Hacker News new | comments | show | ask | jobs | submit login
Bitcoin is no long the only game in crypto-currency town (economist.com)
66 points by edward 5 days ago | hide | past | web | favorite | 90 comments





Bitcoin 10 minute block generation times is an inherent problem that unless changed, will doom the protocol to its eventual death.

It's like trying to make a 56k modem work, when broadband is available.

When litecoin came around a few years ago these were abstract problems, but now that transactions with $10 fees take 36 hours to confirm... the issues are real and a switch needs to happen. Lightning network or not, these hacks to speed up the network unequivocally make more sense to do on a network with faster block generation times. Necessity will force this outcome.


I know this probably sounds like a stuck record at this point, but Lightning Network can solve this, and will relegate the actual blockchain as an arbitration/settlement layer where the 10 minute blocktime isn't a problem.

LN's security model relies on being able to close a channel before a certain block. This means that there will be periods of time that paying the fee to close a channel in time will be cost prohibitive and the security property evaporates.

That being said we really should see exchange -> exchange txs be conducted over large LN channels, it could off load a lot of pressure off the chain.


There is a method (which is already in LN) to prevent this which will stop the nSequence countdown if blocks are congested. From the LN paper:

>For example, if a parent transaction output is spent by a child with a nSequence value of 10, one must wait 10 confirmations before the transaction becomes valid. However, if the timestop flag has been set, the counting of confirmations stops, even with new blocks. If 6 confirmations have elapsed (4 more are necessary for the transaction to be valid), and the timestop block has been set on the 7th block, that block does not count towards the nSequence requirement of 10 confirmations; the child is still at 6 blocks for the relative confirmation value.

>This gives sufficient time and block-space for transactions at the current auxiliary timestop block height to enter into the blockchain, which can prevent systemic attackers from successfully attacking the system.

It's not perfect by any means (a malicious miner with a significant portion of the hashpower could never set the timestop bit and spam the mempool at the same time), but it mostly removes this worry in a "normal congestion" scenario, and makes it require a pretty large portion of global hashpower as well as very VERY deep pockets to pull it off in a "malicious" scenario.


the main repo doesn't seem to have the string timestop in it. Nor does any of the bips. What BIP was this implemented in?

Seems like a decent idea, although I don't see why miners would be incentivized to set timestop and the paper suggests that timestop is set to 1 whenever the block is full. The blocks are almost always full, so is it just up to the miner to choose if time is stopped or not?


I actually don't think there is a BIP for it yet... So I guess I was mistaken saying it's already part of LN (it's part of the paper, but not of any implementations).

But the idea relies on miners "doing the right thing", and in this case since there isn't any downside for following this rule, the assumption is that miners will act in good faith.

>The blocks are almost always full, so is it just up to the miner to choose if time is stopped or not?

It's not necessarily if the block is full that the bit is set, but rather if there is "network congestion". This can mean many things to many people, but the assumption is that if fees go up significantly over a short period of time, that could indicate congestion which miners would set the bit on, or perhaps if the mempool of pending transactions gets over a certain BTC amount, it could be set.

Also, I completely forgot to mention the other piece that makes all of this much less worrisome. If you need to broadcast any transactions to prevent "theft" of your BTC, you will receive all of the BTC in the channel by doing so. This gives you the ability to make a child-pays-for-parent transaction with the UXTOs from the channel closure (in a contract revocation scenario) with an absurdly high transaction fee to guarantee inclusion in the next block (basically using ALL of the other party's money as the transaction fee).

Technically, the other party in the channel could also do a CPFP to try to get their transaction included before yours, but as long your side of the channel never went to 0 at any point, you will always be able to "outspend" the other party for transaction fees.


10 minute block generation time is not a problem at all. What's a problem is every block being full.

Can you name one good reason for having it be 10 minutes rather than 30 seconds?

If you can't then you are going to lose this argument.


That problem isn't a result of the 10 minute blocks. A 0-confirmation transaction normally is just fine. These are available on Bitcoin Cash.

A 0-confirmation transaction normally is just fine.

It's really not fine at all.

The competition is full confirmation transactions on a debit card or bank account in seconds with much lower or zero fees.


> It's really not fine at all.

Actually, it really is. I've been accepting 0-conf transactions for nearly a year and had negligible double spends. https://incoherency.co.uk/blog/stories/zeroconf-payments.htm...

