Live data from Hacker News

In defense of cryptocurrency

blog.cryptographyengineering.com

291–300 of 578 posts

Re: In defense of cryptocurrency

#291
post #248

Earlier quoted context omitted.

Your response is so full of gaping holes it's kinda crazy. I'm not sure I can even begin to address them. > I think you've overlooked this in the grandparent, because you've probably never had the experience of going to a low-tech county property registrar's office to pull a record. No, I haven't overlooked this at all, in fact it's bang on with what I'm trying to get you to understand - that pulling a record can be…

You are vastly over-estimating the scale at which local government operates and I'm not saying this is 100% about signing. Searching, pulling GIS files, whatever, it's all essentially the same problem. Your argument is akin to being anti-Twitter because the local government could build a notification system just as well. Maybe San Francisco could, but most of America cannot. Twitter's core function is that it lets yo…

> You are vastly over-estimating the scale at which local government operates

Not really, they could all deploy the same DB solution, we could open source it and containerise it and give it away for free. It could be run as a managed service for a small fee. There's no need for each local government to design their own or even run their own.

Exactly the same as with a blockchain solution.

> Your argument is akin to being anti-Twitter because the local government could build a notification system just as well

My argument is not in the least bit akin to being anti-twitter. My argument is more like saying twitter is possible with multiple different back-ends, conceptually. I'm saying look, your short message service can be implemented in a bunch of different ways - in the back end it could run on a NoSQL server of some sort, or some sort of RDBMS, of which there are many, or we could probably mash something up with SQLite to get off the ground. The important thing that's addressing the need is to build a short message service.

In contrast you're there insisting that only an MSSQL system will enable twitter. But with MSSQL we get archiving for free! And with MSSQL we can just UPDATE a record and it will be changed! Huh, how about that?! Nobody else can do that!

And I'm here scratching my head and saying that's just an implementation detail, sure, it could work, there will be license fees etc, but it's not the only way, and we still need to do the rest of the work.

> If parties agree on data sharing protocols, which is something we've all been talking about forever, then the network gains value.

Yep. And those protocols are still going to have to be designed and deployed by someone, a blockchain solution doesn't magic these into existence any more than any other solution does. APIs and UIs will still have to be built. Data schemas will still have to be defined. It doesn't reduce engineering effort.

> He's literally talking about ownership rights tokenized for convenient settlement with a governance layer that can be overridden by officials.

Sure, and maybe it would work great, but that doesn't require a blockchain either, it doesn't need to be an NFT on a public blockchain to provide that low-friction experience.

Throwing 'blockchain' at a massively complex situation doesn't solve anything in and of itself. It seems to me that saying 'blockchain' allows its proponents to mentally skip to the end of the process, where the value actually comes from one of the early or middle steps.

  Step 1: Figure out how to encapsulate land registry records in a simple but comprehensive way
  Step 2: Make legislative changes such that these records are now the source of truth
  Step 3: Expose these records in an easily digestible and transferable way
"Blockchain" is part of one of several solutions to step 3. That's literally all it is in that picture.

(edit/addendum - > If you can't tell, I was a fan of the "semantic web" idea.

I always found that really naive, perhaps I just never understood it, but it seemed to rely on honest actors categorising their content in ways that would allow it to be retrieved, catalogued and processed correctly, then presented to users. Some of the professors at my University, Southampton, were well into it. Even back in the late 90s I was jaded enough to know that anything relying on people to be honest and not use all sorts of SEO tricks to game the system was doomed. I know there was more to it than that, and some of its ideas did filter into the modern web in various ways.)

Re: In defense of cryptocurrency

#292
post #248

Earlier quoted context omitted.

Your response is so full of gaping holes it's kinda crazy. I'm not sure I can even begin to address them. > I think you've overlooked this in the grandparent, because you've probably never had the experience of going to a low-tech county property registrar's office to pull a record. No, I haven't overlooked this at all, in fact it's bang on with what I'm trying to get you to understand - that pulling a record can be…

You are vastly over-estimating the scale at which local government operates and I'm not saying this is 100% about signing. Searching, pulling GIS files, whatever, it's all essentially the same problem. Your argument is akin to being anti-Twitter because the local government could build a notification system just as well. Maybe San Francisco could, but most of America cannot. Twitter's core function is that it lets yo…

> The idea is that a universally trusted, always-on database allows you to build new types of applications.

