Most modern RNGs should be faster than memory bandwidth (when optimized), so unless your list is small enough to fit in cache, its unclear if this is faster?
Why should they? If demand for X goes up, and price goes up, the whole point of the price system is to incentivize all users of X to use less and producers of X to make more.
Here's an example where the benefit is obvious:
Say I set up a desalination plant for a city with water shortages. I use a lot of electricity; this drives up electricity prices. But I also sell the more scarce water, lowering the price of water. If I make more selling the water, then to first order I'm helping people overall- they value the water more than the electricity. Perhaps I put other, less efficient desalination plants out of use, or simply other uses of electricity that people value less than the water I provide.
Yeah, the issue with your example is that it's not a desalination plant serving local customers, but rather the entire world via a network. So rural customers have to pay for negative externalities that benefit others geographically distant from them. The benefit is not purely localized. Of course data centers bring jobs etc. I don't think we should ban them... but community level politics exist for a reason -- those that live in a geography should control the land surrounding them.
There isn't an externality here in the traditional sense. No one has a right to purchase anything at any particular price. Should people be disallowed from bidding in auctions because it increases the price of the item in question?
No, people who live in geography absolutely should not control the land surrounding them. That's effectively land hoarding, and incentivizes deeply counterproductive NIMBYism. People should primarily control the land they own.
Because utilities are regulated monopolies that exist to benefit society. The companies building data-centers are well aware that the system isn't setup to handle a customer that suddenly doubles or triples the electric demand of a whole town over night. They are taking advantage of that until someone steps in.
Electricity production isn't a free market in the vast majority of municipalities. Market systems only work when demand and supply are relatively unconstrained. In this case, supply is strongly constrained.
Although this is a neat idea, the author appears to be confused about how generative modelling works.
They repeatedly claim that previous approaches rely on factorization to reduce things to guessable chunks to avoid the 'blur problem'. This is a misunderstanding. Previous approaches solve this problem by modelling a distribution as output rather than a point. Factorization is one way of representing the distribution, but the key point is that even for the small chunks we predict an output distribution, not a point estimate.
Factored models have no issue producing TV static as an image. Each individual guess is impossible, and yet they don't produce a constant grey averaged image.
Normalizing flows, though not considered particularly efficient, are not factored at all, do not proceed in small steps, and yet still have no issues with blur because they predict a distribution.
Their approach appears to be a hard version of a latent variable model. It may be a good idea, but it isn't a fundamental change in the way they suggest.
Can you find true monopolies in the free market, in the last 100 years, that have lasted for at least 20 years, without some kind of government backing/suppression of competitors? I only know of one.
Things only count as monopolies if they have monopoly pricing power; Google is not a monopoly, they cannot even charge 1 cent per search.
In practice, despite economies of scale, large companies age and become bloated, inflexible, etc. So there's typically plenty of room for new companies to compete.
If a monopolist buys all upstarts, then that creates an even stronger incentive to make competing businesses! You don't even have to succeed in the market, you just have to get bought up! The more they buy out, there stronger incentive there will be to enter the market. No one can do this indefinitely- except with some kind of government support- say a regulation that makes it difficult to enter the market.
Your definition of monopoly isn't what people are commonly talking about when they say "monopoly".
Granted, you have the correct definition, just not the colloquial one.
What people say "monopoly" what they are really talking about is industries captured by few big players. For example, grocery stores. There's not a true monopoly there but there is damage done by the fact that there's very little competition in that space. Both to consumers and product manufacturers. These big grocery chains use their leverage to get prices smaller chains can't compete with. Further, they have been trying to consolidate (see Kroger Albertsons merger) to further exert anti-trust behaviors.
A clear example of the damage done by this sort of hyper consolidation is what's happened to the memory market. Multiple times the 3 big players have been hit with anti-trust lawsuits because of the price fixing they've engaged in. It's true they are not true monopolies, but they exert exactly the problematic behavior of a true monopoly. It's very unlikely that the free market would ever fix this.
Groceries are an excellent example of competition. Profit margins on groceries are razor thin. The large chain leverage results in lower consumer prices. Large grocery chains don't have anomalously high profits, ROI, or really anything that would indicate harm to consumers. Groceries are cheaper than ever compared to median wages.
