Hacker Newsnew | past | comments | ask | show | jobs | submit | more aeternum's commentslogin

They also confiscate those hard-to-copy tokens if you acquire a sufficiently large quantity and attempt to leave the country with them.


I see a lot of hand wringing about this; but for 99.99% of people the banking layer and bureaucracy of modern monetary systems is a feature that protects them from fraudulent transactions, people stealing their credit card number, and businesses charging them and not delivering goods. These are generally good things.

Yes it is possible for the state to inflict violence on you, and if the state wants to, it probably will do so. Putting your money into internet tokens instead of state backed money will probably just get you tortured more until you give up the keys, or die. Crypto isn't some "one weird trick" to prevent the state from taking your property and possessions.


> for 99.99% of people the banking layer and bureaucracy of modern monetary systems is a feature that protects them from fraudulent transactions, people stealing their credit card number, and businesses charging them and not delivering goods. These are generally good things.

Let's go through these. To begin with, "fraudulent transactions" is redundant because that's either someone stealing your credit card number or someone you paid not doing what they said. So let's consider those two:

> people stealing their credit card number

This is the problem caused by the existing system, which is designed with such poor security that breaching a merchant allows the attackers to make charges to their innocent customers' cards at a different merchant. They get zero credit for providing a mitigation to the problem they created themselves.

> businesses charging them and not delivering goods

This gets sold as a benefit, but it's also a cost, because then it becomes a mechanism to commit fraud. People go to a business that does deliver the goods and issue a fraudulent chargeback. The merchants then have to pass the cost of that onto everyone else, which means that it's also a fraud against every other customer.

Meanwhile we have other solutions to that problem that don't do that. Established businesses don't want to ruin their reputation. If someone rips you off you can sue them. Sometimes you're just paying someone for something they're already delivered.

And most importantly, there instances when you would trust someone to deliver the goods independent of the payment system, and other instances when you wouldn't. Which is why you want both payment systems to be available instead of just the second one, so you don't have to pay for the chargeback fraud when you don't need to buy your trust from the payment system.


Very similar arguments were made for slavery. Giving up freedom for a promise of safety rarely the right choice.

While it is possible for the state to inflict violence, it's relatively difficult to scale. The state can freeze your USD accounts with the stroke of a key (as they did for Russian accounts recently). Whereas rounding up and torturing all those account-holders is just obviously infeasible.


Wait until you learn how much of life is gambling + entertainment. Is working for one of the many SV SaaS companies really that much different?


Seems foolish, I've been a long-term shareholder in Micron but this seems aimed at short-term profit maximization. So I guess I'll be a short-term shareholder now.


1) More of an emergent behavior than a dark pattern. 2) Imma let you finish but hallucinations was first.


A pattern is dark if intentional. I would say hallucinations are like CAP theorem, just the way it is. Sycophency is somewhat trained. But not a dark pattern either as it isn't totally intended.


Hallucinations are also trained by the incentive structure: reward for next-token prediction, no penalty for guessing.


That's not a matter of training, it's an inherent part of the architecture. The model has no idea of its own confidence in an answer. The servers get a full distribution of possible output tokens and they pick one (often the highest ranking one), but there is no way of knowing whether this token represents reality or just a plausible answer. This distribution is never fed back to the model so there is no possible way that it could know how confident it was in its own answer.


You could have the models output a confidence alongside next-token then weight the penalty by the confidence.


I believed this quote when I first read it.

I now see that both the heroes and villains in Atlas Shrugged turned out to be far more real that I could have imagined (the engine was still unnecessary BS though).


The central mythical figure in the novel is John Galt, a great inventor. What more critical piece of machinery could Rand have him contribute than an engine, the driver of the industrial revolution? It doesn't work to simply remove it from the plot. It would need to be replaced by something else, like Taggart's railroads, Rearden's steel or Roark's buildings.


It didn't have to be a mythical engine that ignores the most basic laws of physics. Just like with Rearden's steel, a more optimized version is often sufficient to change the world.

Both Elon and Bezos have significantly decreased cost to orbit and are very close to full reusability. This is alongside their other feats like a global logistic network that gets pretty much any product in the world to your doorstep in 1-2 days, self-driving cars, neural implants that enable mind-controlled computing. If that were in the novel, no one would believe it.


See also Heinlein's Waldo (1942) where the hero invents a receiver for an inexhaustible ambient energy source. I wonder if Rand read Astounding. To me, a good hard scifi story is allowed up to one impossibility anyway, as long as taken seriously.


A MacGuffin is just something that drives the narrative.


I think this quote is the kind of thing that sounds smart, but is actually devoid of meaningful criticism.

As you said, Atlas Shrugged touched a real conflicts that are rarely addressed. Is it kind of obtuse/allegorical? Yes. Would I like it to be a bit shorter? Yes. But it’s ideas seem generally right, the subject matter important, and under discussed.


But the "heroes" of Atlas Shrugged are, in fact, the villains of our modern world.


