Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

A faster, more efficient cryptocurrency

news.mit.edu

Post reply on HN
Print view View original

331–340 of 839 posts

Re: A faster, more efficient cryptocurrency

#331

Post by ur-whale »

sbenitoj wrote: post #327
ur-whale wrote: post #319

Earlier quoted context omitted.

>this is why the market settled on gold Gold, which, by the way, has an extraordinarily high energy and environmental cost to extract.

That’s not a bug, it’s a feature — if it were easy to extract (like aluminum), it would have been much more inflationary and thus not a “good” choice for money.

I didn't claim it was a bug. I was trying to make the point that all the tree huggers that scream bloody murder because of Bitcoin mining costs never look at the actual environmental cost of existing systems.

Re: A faster, more efficient cryptocurrency

#332

Post by dorgo »

hanniabu wrote: post #290
sbenitoj wrote: post #252

Earlier quoted context omitted.

Exactly, any supply of money will do, no need to permanently increase the amount in circulation — this is why the market settled on gold (which has an inherently low inflation rate due to the difficulties of mining) and impossible to pass off counterfeit gold at scald.

You need inflation for proper incentives. It's like a tax on the entire network to keep it supported. Without proper incentives, the game theory collapses. The difference here is that inflation is predetermined and set instead of printing money on demand like we do in the traditional financial system.

There are crypto currencies with built in inflation.

Re: A faster, more efficient cryptocurrency

#333

Post by sbenitoj »

ryan_j_naughton wrote: post #311
sbenitoj wrote: post #248

Earlier quoted context omitted.

I 100% disagree, the central planning of money is (1) what causes inflation, and (2) greatly exacerbates (and in many cases creates) financial crises. Any supply of money will do, the idea that money should increase in supply along with population growth (or some other arbitrary metric) otherwise we’ll permanently live in a deflationary disaster is an imagined monster — we live in the here and now, therefore people h…

A low, but positive, inflation rate is of critical importance. > Any supply of money will do. That is just nonsense. If you have a fixed money supply, given that economic growth is happening, would result in a naturally deflationary currency. That is a terrible place to be for the system as a whole as no one has any incentives to spend or invest in anything as money itself will simply gain value over time just sittin…

Prove it, show me the world where we had a stable monetary supply that resulted in people having no incentive to invest in anything — where’s the experiment?

Again, you HAVE to spend money to live (otherwise you die). It’s impossible to live without food and water, most people want to live in a house, most people want entertainment, etc. None of those things can be had without spending money.

Also, if your argument is correct, why don’t people put 100% of their money into the stock market (which reliably goes up over long periods of time) and instead choose to spend it?

Re: A faster, more efficient cryptocurrency

#334

Post by BenoitEssiambre »

sparkie wrote: post #312
BenoitEssiambre wrote: post #249

Earlier quoted context omitted.

You don't have to ask that question. If they really cared about the tech, they would make sure their tokens are not deflationary. Deflationary tokens are not much good as a currency but they are good as a Ponzi scheme.

I think here's your answer: It isn't just about the tech. This is what most "blockchain" enthusiasts get wrong. They think it is just about technology, and thus, they can come up with "a better bitcoin." They're missing that the predetermined monetary policy is the key innovation of Bitcoin, and that by creating a new "blockchain" which prints new money, they are shooting themselves in the foot, because they can't ev…

A predetermined monetary policy is fine but the fact that it is predetermined to be deflationary is a fatal flaw.

I would have much more faith in a cryptocoin's potential as a currency if the predetermined monetary policy was something like Friedman's k-percent rule: https://en.wikipedia.org/wiki/Friedman%27s_k-percent_rule

Of course this would vastly reduce the Ponzi aspect for early owners which might remove most of the appeal.

And I still wouldn't have much faith in it since I doubt being decentralized is enough of a benefit for people to switch away from government currency and normal banking.

Re: A faster, more efficient cryptocurrency

#335

Post by Vervious »

xiphias2 wrote: post #57

As usual assumptions are much stronger than what Bitcoin has. I'm really sorry about the fact that this currency is voted up so much.

Well, in some sense that's a good thing. Bitcoin relies on a synchronous network - an honest majority of online miners must be honest. Algorand's safety assumption is stronger - the honest stake need not be "always online", and the system cannot fork even under worst-case asynchrony.

As a result I certainly don't trust Bitcoin with anything more than pocket change. I would, on the other hand, put money on Algorand. A Bitcoin attacker could conceivably DOS a handful of honest miners/network relays (given that mining seems pretty centralized), sequester a third of the mining network, and mint money.

No matter how much DOS power you have, you cannot mint money on Algorand without a substantial proportion of the stake.

Re: A faster, more efficient cryptocurrency

#336

Post by justicz »

thesausageking wrote: post #322
justicz wrote: post #225

Earlier quoted context omitted.

(I work for Algorand) This is false. The BFT protocol that Algorand uses guarantees that each round may have exactly one block (up to our security assumptions with regard to honest stake); this is one of the protocol’s “safety” guarantees. In the case that a large number of participating nodes crash, the network will refuse to produce a block until a sufficient amount of stake begins participating again. Nodes that f…

Calling the case of the network halting "safety" doesn't help things. To walk through a plausible example: let's say 3 Chinese services control 35% of the stake of Algorand. Without warning or any public announcement, the Chinese government steps in and forces them to turn off their servers. The Algorand network will now stop functioning as it can't reach 2/3rds. Hours go by and no one is able to transact. At this po…

In this case the Chinese government can kill the liveness of the network (preventing it from producing new blocks), but not the safety of the network (they can't fork the chain).