The memory market is international, its not clear what anti trust even means here. Even so, look at the price of memory over time, say the last 20-30 years. Adjust for inflation and its even more extreme. Sure, it is up recently with AI demand, but the general trend is drastically reducing cost and increasing quality. Memory technologies are evolving quickly and it's a capital-intensive business, so its not surprising that there are a few leaders. There's no guarantee of instant competition in free markets.
Cartels rarely last long without enforcement. There's such a strong incentive to defect and secretly sell more under the table. Even OPEC could not maintain one. Memory right now is expensive simply because demand outstrips supply and there is significant capital investment and lead time to increasing supply.
> Large grocery chains don't have anomalously high profits
Define "anomalously" What do you mean by this? You'd admit that large grocery chains do have high profits right? Even with "razor thin" margins.
Large grocery outlets do love to claim that their margins are razor thin yet somehow each of them are some of the wealthiest in the nation. How can that be if they are so good for the customers?
> its not clear what anti trust even means here.
Price fixing [1]
> There's such a strong incentive to defect and secretly sell more under the table. Even OPEC could not maintain one.
OPEC is the exception, not the rule. And enforcement is a lot easier when there are fewer players. The reason OPEC has a hard time enforcing it's prices is because there are are too many players. Were they standard oil, they'd almost certainly be able to maintain their prices and discipline.
Cartels very rarely dissolve on their own. The free market naturally tends towards consolidation and cartels forming.
Grocery store profits aren't larger than any other industry. Nor are they larger than in the past. Large grocery store chains do not have high profit margins. Sure, the total profit may be large if you run 1000 grocery stores.
What do you mean by they are the wealthiest in the nation? None of the richest people in the world built their fortunes on groceries?
Price fixing is a phenomenon. Anti trust is a policy agenda. What entity would carry out the proposed agenda, when the parties supposedly fixing are in different countries?
How long did the fixing agreement last successfuly? DRAM has famously gone through booms and busts where supply outstripped demand and led to very cheap prices and no profits.
OPEC worked about as well as any cartel. Could you give an example of a cartel maintaining their position without heavy government help? Even the drug cartels in mexico, with extreme willingness to coerce, cannot maintain a monopoly and have competed drug prices in the US down heavily! If you consider memory a cartel, then it has only remained so by... drastically improving the quality and reducing the cost of its product? Hardly a problem.
> None of the richest people in the world built their fortunes on groceries?
The Waltons (Walmart) aren't some of the richest people in the world?
> What entity would carry out the proposed agenda, when the parties supposedly fixing are in different countries?
Few can, to be fair. The US and the EU are some of the few that can due to the size of the markets underneath them. A company getting excluded from their markets is enough for them not to want to risk anti-trust enforcement.
China can to some extent as well, which is why you see the likes of Hollywood bending over backwards to meet Chinese standards in media. It's a large enough market that they don't want to miss out.
> How long did the fixing agreement last successfuly?
I see no reason they wouldn't have gone on indefinitely. The only thing that ultimately ended them and caused the busts was antitrust enforcement.
> Could you give an example of a cartel maintaining their position without heavy government help?
Modern examples are few and far between, and that's because the historic examples failed to use their money to buy off enough government officials to cement their positions.
Standard Oil would be an example of this. They did spend money to avoid being regulated, but they didn't ultimately use that to push policies and laws to cement their position in the market. They simply bought out competitors as they rose up.
Bell Labs/AT&T is another example that didn't use their influence to cement their position. They got their monopoly fairly naturally. The closest you could argue is they got utilities access laws passed by the government.
An even older example would be the East India Company. They became so big that they actually took over land and established themselves as a government. Basically the only reason they fell was because of an armed rebellion.
> You'd admit that large grocery chains do have high profits right? Even with "razor thin" margins. Large grocery outlets do love to claim that their margins are razor thin yet somehow each of them are some of the wealthiest in the nation. How can that be if they are so good for the customers?
"First, consider margins. These are often wafer-thin. American supermarkets like Kroger and Albertsons collected just 1% of their sales in operating profits last year. Even Walmart, with its 4% margin, hardly looks like it is milking customers. It is a similar story in Britain and France, with margins only a little fatter in Australia and Canada. The real cream has traditionally been elsewhere in the supply chain, such as with the big food manufacturers. Over the past decade the average operating margin across General Mills, Kraft Heinz, Mondelez and PepsiCo has been 13%."