If you think a government bureaucrat or judge will trust a blockchain because "proof of X" when a dispute does happen, you've got some weirdly conflicting beliefs about government. Furthermore, if you think they would trust it more because it's public then they would trust a DB that, say, Price Waterhouse Coopers were running that any county clerk could sign in to with an audited account and put some information in, then again I really don't understand what is your mental picture of the government.

However you slice it, the fact remains that, if there is ever a dispute on the ownership of your property, the government will want to consult an official paper trail of how that ownership came about. The more layers of abstraction you add, the more places your trail could be attacked, the longer it will take to do it, the more expensive experts will have to be hired for the trial and so on.

Permissionless databases also lend themselves well to a kind of DoS attacks, where someone can trivially bury the real data in mountains of low-effort fake data (perhaps with some high-effort fakes thrown in as well). If the county clerk's intern can mint an NFT for each house in the real registry, I can easily mint 100,000,000 fake NFTs for the same houses as well - good luck disputing the ownership then if your house purchase wasn't registered in the paper registry.

A permissioned DB run by a trusted entity (which could easily be some company, even though I hate that this is what usually happens, and would much rather see the government handling internally) that is responsible for auditing who has access to modify the digital registry and so on is the only possible solution for improving the bureaucracy of this. And even then, it will only improve it so much, since you will still need to send documents that prove you are who you say you are, and the other party is who they say they are, and you both agreed to the contract that made this transfer possible. Of course, much of this can also be fixed by using a government-certified digital certificate proving your identity (this is a common service in the EU at least).

Re: In defense of cryptocurrency

#293
post #55

Earlier quoted context omitted.

He says why he cares are the end. That it could be a viable payment system, if you solve the scaling and energy and regulation and a few other things.

It does not sound plausible that all of this engineering, required to make these systems even viable in the real world (and, we hope, preventing it from eventually demanding approximately as much energy as Eastern Europe, despite the weak incentives to actually accomplish that) is the best, fastest way to lower transaction fees for the working class.

It already does lower transaction fees for the working class. Maybe not Bitcoin specifically, but the cryptocurrency industry has spent more than a decade vastly improving on the concept.

The legacy financial system - even just the electronic parts - are meanwhile the product of many decades of development, and it still ain't viable for large swaths of the working class worldwide.

Re: In defense of cryptocurrency

#294

Earlier quoted context omitted.

I do find at least amusing to point out you started this thread with a comment that adamantly stated that "There's nothing unique or special about it [blockchains]" Only to immediately follow it up by conceding your main point > "he PoW system used is one way to prevent Sybil attacks when distributing work among untrusted peers, and that's it. Novel, neat, but useless" Ok, so you shifted the goal post from "there's n…

> I do find at least amusing to point out you started this thread with a comment that adamantly stated that "There's nothing unique or special about it [blockchains]" > Only to immediately follow it up by conceding your main point > "he PoW system used is one way to prevent Sybil attacks when distributing work among untrusted peers, and that's it. Novel, neat, but useless" I don't see any goalpost moving. PoW is not…

> I don't see any goalpost moving. PoW is not blockchain.

Blockchains are exactly 3 things

- A ledger of "things"

- Consensus to agree on "things"

- A state machine that that transitions based on "things"

PoW is how Bitcoin achieves consensus. The brute-force operations that "waste" energy is actually a highly robost leader election algorithm.

So sure, "PoW is not a blockchain" is technically accurate, but a useless statement on its own.

OP states there is nothing unique or special about blockchains, then notes a unique and special fundamental aspect that enables (some) blockchains.

Re: In defense of cryptocurrency

#295

Earlier quoted context omitted.

It wouldn't be better "in every way", though. A blockchain is much more resilient; the equivalent would be to have hundreds or thousands of redundant SQL databases around the world and somehow keep them all in sync. By the time you've done that, you've very likely reinvented a blockchain, at many times the cost of just using an already-existing public blockchain. > monitoring and reversing payment transactions I disa…

>Transaction reversal is indeed outright harmful to honest vendors, and is only really necessary for the legacy systems because the legacy systems have borderline zero protection against dishonest vendors retaining customers' payment information and pulling money from it This isn’t the only form of dishonesty. There’s cases where you initiate an honest transaction, the vendor ships you something you didn’t ask for be…

You can simply opt in to those kinds of setups.