Your question has changed into: "doesn't the system break when a malicious party controls 35% of the stake?"

To which the answer is yes. If a malicious party can control 35% of the stake, that is very bad.

I should add, though, that if you know your machine is going to be taken down, you can mark your account as "offline" via a special transaction, which will allow the network to proceed without you by increasing the probability that others will be chosen to be on a committee.

Re: A faster, more efficient cryptocurrency

#337

Post by Sir_Substance »

semi-extrinsic wrote: post #263
Sir_Substance wrote: post #237

Earlier quoted context omitted.

The key word in this discussion is potential.

Ah, but you don't need crypto to have that potential. Anyone could issue their own currency, and try to convince people to use it for actually buying and selling stuff. But that convincing part is really the key, and crypto very explicitly doesn't solve the problem that businesses will need to trust the issuing entity before they switch to using a new currency. People and companies want to have a central, stable issu…

>Anyone could issue their own currency, and try to convince people to use it for actually buying and selling stuff. But that convincing part is really the key

Absolutely true, see EcoCash. It's a fascinating case study.

>and crypto very explicitly doesn't solve the problem that businesses will need to trust the issuing entity before they switch to using a new currency.

Trust is a relative thing. EcoCash is issued by a mobile phone provider. That should be crazy (who would trust verizon to issue currency?) except that _literally no one_ trusts the Zimbabwean government to issue currency any more, so the phone provider won by dint of a) being in the right place at the right time and b) being relatively more trustworthy.

Actively building trust is something you need to do if you've got a pile of venture capital burning down that you need to pay back before you go broke. However, cryptocurrencies are usually not VC-backed products, and therefore don't have time-bombs strapped to them. Bitcoin is doing extremely well in Venezuela right now as faith in the bolivar drops through the floor. Cryptocurrencies are at this point extremely well position to be conveniently accessible any time people's faith in their national currency drops enough for them to start looking elsewhere. This is a slow burn situation.

>People and companies want to have a central, stable issuing authority they can trust.

I'd love to see a citation on that. People trust a wide array of deities and insist that the deities other people trust don't exist. This is a Homo economicus argument, it seems unlikely that we can quantify what people want and how they trust so easily. My anecdotal observation is that people want to have something that sounds plausible enough that they feel comfortable not researching the details. If people wanted a central stable issuing authority they wouldn't accept fractional reserve banking.

Re: A faster, more efficient cryptocurrency

#338

Post by cherafel »

malloreon wrote: post #16

I have long wondered how cryptocurrency fans would answer the question: ‘Would you be happy if the crypto-utopia you bring up happens in the next 10 years, and all value is stored/transacted through a cryptocurrency, but it was a coin that you do not possess now, nor could you transfer any of you current currencies into it?’ Say tomorrow someone releases the one true coin, but no one notices. All other cryptocurrenci…

I'd answer in the negative here.

Most cryptocurrency fans seem to be optimistic investors looking to join in on the massive returns seen by investors in 2016-2017. Most of them care more for generating 'hype' and seeing an increase in the value of the coin they hold, just so they can dump their holdings later. I don't think they'd be too interested in a coin that inflated in value that they didn't hold (they'd probably instead promote something they did hold). Most cryptocurrency fans do not hold one particular coin, and generally change which coin they advocate for pretty often.

Cryptocurrency communities are generally pretty toxic and non-constructive environments. I've been part of a couple, and I've noticed multiple times a massive decline in the quality of the community (and, sometimes, in development as well) as the cryptocurrency got more popular (and more people joined the community).

Re: A faster, more efficient cryptocurrency

#339

Post by x2f10 »

throwaway415415 wrote: post #235
x2f10 wrote: post #211

Earlier quoted context omitted.

Q: How erosive is it to the image of cryptocurrencies that there's so many coins? It seems that every week there's a new coin that does XYZ better. Often, it has no infrastructure (hardware or software) other than a limited number of exchanges that are used as speculation.

Is that really a good question to ask someone who is pushing the state of the art in this domain?

Yes - isn't that obvious? You need infrastructure to use a 'state of the art' currency as a medium of exchange.

Re: A faster, more efficient cryptocurrency

#340

Post by sbenitoj »

hanniabu wrote: post #318
sbenitoj wrote: post #304

Earlier quoted context omitted.

“You need inflation for proper incentives. It's like a tax on the entire network to keep it supported. Without proper incentives, the game theory collapses.” What, specifically, do you think the consequences of a stable money supply are?

No incentive to support the network if there's no block rewards. The transaction fees are not sufficient and just meant to prevent spam. It's also better to tax the network as a whole (through this set inflation) than to tax for each usage.

I misunderstood you at first.

That’s an interesting speculation, but it remains to be seen what will happen when inflation is near zero.

If people value transacting on the network, then they will pay the necessary fees to keep it running. If they don’t then it will go down.

I suspect this isn’t the big issue people think it is, regardless there’s no way to “prove” that transaction fees won’t be enough — everyone from miners to speculators by owning mining equipment and Bitcoin are specifically betting that transaction fees will be enough.

Post reply on HN