Do you know what an operating margin is? If you don't you can easily find out. Your comment is wrong in its entirety and you could have very easily discovered this before subjecting others to it.
"yet somehow each of them are some of the wealthiest in the nation."
Who even knows what on earth you meant by this. Some of the wealthiest what, businesses? That's something certain large grocers and some other businesses get with scale - it doesn't mean a business necessarily has 'high profits,' which was your entire thesis.
Looking at your comment history, you have a tendency to state things with a high degree of confidence when you often know very little about them. You should strive to do better - perhaps research and read more and type less.
I'd say exactly the same about antitrust. Can you give an example of a lasting monopoly not backed by some government policy? At any point in the last 150 years?
Sure. The most straightforward example would be utilities in much of the US. Companies like PG&E are literally granted a monopoly in a region.
Another example would be epipens. There is a competitor, but unlike most drugs, if the doctor prescribes you an epipen, the pharmacy can't give you a generic substitute, it must be name brand.
Tesla is being heavily protected from Chinese competitors by US auto tariffs.
Credit Ratings agencies (in large part responsible for 2008 crash) have direct government granted status.
Taxi medallions, Jones act, sugar tariffs.
These are the ones that come to mind right away, but there's a whole slew of stuff- any regulation that's easier for a big business to comply with than a small competitor, for example. Even worse if the big business manages to influence the regulation to enhance this effect. Personally I'd guess Boeing is an example of this; yes, aviation does have high capital costs but I suspect that FAA regulation + Boeing cozy relationship makes it much harder for new entrants. Of course, this isn't to say the FAA regulation isn't helpful or doesn't improve air safety, just that it also has this effect.
If you wanted to do the most to improve competitiveness possible, by far the most effective strategy would be free trade. Trade barriers prevent competition from overseas, making monopolies much easier to maintain.
Sometimes yes, sometimes no. Depends on the barriers to entry as an example. An effective anti-monopoly strategy needs different tools at different times.
Free trade has little impact on monopolies. The biggest corporations in the US are international and the massive amount of free trade we've had from Reagan to Trump's second term has only resulted in these companies becoming even more consolidated.
What free trade enables is for monopolistic corporations to take control of foreign resources to circumvent and undermine US labor laws. They'll buy the rights to the same crops growing in Mexico or Candida for cheaper than the US to undercut US farmers. None of that changes the price they are selling goods in the US and in fact ultimately results in prices going up in both Mexico and Candida.
You claim that a law that prevents or hinders you from making transactions with certain parties, while privileging other parties by exempting them from these restrictions, doesn't make it easier for the privileged parties to form a monopoly?
No, I claim that we spend 3 decades of free trade and we didn't see corporations get smaller. Republicans and Democrats had a free trade policy since Carter really.
If it were true that free trade results in more competition, then you'd expect that there would be more companies competing, not fewer. We see fewer companies.
Your theory has been tested in reality and it fails.
Free trade isn't the issue, the issue is monopolies. When there were 60 or 80 companies all in the same industry but in different states or counties, the economy was much healthier because those 80 companies had to have their own independent staff for functions like accounting. And they also had to compete with each other at the edges of their jurisdictions. None of them could become powerful enough to set policy across the nation and if a corporate policy was beneficial to the consumer, one of their competitors could enact it to compete. Now that there are only 2 or 3 nationwide providers of most things, those companies can just set their own policies whether hostile or not and the only control we have is regulation.
If you want to eliminate regulation and promote free trade then the best thing you can do is oppose the formation of monopolies and encourage enforcement of anti-trust everywhere.
There are many other factors affecting the size of corporations. For example, a significant increase in the amount of regulation in other sectors of the economy.
More importantly, the raw number of corporations isn't how you measure a monopoly. Profit margins are the key sign of monopoly pricing- but in practice, corporate profit margins have been relatively stable in percentage terms over time. Prices for virtually all consumer goods are way down in real terms over the last 30 years.
In particular, trade might get you more competition with fewer corporations, but via competition with overseas producers. These might show up as a few importing corporations, representing a whole slew of competitors abroad.
