100 years of whatever this will be • Hacker News Board

Re: 100 years of whatever this will be

#191

Post by colechristensen » Thu, Dec 02, 2021, 5:04 PM UTC

AlexanderTheGr8 wrote: post #91

chmod775 wrote: post #64

Ignoring the first couple questionable true-isms, a lot of these bullet points are just choices the author locked themselves into. The author is painting a picture in which AWS is the internet, or somehow representative of pricing. I don't have half of these problems. I have run services for $6,000/year on bare metal servers that would have cost $240,000/year on AWS at the time. Likely the author is unwilling to make…

replace AWS with and every bullet point is still True. Every website (the Internet) needs to be hosted. If not on a cloud provider, you are going to have to host it yourself, which is a ton of more work and more points of failure. For cloud providers, point of failure occurs if you don't pay them. When you are hosting yourself, there are tons of points of failure (electricity, maintenance, etc.)

It’s not really all that much more work. More like people who don’t know how to cook talking about how unreasonable making dinner instead of ordering it would be. Honestly the biggest benefit about the cloud is actually just hiding the details of how much things cost so that management can’t pinch pennies when it comes to the fine grained cost of operating tech and doesn’t have to spend the time making reports about it.

There are plenty of hosting options out there that just provide you with bare metal hardware and take care of the lower level maintenance.

Re: 100 years of whatever this will be

#192

Post by pphysch » Thu, Dec 02, 2021, 5:04 PM UTC

> Big governments really do nonspecifically just suck a lot.

> We don't need deregulation. We need better designed regulation.

There is a major contradiction in this line of reasoning that the author never addresses.

The only way to have successful regulation is to have a powerful regulator, a powerful governing organ. The government needs to be the most powerful "firm" in the market.

Otherwise, there will be a more powerful private firm or "trade association" (read: cartel) that is the de facto "shadow government" of one or more economic domains, that the de jure government cannot hold accountable. We see this a lot in America since the 70s, mainly expressed through regulatory capture.

Re: 100 years of whatever this will be

#193

Post by cs702 » Thu, Dec 02, 2021, 5:05 PM UTC

In practice, any distributed system that doesn't have an explicit hierarchy evolves to have an implicit hierarchy, because economies of scale and network effects favor the increasing concentration of resources and connectivity, respectively. The OP is right about that.

In any distributed system, those nodes that by luck or skill become more cost-efficient and inter-connected tend to become even more cost-efficient and inter-connected, because all other nodes want to connect via the most cost-efficient, most inter-connected nodes.

This dynamic is self-reinforcing. We are all familiar with it.

Without some form of regulation (historically centralized, and always imperfect), distributed systems like market-driven economies, modern financial markets, and information networks evolve to have growing inequality of resources and connectivity over time.

I'm not aware of any exception.

Re: 100 years of whatever this will be

#194

Post by dsign » Thu, Dec 02, 2021, 5:05 PM UTC

thesuitonym wrote: post #136

>Centrally planning a whole society clearly does not work (demonstrated, bloodily, several times). I really, really hate this argument. If it's true, why does the CIA feel the need to destabilize countries with centrally planned markets? Why has China been so successful despite what is widely believed to be a failed premise? Central planning works as well as unplanned markets work: That is, sometimes they work, somet…

> Central planning works as well as unplanned markets work: That is, sometimes they work, sometimes they don't, and it mostly comes down to external factors.

I have first-hand, extended experience with a hard-core centrally planned economy. It most definitely doesn't come down to external factors. Everything I saw and experienced was self-inflicted. And things keep getting worse.

Re: 100 years of whatever this will be

#195

Post by ro_bit » Thu, Dec 02, 2021, 5:05 PM UTC

> The most reasonable daycare and public transit in the Bay Area is available only with your Big Tech Employee ID card.

What's this about?

Re: 100 years of whatever this will be

#196

Post by xbar » Thu, Dec 02, 2021, 5:05 PM UTC

endisneigh wrote: post #182

zeroxfe wrote: post #149

Earlier quoted context omitted.

These arguments are not helpful because their metrics and criteria for what constitutes "works vs. does not work" are different from yours. > Why has China been so successful despite what is widely believed to be a failed premise? They're successful as an economy, not as a society (obv, using my metrics.)