The small number of double-spends that did occur since I wrote that blog post I think are mostly due to bad luck (and opportunism) rather than malicious intent upfront, and represent only ~0.1% of payments.


Actually, it really is. I've been accepting 0-conf transactions for nearly a year and had negligible double spends.

So your approach to preventing fraud is... to trust people not to defraud you?

Glad it worked for you with bitcoin and the particular people using your service, but that's not really a good basis for a payment network. There are multiple other problems with bitcoin as a payment network and currency, the delay in transactions (which prompted your experiment in ignoring a key part of the currency) is only one of them.


All payment systems have some system of loss. Visa/MC have fraud and error that eats away small percentages and relies largely on trusting people not to defraud you.

If you run Starbucks and serve 1000 transactions per day, you’ll have some fraud. If 0.1% are fraud, that’s not too bad.


This isn't a system of loss (systemic loss?), it's an abdication of responsibility - it's giving up on the payment system having any sort of fraud protection.

You just used the same argument that some parents use when they decide to not vaccinate their children. They essentially rely on the fact that everyone else vaccinates so they don't have to. This trend catches on and now measles are at a 20 year high.

Interesting that a bunch of arm-chair downvoters "know" this won't work, but in practice, it does...

Things that work out fine when only a few people are doing them might not work so out so well when a lot of people are doing them. See: the entire history of Internet security.

Sure but

1-The current solution is worse on this dimension.

2-Since Bitcoin Cash doesn't have replace-by-fee this is impractical for any small transaction.

3-Most of the situations that need to be immediate are small--with the large ones it makes sense to wait.


Debit and credit-card transactions don't confirm for about 120 days, which is when chargebacks become slightly more difficult. Those chargebacks also cost the merchant not only the price, but a $15 chargeback fee.

Chargebacks (i.e. unhappy customers) are a part of any business. Remedies might be business or legal, but you can't just wish them away by excluding them from a payment solution.

I have no idea why you think that's related to time to confirm that a customer has money to pay for goods, which is an entirely unrelated problem which bitcoin solves badly in comparison to current payment networks, and due to its distributed nature is unlikely ever to solve well.


>seconds

At PoS systems in the US, it's typical to have to wait 30+ seconds for chip cards to go through. I'm guessing that this has more to do with the PoS system / slow internet connection for the store / slow network on the back-end card network, but cryptocurrencies would need to be sub-minute to be ~acceptable for in-person retail transactions.

Until that is solved, until cryptocoins are more than speculative securities where the value is only the current market price on a handful of exchanges, and until you can use them to realistically buy and sell things instead of using cash or a credit card, I don't consider them a "real" currency (by my own definition). I DO hope that it gets there some day, but I don't see that happening for at least another 5 years, likely with quite a few people who put a significant amount of money into it getting burned quite badly.


30s isn't really typical; the average is in the ballpark of 10s. When I worked at Square we got our average down to 4.2s [1]. Modern Verifone terminals seem pretty quick as well. Long term, "instant" contactless will probably become the norm (though it's taking hold very slowly in the US).

[1] https://squareup.com/townsquare/weve-chipped-away-at-emv-tra...


I have absolutely never had to wait 30+ seconds for a credit card transaction that used the chip, so I don't agree with your assessment of that amount of time being typical. Generally, for me, I think it takes about 3-5 seconds from the time I confirm the dollar amount to the time the confirmation comes back.

Sure, this varies. If your POS system uses dialup, it can take tens of seconds, if it uses wifi it will be just a few seconds. If it uses contactless it can be sub-second. But in principle it can be as fast as the network and one trusted node.

The fundamental problem with bitcoin is that it is decentralised and trustless, and those two properties preclude fast transactions at a global scale - for that you need a federated system with trust between nodes. If you remove those properties, you're no longer using anything resembling bitcoin, and why not use a system with proper validated parties in the first place?


But even with BitCoin you can submit the transaction and see it unconfirmed on multiple nodes within 5-30s. It may not be confirmed for hours, but it’s certaibly actionable and trustworthy enough for merchants.

The bigger problem is having full blocks forever so it never gets processed.


Such time is possible if terminal is using dial up and establishes connection only when transaction started.

Seen it couple times.


> The competition is full confirmation transactions on a debit card or bank account in seconds with much lower or zero fees.

