I disagree with the sentiment. Who is made worse off, who is made poorer because he held Microsoft shares? OP describes it as a "policy failure"... what kind of social wrong does a "policy" which would have forced him to sell his shares in 2015 correct?
Wealth doesn't circulate. Currency circulates. They aren't the same thing. And a "concentration of wealth" can't lead to inflation or debasement, I'm not sure what that sentence means.
Pretty ironic that you'd call out economic chops when this is a very basic and well understood principle. Perhaps be more cautious and less confident in your presuppositions going forward.
Yes and: Our economy now has fewer young(er) small- to mid-sized companies. The kind that have historically been the engines for creating jobs, for creating wealth (vs merely transferring it), and innovation.
I was excited by all new green energy startups due to Biden Admin's policies (BIL, IRA, etc). Oh well...
I know you know all this; I'm just compelled to repeatedly point out the obvious.
Others have given some answer to who was made poorer by Ballmer holding Microsoft shares, but I'd argue that this is the wrong question. Instead of looking at a specific individual, we should look at systems.
A system that allows this kind of extreme wealth accumulation is quite fundamentally at odds with democracy because extreme wealth can be and is in practice used to influence politics in a way that undermines democracy.
Some people might not care about that, but if your goal is improving the outcomes of the largest number of people, then pretty much everything else is secondary to having a functioning democracy.
This is genuinely the only response I ever get - that wealth can be used to influence politics. In my view this is a poor argument for two reasons.
1. The amount wealth actually influences politics is hard to measure but likely much lower than most people assume. Trump was outspent significantly both times he won. Bloomberg dropped $1B in a couple months and won nowhere but American Somoa. Probably the two biggest boogiemen, Koch and Soros, have spent billions over the years on their causes, and the present administration and general overton window is actually something neither of them like! The nonelected king-makers in American and EU politics are not actually wealthy people at all, just those with a lot of accumulated political capital, for instance, Jim Clyburn who single-handedly gave the 2020 nomination to Biden.
2. The amount that it takes to finance initiatives is much lower than centibillionaire level. Is the original $20B OP mentioned not sufficient to finance some ballot initiative? Why is it the increase to $130B that causes concern? The truth is that even a wealthy non-billionaire can easily do that, or bankroll someone's run for congress, or fund a partisan thinktank. The maximum level of wealth you'd have to set your ban-limit at would be problematically low.
It's interesting how quickly my brain developed an AI detector for written language (this blog post screams ChatGPT).
I wonder if it will stay effective, or if LLMs figure out a way around it? Or maybe it's just that this is the new way that technical blog posts are written, sort of how nearly all press releases feel univocal.
Is there a term for this style of writing? It's like a longform linkedin post. I don't mean this as a criticism, it just seems like written language is evolving to better capture our diminishing attention spans.
I've had a hunch it's about the shrinking width of pages. The text-width is about 11-13 words (at a max-width of 728) and the author seems to have written around this.
Because it's so thin, anything longer than 12 words becomes a paragraph which "slows" the reader, so there are a lot of punchy short sentences. This style would look silly if it was written on a wider left-aligned blog.
That's funny you mention text-width, I've noticed that with "punchy" articles like these.
Alternatively, when I see the old-school blogs that fill the entire page-width, I get the instant feeling I'm about to read something opinionated and quixotic.
Sure enough, the last HN article with that style hit it-- let's rescue the web by sending around WASM-blobs to be rendered to a common Wayland-like compositing surface. Thanks again, default CSS!
It gives me serious TED talk vibes. I guess it could be called that. Or maybe it's the twitter thread style?
A lot of catchy one liner hook sentences ("they're literally removing concession stands from NFL stadiums!") that sort of add up to make the author's point.
I'm pretty sure fact checking line by line will make the whole thing less impressive.
I am a fan of Carroll Quigley's style of writing (Georgetown University professor and ole Slick-Willie's mentor). I find having DeepSeek restate articles in his style to be much more enjoyably ingestible.
Wow, I just followed your suggestion and the output text was exactly the style of writing I enjoy and try to employ myself. Gonna check Quigley out. Thank you!
... which is intended to present detailed, long-form treatments of a subject yet, at the same time, provide something interesting and actionable in a brief moment.
Yeah, it's got that "visionary thought leader" vibe dialed up to 11. There's some interesting analysis in there, but it leans hard into the grandiosity—like he's the only one seeing the big picture while the rest of the world sleepwalks into a $100 trillion shift.
Hype cycle slop reminiscent of Business Insider, etc. Reminds of me engagement bait threads on Twitter which always include the “thread” emoji and the downward pointing finger emoji.
Incorrect. I worked for an Ag tech company for almost a decade. Ten years ago, farmers were downloading high resolution satellite imagery of their fields directly to their GPS enabled precision sprayers so that the spray rate would adjust continuously based on the vegetative index of the land they were over. I don't know what state of the art is today, but I imagine it would surprise folks who aren't in the field.
Right. There's a lot going on. Lots of stuff is measured that didn't used to be be. A big controversy is over who owns that data. Deere tries to own it.
Vision systems for weeding are effective. Instead of spraying everything, cameras look down as the sprayer is pulled behind a tractor, and only weeds get sprayed, or zapped, or hammered.
Uses far less pesticide. Deere and others sell this. Deere prefers to sell this as a service, where farmers pay fees when the vision system is enabled.[1] There are a bunch of autonomous robot startups with similar systems, but none seem to have grown much.
Tomato picking robots have been demoed since at least 2018, but still aren't used much.
There are at least half a dozen startups. It's not hard to do with off the shelf robots,
but not cost-effective yet.
Automated cow milking is used in New Zealand and Australia. Those countries have extensive agriculture but not much of an underclass, so they have to pay workers real money.[2]
Automated meat cutting is also used in those countries.[3] Mostly lamb, which isn't a big thing in the US. Fully automated beef lines don't seem to be available yet, although the company that builds the lamb systems is getting there.
Vision-based sorting is automated, fast, and cheap.[4] That's why when you buy packed berries, there are no bad ones any more.
All the big field crops - corn, wheat, hay, soybeans, cotton - were mechanized decades ago,
of course.
This is an unfortunate comparison. I actually chose Next.js because of its similarity to Rails - it's a batteries included, opinionated framework that favors convention over configuration (though it's not sold that way since these are not the currently trending buzzwords). There's absolutely nothing preventing you from using both tools. Rails works great as an API supporting a Next.js UI.
I'd say Next.js is the opposite of a "batteries included" framework. No abstractions for ORM, background jobs, sending emails, managing attachments, web socket communication - all very basic stuff when dealing with a production application.
It is a batteries included _front end_ framework. You don't need to worry about compiling, routing, code splitting, etc. Most of the things you described should be handled by the back end service
Hey, let's be fair here: Rails also doesn't have built-in authorization. You need something like Pundit or CanCanCan if you don't want to built it yourself.
Also Rails only recently got authentication. For more than a decade you needed Devise or something else.
I mean it has a router (2 actually), and NextAuth seems to be becoming something of a standard for many Next devs.
Meanwhile.. last I checked you still had to choose how you were going to roll your own auth in rails. Are people not often just installing bcrypt and adding a users table with their password hash? Or is there a generator for all that now?
Anyway, I disagree with the idea that Next is Rails-like. Adonis is probably still the closest in the JS/node ecosystem, though Redwood might also serve a similar niche for the types of apps it works for.
Next and the other "frontend metaframeworks" (as they're called now), are certainly much closer than the most popular choices 7 or 8 years ago (often cobbling together React and Express and an ORM like Prisma, making a bunch of other decisions, and then doing a bunch of the integration work by hand)
Right, so Devise seems like for rails it's what NextAuth is for Next? Though I don't know if there's anything equivalent to rails' code generation yet.
All these features are stateful or realtime. In a cloud/serverless world, they are all separate managed services ("compute/storage separation"). That's the trade-off of Next.js, greater productivity by standing on top of more hosted dependencies. Theoretically unlimited (within datacenter limits) scaling, bottlenecked only by your credit card.
By all means use Django if you specifically want to work in python but otherwise if you really want a Rails-esque framework why not just use full stack Rails?
You get much out of the box with Rails 8 now like deployment, caching, job queue, Hotwire, Turbo Frames and mobile.
DB access (drivers are automatically started, connected, and wired for use), queues, cron jobs, websockets, uploads, API helpers, simple routing, caches, indexes...
It gets ignored, but there are (sane) options. I'm quite proud of the APIs, too. Easy to learn, tidy, and everything just works.
Honest question - have you ever tried to use the actual web components API? It very much feels like a set of primitives to build a framework on top of. I can't imagine the amount of boiler plate you would need to write to just use web components.
An alternative interpretation of the Marshmallow Test is that it is a measurement of trust as much as it is of self control. If you don't believe that the researchers are going to give you the two marshmallows, then you're not going to wait.
I suspect (or hope) that many professional psychologists are beginning to doubt that data acquired in these contrived laboratory settings can provide a window into actual human behavior at all.