By which criteria are they not successful as a society?

The ones that measure the success of a society. WEF's Global Competitiveness Report and Amnesty International, for example.

Re: 100 years of whatever this will be

#197

Post by whatshisface » Thu, Dec 02, 2021, 5:05 PM UTC

hn_throwaway_99 wrote: post #152

notpachet wrote: post #49

All of the current histrionics that we hear from DeFi advocates regarding escaping the evil centralized Banks and Regulators and what not -- you know what they remind me of? They remind me of what I hear from junior developers when they're tasked with fixing an antiquated, broken software system. These developers typically lack the experience and the patience necessary to fix the system incrementally from within, so…

Thanks, I think your comment clearly explains some ideas I've had for a while but have been struggling to elucidate. I think it's also why you see so much "pendulum-ism" in the tech world. Something about the current paradigm is difficult ("monoliths make it hard for large teams to release software quickly and independently!"), so then a new paradigm comes along which perhaps addresses some of the shortcomings of the…

I don't believe the broader financial system ever tried out making transactions impossible to reverse. They might not have ever had the power to do it.

Another thing to bear in mind is that none of the participants in the financial system have ever been interested in designing a financial system. They have been interested in making money.

Re: 100 years of whatever this will be

#198

Post by jakub_g » Thu, Dec 02, 2021, 5:06 PM UTC

Joeri wrote: post #157

_9omd wrote: post #92

Earlier quoted context omitted.

You make a good point about software development in general, but I don't think it applies to things as fundamentally revolutionary as crypto. Sometimes system _do_ have to be completely re-thought from the ground up. I don't believe it's possible for our current financial systems to morph into an open network for storing and transmitting value (the internet of money as some call crypto).

Let’s assume (and that’s a big if) that crypto is a fundamentally better foundation to base finance on. Even if it is, we cannot presently know that it is, and we cannot predict what its unique failure scenarios will be and how to counter them. Therefore for me arguing a headlong dive into crypto is indeed like a junior developer arguing for the big rewrite. I also fail to see the fundamental difference between crypt…

> Anyone can mine gold

Bizarre comment - there's only a handful of places in the world where gold is located (most countries have close to zero), and they're probably controlled by some powerful private entity and probably under a gov license.

Re: 100 years of whatever this will be

#199

Post by dirtshell » Thu, Dec 02, 2021, 5:06 PM UTC

kkjjkgjjgg wrote: post #132

"I can still enter into a contract with you without ever telling anyone. I can buy something from you, in cash, and nobody needs to know. [...] As long as the regulators are doing their job." Is he not aware that regulators in many places work towards making that impossible? What are examples of good, helpful regulations, does he cite any? Or is it just a vague feeling that regulation is good, and governments are goo…

Its sort of a given to learned people that regulations are good. The fact you don't live in a factory town bartering cigarettes for Amazon Coin you use to buy your meals with is a testament to this. Or that your company isn't forcing you to work at gunpoint. The US allowed corporations to run free during industrialization for just 60 years and they managed to exploit people to such an extent that even the hands-off US government stepped in and started pushing regulations. When able to relentlessly pursue profits, companies will stop at no length to increase profits, and we have seen this all across the world. I believe your partisanship is clouding your judgment.

The author is also clearly not a leftist based on the contents of the article, and his allusions to centrally planned societies (aka their understanding of communism) as a fool's errand.

Re: 100 years of whatever this will be

#200

Post by lordnacho » Thu, Dec 02, 2021, 5:06 PM UTC

You know what makes a distributed system work? Failure.

Now and again, nature arranges for a catastrophe. An asteroid, too much oxygen, lignin building up, nature has cleaned up after and built on top of the rubble.

Over the last 20-odd years, we didn't let a real economic catastrophe happen. There are millions of people working the wrong job. Loads of people have a made a ton of money providing nothing useful at all. Without a periodic shakeup, the bullshit merchants take over the economy. People who talk a good game but are never held to account. Who is swimming naked? There is too much money to know.

Part of the reason is centralization. Unfortunately, we don't really have 200-odd governments. A lot of them think the same way about the issues that matter, and a very small group influence all the others.