No. Those do not compete with trustless digital money. If you think that then you don't understand the point of Bitcoin. It's like saying landlines are better than mobiles because you can never misplace them.


I honestly don't know a ton about cryptocurrencies, but I though confirmed transactions were.....more or less whole point? can you give me an ELI5 version of why 0-confirmation transactions are ok?

For small transactions, it's often a tolerable risk. You amortize any losses to double-spends and compare that to e.g CC fees.

For larger amounts, you want confirmations, but then it's generally alright to wait a little while for confirmation, too.

Note that replace-by-fee means this only works with Bitcoin Cash.


It depends on the use-case. If you look at credit card transactions, they are only "confirmed" after 120 days (when chargebacks become harder...not impossible). A 0-conf transaction will work for all scenarios where you would trust a credit card with about the same amount of certainty. For those other scenarios, 10 minutes is better.

I think it's really only comparable for online transactions, where all credit card transactions are Card Not Present and you aren't able to 100% guarantee that the delivery went to the correct person or that the card wasn't stolen, making fighting the chargeback sometimes not worth it. A brick and mortar doing swipe transactions should have everything they need to effectively fight chargebacks, so I wouldn't consider that money to be 'gone' the way it would be in a double-spend.

0-conf isn't fine in many trustless scenarios with Bitcoin because of Replace By Fee.

Agreed, but Bitcoin Cash intentionally has not added Replace By Fee for this reason.

What the hell is up with economist.com proofreading. A grammatical or spelling error in the article title is a red flag to me that screams "I'm a piece of garbage, don't read me".

Time to reestablish quality over quantity.


Haha, I saw the same thing. No way I would pay 20 euro to read that.

I used to play a game when I'd read TechCrunch for whatever misguided reason: I'd read until the first obvious spelling or grammatical error. I don't believe that I ever made it beyond the second paragraph. US journalism today is largely awful clickbaity crap, and the ad revenue encourages a race to the bottom for content quality. It'd be awesome if there were a way to align incentives for this to not be case. So far, nothing I've heard of seems to be working very well aside from running ads and trying to sell subscriptions, sometimes despite low-quality content.

The economist's dirty secret is that it's mostly written by a bunch of cheap graduates.

In case the error gets fixed, will you guys clarify what you are referring to?

Do you want it to say "crypto-currency game in town" instead of "game in crypto-currency town"?


"no long" -> "no longer"

title: "Bitcoin is no long the only game in crypto-currency town"

Let’s see how long they will take to fix

Exactly, if they couldn't even get the title right, what else in the article is going to be incorrect?

Erm... it hasn't been the only crypto in town for a long time now. But I guess these kind of publications have only relatively recently caught up with the fact Bitcoin exists at all.

Bitcoin dominance dropping below 50% was not close to the realm of possibility just 2 years ago. Bitcoin is now only at ~33% of dominance.

I wish people would stop using market cap as if it was meaningful for cryptocurrencies. A lot of these coins are traded at low volumes on unregulated exchanges that allow wash trading and other means of manipulating prices.

Even were that not the case and they had meaningful price discovery mechanisms 'market cap' would be a meaningless metric because if someone were to buy all the coins, they'd have destroyed the value of them because they've ruined the network value of the currency. This isn't the case for stocks in an actual company, because if you buy all the shares of a company with a 10 billion dollar market cap, you now own a company worth approximately 10 billion dollars.


I recently created my own cryptocurrency, UCaetanoCoin (UCC), with a supply of 100 trillion and one coins, all owned by me.

I sold one to a friend of mine for $0.01 yesterday, which made me not only the richest person in the world, but also the first trillionaire, as the coins I own are valued in $1T USD.

And my UCC has a market cap of $1T USD (and one cent).

Then I exchanged a single coin with myself for $0.01. As my coin has zero transaction cost and instant transactions, I traded it back and forth with myself 100 quadrillion times.

This generated a total traded value UCC-USD of 1 Quadrillion Dollars yesterday.

Anyone wants to invest on my UCC?


Show me your white paper :D


You win hacker news today for this double hitter of comments. I love your mental model regarding market cap and implied valuation. Your white paper is more honest then most!

> You win hacker news today

Oh, thanks, although winning hacker news is quite irrelevant when you're a trillionaire.

Now if you excuse me, I'll retire to my orbiting chateau, it's tea time at my geostationary orbital slot.