Strong disagree, the true villains have always been those attempting to convince you that human ingenuity and invention is bad. Without even mostly unknown inventions like the Haber process, over half of humanity would be starving right now.

Industrialists and inventors are the great heroes of our time.


> Strong disagree, the true villains have always been those attempting to convince you that human ingenuity and invention is bad.

Those don't exist. Effectively nobody actually believes that, or is trying to convince anyone else of it.

And if you think that "we should make sure no one person or small group has too much power (with wealth being a form of power), because we've seen the bad outcomes that produces" equates to "human ingenuity and invention is bad", you have been swallowing propaganda whole.


Europe in general has become the world leader in tech company extortion. If you run a tech company, be very careful about selling into that market. You should generally have an army of lawyers before making the leap.


You mean the EU regulating their market, right? I think TFA is about shareholder voting rights. Two different things.

> Norway's $2 trillion wealth fund said on Sunday it would vote for a shareholder proposal at the upcoming Microsoft annual general meeting requiring for a report on the risks of operating in countries with significant human rights concerns.

The Norway sovereign wealth fund is a large pension fund holding the profits of exploiting their natural oil reserves. It’s not related to the EU.


IMO it's not different, I've been at multiple companies shaken down by these kinds of groups. It goes like this:

1) A) Shareholder proposal bullied through via questionable means like buying votes from index-fund vote providers (or endowment/pension/wealth funds) that put other interests ahead of fiduciary interests. (and/or) B) Weak but expensive-to-fight lawsuit

2) But don't worry, we have a consulting arm that will do the reports for you. If you pay them, we can guarantee it satisfies the shareholder proposal, and we will drop the lawsuit.

Sometimes human rights, sometimes ADA, sometimes environmental, the playbook is basically the same.


Number 2 would be quite damning regardless of anything else, so if there's a lead for that definitely let us know. Assuming it's just nr 1, though: I think I see what you mean but at that point, I guess.. don't hate the player, hate the game? Bit overboard to call it "extortion".

And, yeah, if you're a >$tn company going into the EU, or even just giving out shares, I agree: consult lawyers :) sound advice. In fact they probably did.

On its face, I don't see a problem. If the situation is as insidious as you describe, I definitely agree with you. But then that's the story. Not TFA.


Investing 50 billion into a company and then requiring at least a minimal amount of accountability about the company's business practices is called 'extortion' now? That's hell of a hot take :) If the company has completely lost its moral compass when doing business, it is entirely normal for shareholders to be worried about the company's future.


You invest momey to make even more money. Did MS broke any US laws making more money for Norway? No.

If Norway has an issue with Microsoft, they should sell their shares in MSFT. Considering they’re not doing that, they are being dramatic for drama’s sake.


What do you think shareholder meetings are for?


My point was to the fact that they chose to have a fuss about what they're going to vote for, especially considering that they're a very small shareholder of MSFT.


You don't get to say why other people invest their money.


And the US is a world leader in corporate warfare supported by the three letter agencies.


they're a shareholder

it's THEIR company


How so?


Different than when the railroads dominated the market, or the industrials dominance in the 20s and 30s, or the nifty fifty, or the communication dominance in the late 90s?

Does the S&P 493 reveal a different economy or does it reveal that the author published an article based on feelings instead of research?


All those ended pretty badly, right?


Problem is, eventually everything ends badly. You can make this argument about "average S&P500 company", and the time in which it ends badly has gone down over time. In the last 10 years the average S&P500 company ends badly after about 18 years.

If memory serves it was about 30 years in 2000, and much longer before that.

By contrast most of these mania lasted a lot longer.


Seems more correct to say with different choices rather than the right choices.

For example plenty of energy likely means far more nuclear reactors, and the unfortunately truth is that most can be used to breed plutonium, so while we could be living with energy abundance, we also at least slightly increase the chance of global nuclear war. (IMO probably still worth it but it's no longer a super clear right/wrong).

With food, the problem is not producing it, it's the distribution. You aren't going to get it distributed in many countries that need it without overthrowing governments which will include lots of killing and bombing. Now it being the right choice to feed everyone is much murkier.


That person will get hit by the car and all their economic value will be lost.


That's an argument to reduce car usage and improve bike infrastructure


Not really because reducing car usage makes it worse. Less car congestion means higher potential and actual speed deltas. Increased likelihood of hitting a ped/bike rather than another car.


TOS in their current form should be made illegal.

Users don't/can't read them and service providers know it. There's no actual agreement because anything (kid, bot, dog) could click the button.


I wouldn't say they should be made illegal, but it would create a looot of jobs if customers had to participate in a workshop before being allowed to use some hardware or software the TOS of which contain--theoretically and/or practically--rather dodgy stuff. ...


I wouldn't say they should be made illegal

Or at least not enforceable since I can put just about anything in a ToS such as giving me your first born, all your money or any other shenanigans that do not strictly violate a law. Adding to this most here have violated the ToS/AuP on just about every platform they utilize at one point or another even if they do not realize it.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: