin, value-added sectors where quality, rather than cost, are the competitive factor.

Another important detail lost in the invocation of "cheap gas" is that Europe as a whole, and German in particular, has never had particularly "cheap gas." European natural gas prices have long been well above those in the US, and Germany's have been above the European average for the last 15 years or so.

I agree with both Tooze and the majority view that, notwithstanding the above points, Germany's post-war energy policy has been a catastrophe. It has certainly limited its manufacturing potential. But the effect of the Ukrainian war on its _already_ limited manufacturing sector, due to _already_ not-so-cheap energy, tends to be exaggerated.

[1] https://adamtooze.substack.com/p/chartbook-150-why-cheap-rus...

Air travel accounted for 2.5% of CO2 emissions in 2020, and its total contribution to global warming was probably closer to around 3.5%.[1]

Only 11% of the world's population travelled by air in 2018, with at most 4% taking international flights, and 1% of the world's population accounting for more than half of total emissions.[2]

Passenger air travel is projected to grow by about 44% by 2050,[3] and will probably take up an even more substantial slice of overall emissions by then because technologies to decarbonize air travel (other than direct carbon capture) do not yet exist.

The argument you are making is probably least compelling when applied to air travel compared to any other form of consumerism, and HN's readership (generally speaking) is uniquely culpable here.

[1] https://ourworldindata.org/co2-emissions-from-aviation

[2] https://www.sciencedirect.com/science/article/pii/S095937802...

[3] https://www.eurocontrol.int/article/aviation-outlook-2050-ai...

“Curtis’s stitched-together compositions are less collages than they are Rorschach blots: look into their murk, and you can find your own worldview confirmed.”

This passage, which I see as being core to the author’s critique, doesn’t really jive with me. Basically he’s saying these movies are enigmatic, they offer space for reflection, they resonate with a lot of people across the ideological spectrum.

Well, you can look at that cynically, or you could say that’s precisely — almost definitionally — what makes them effective.

To go a little further than that: I don’t think it’s a fair claim. Each of his movies since Bitter Lake have had the same general arc of “emerging ideological apparatus promises to resolve social contradictions and empower the common person, fails to do so.” In this sense, his narrative angle is broadly _anti-confirmatory_.

Then there’s the aesthetic critique, which, whatever. I really can’t fault someone for finding fault with Adam Curtis’s style. And it’s least overbearing in TraumaZone out of all his movies I’ve seen, so I get why this author favors it.

As it happens, I’m about halfway through TraumaZone right now. It’s great. Poses some interesting questions, doesn’t offer any easy answers.

+1 daydream's comment on LLMs.

But I think the steam on FP hype was already starting to run out for unrelated reasons.

The briefest account I can offer on "peak FP" from my (web developer) perspective is that

1) functional programming was already enjoying a moment of renewed interest and vitality due to the increasing ubiquity of multi-core,

2) React/Redux -- which ~solved[1] many problems with increasingly complex frontend web/mobile state management -- really started to become mainstream around 2016-2017,

3) Node/TypeScript was in the midst of an popularity / enterprise adoption explosion (in part due to #1) and only served to amplify the general enthusiasm around FP among JS-literate engineers (in part due to #2).

In the intervening years, the React paradigm more-or-less "won" and multi-core is taken for granted. A huge chunk of our industry has probably never known a time where FP wasn't celebrated. For that reason, it no longer seems to be answering any pressing problems, and naturally there are fewer articles being written about it.

[1] I expect this to be a point of consternation, since this is HN, but the point is that React was at least _perceived_ to have been an antidote to a variety of issues people faced with Angular, Backbone, plain old jQuery apps, etc. and the unifying theme of those issues was (rightly or wrongly) perceived as "OO / mutable state bad."

I think the reluctance to blame phones is not just that people have a personal, dopaminergic fixation on them.

There are a couple of other factors at play:

1. Thinking people, especially of the sort that frequent Hacker News, want there to be a more complex and multivariate reason to the "big problems" like loneliness and alienations. It feels wrong to just say "phones bad."

2. Many of us grew up with emerging technologies, like video games and the internet itself, that were reflexively rejected by generations older than us for reasons that seem poorly thought out in retrospect. We pattern match on the type of blithe Luddism we grew up with, and are instinctively reluctant to say "phones bad" for fear of falling symptom to it.

But I agree with you, and the thesis of this article. The "phones" argument has such obvious explanatory power for a wide range of regressive social phenomena that there needs to be overwhelmingly strong evidence for some other catalyst to drop it from consideration.

I would greatly appreciate a moratorium on this genre of article until there is compelling accompanying evidence that a meaningful portion of ChatGPT's users are unaware of these shortcomings. I have yet to encounter or even hear of a non-technical person playing around with ChatGPT without stumbling into the type of confidently-stated absurdities and half-truths displayed in this article, and embracing that as a limitation of the tool.

It seems to me that the overwhelming majority of people working with ChatGPT are aware of the "con" described in this article -- even if they view it as a black box, like Google, and lack a top-level understanding of how an LLM works. Far greater misperceptions around ChatGPT prevail than the idea that it is an infallible source of knowledge.

I'm in my 30s, so I remember the very early days of Wikipedia and the crisis of epistemology it seemed to present. Can you really trust an encyclopedia anyone can edit? Well, yes and no -- it's a bit like a traditional encyclopedia in that way. The key point to observe is that two decades on, we're still using it, a lot, and the trite observation that it "could be wrong" has had next to no bearing on its social utility. Nor have repeated observations to that effect tended to generate much intellectually stimulating conversation.

So yeah, ChatGPT gets stuff wrong. That's the least interesting part of the story.