The flipside is there are times I've been unable to pay with cards because the processors deem it to be too large of a risk, despite me being happy risking my own money. Or, see the risks people take running businesses on PayPal - large companies able to reverse transactions / hold your money solves some issues but raises others.

Re: In defense of cryptocurrency

#296
post #59

>Proof-of-stake systems are not perfect: they still lead to some centralization of power, since in this paradigm the rich tend to get richer. What does this mean? Is there a central node in PoS, or are they saying that power is proportional to stake size?

Power is proportional to stake size. This can be (and is) mitigated by penalizing stake pools that get too big relative to the rest of the network, thus encouraging users to stake on other pools.

Re: In defense of cryptocurrency

#297

This is a genuine question. Is there a defense of cryptocurrencies that doesn't follow the pattern of, "yes, that is indeed true, however here is an edge case or counterexample which either doesn't exist yet, or is not in wide use so it's not actually a problem"? Every article or comment defending cryptocurrencies has followed this pattern and it isn't convincing to me. I'd love to have my mind changed, but when ever…

This is also how almost any technology can be dismissed.

Re: In defense of cryptocurrency

#298
post #98

"transaction reversibility is not about the ledger, but rather about the transaction rules that a currency uses. A reversible currency requires that someone anoint this trusted party (or trusted parties) and that they use their powers to freeze/burn/transact currency in ways that are at odds with the recorded owners’ intentions. And indeed, this is a capability that many tokens now possess" I think this is arguing th…

Many real world systems are "nearly trustless". Of course you still need some sort of court system if someone decides to break bad. But in the 99%+ of times you're in the happy path, economic transactions occur based on autonomous rules encoded in software. The analogy I like to use is what's harder to buy a $1 million house or $1 million of Microsoft stock? The former process takes weeks, and dozens of man hours fro…

For me that creates two big questions:

(1) Do we really want high frequency markets in real estate? Is that socially and economically desirable?

(2) Land and property records so far have survived centuries (if not millennia), can we hope that an NFT solution will have the same long-term compatibility?

Re: In defense of cryptocurrency

#299

Earlier quoted context omitted.

The idea is that if you have more tokens, you will reap more reward in PoS. This is because you receive around 5% return on your stake, so if you set up multiple validators you will earn more per year. Note that PoW works exactly like this too, except far worse due to economies of scale. Purchasing mining facilities in bulk will give you a much better deal than buying rigs as a home miner. Whereas in PoS, both the ul…

Your stake is really growing relative to the total supply as fast as a smaller holder's though. Like anything, rich have more opportunities to get richer, but I don't think these problems are specifically egregious in PoS. For chains which have on-chain governance, there's also the possibility that large holders will vote to change distribution in ways that favor them (though this hasn't happened yet), but this is a…

> And there are theoretical solutions to this as well; projects like Bright-ID are working on proof-of-identity / proof-of-humanity.. maybe we'll see a proof-of-stake blockchain with governance tied to identity in the future?

This is what I'm hoping for. As soon as this proves viable, this opens up the door for things like automated international UBI.

Re: In defense of cryptocurrency

#300

Earlier quoted context omitted.

Many real world systems are "nearly trustless". Of course you still need some sort of court system if someone decides to break bad. But in the 99%+ of times you're in the happy path, economic transactions occur based on autonomous rules encoded in software. The analogy I like to use is what's harder to buy a $1 million house or $1 million of Microsoft stock? The former process takes weeks, and dozens of man hours fro…

Tell me you’ve never bought a house without telling me you’ve never bought a house. Buying houses are expensive because: - Realtors are a state-sanctioned monopoly for real estate transactions and suck up a % of the costs. They also provide the necessary service of aggregating information about the market and negotiating on your behalf - Due diligence on the state of the property. This means inspecting that there’s n…

> Realtors are a state-sanctioned monopoly for real estate transactions

This doesn't even make sense. 'Realtors" is plural, and there are many, which is incompatible with monopolies. Realtors have a monopoly on their trade only because anyone competing with them becomes a realtor, by definition.

It's like complaining about the airlines' monopoly in air passenger transport.

Post reply on HN