There needs to be some way of determining the relative size and scale of these projects. Market cap is useful, but it needs to be normalized given the various different distribution models. So something like fully diluted market cap, or market cap using today's prices and the 2025 supply.

what would be a better measure?

I would consider how many people use them for transactions, if they are supposed to be currencies.

Most people look at the price and the market cap because they can tell that the primary use of crypto-currencies is for speculation.


Yeah, maybe we should look at the average daily transaction count and average fee per transaction.

Yield? Oh wait.

Exactly. You could buy assets that produce things people need or you could buy a random string of numbers and letters.

Obviously we are in a post scarcity world and therefore producing something of value is really a net negative. All hail the inert string of chars! It’s better than something because it does nothing.


"Even Facebook has reportedly started looking into creating a token. Should the world’s biggest social network ever make that move, bitcoin’s days as the leading crypto-currency would almost certainly be numbered" > this is a paradox, blockchain technology's best use case is decentralization of data (and power from companies like facebook). And if Facebook issues crypto coin and it starts dominating the market (not necessarily) then we are back at square one.

>decentralization of data

IMO, having many cryptocurrencies above an arbitrarily-large threshold of USD equivalency (let's say $1b, which according to https://coinmarketcap.com/ right now gives us 40) dilutes the overall chance of cryptocurrencies ever becoming viable for commerce, instead positioning them more similarly to stocks or other speculative investments.

To meet my definition of "succeed" as currencies ("when in actual use or circulation as a medium of exchange"), we would likely need just a few, or likely, one, that works e.g. at brick-and-mortar places. Overstock, Newegg, Steam, etc give good credibility to this idea, but look at which cryptocurrencies they accept (and how the price in those is tied to the real-time market rate).

Also regarding decentralization, and again because these cryptocurrencies largely have value right now only because of their order books on a handful of exchanges, are we actually in a better place? If the feds in the US decided that they wanted to regulate / shut down the big exchanges for whatever reason (e.g. billionaire gets burned and cries to his politician buddies to "fix" it), sure the coins don't go away, but as long as you can't realistically spend them on basic needs like food and shelter, are they actually decentralized? To me, cryptocurrencies are the similar to Chuck E. Cheese tokens that vary thousands of percent in value per year; they have some cash equivalency if you're able to convert them, but alone, they have no real utility.


I agree with you, but that's idealistic. See: Ripple (https://ripple.com/)

For all the talk of Ripple being centralized, they aren't by design, and they have a plan in place to reduce and eventually eliminate that centralization (to a point).

They decided that it would be a good idea to start centralized, then over time wane themselves out of the "validation" of transactions until it's in the hands of a diverse group of validator nodes where no one person has the ability to change things.


How about the part where the founders hold enough tokens that they could crash the price or manipulate the market at will. I understand that they have loose structures in place to keep them honest, but they could absolutely still act badly. We trust that they won't.

There are 2 things here, There is the cut that the founders hold, and there is the portion under the control of Ripple labs.

Ripple labs has a very strict way of handling the XRP they hold. It's not just "lose structures" in place, they are using the Ripple blockchain to escrow 55 billion of the XRP and release it on a set schedule of (I believe) 1 billion per month, and any that is left over at the end of the month is re-escrowed. So they can't just "flood the market" at any time, there is a set limit to how much they can even sell without hardforking the blockchain in their favor.

The founders have free reign on their 20 billion XRP, but I don't see why they would ever use it to crash the price.


> The founders have free reign on their 20 billion XRP, but I don't see why they would ever use it to crash the price.

Profit motive?


Why kill the goose that lays the egg? I mean, if XRP is on its way out then sure, they'll add fuel to the fire, but crashing the price is counterproductive. If they really wanted to cash out and they were smart, they would drip it over time so as not to eat up the market orders on the buy side.

There is important work being done to create a general framework for valuing cryptoassets. We might consider a few key concepts, total addressable market (TAM), percent penetration of that market, velocity, and number of coins outstanding. It's essential to understand the token velocity, and the degree by which a network is effectively capturing some of the value it delivers to its users/token holders.

Chris Burniske, a crypto fund manager writes "these aren’t companies; they don’t have cash flows. Hence, using a discounted cash flow (DCF) analysis is not suitable. Instead, valuing cryptoassets requires setting up models structurally similar to what a DCF would look like, with a projection for each year, but instead of revenues, margins and profits, the equation of exchange is used to derive each year’s current utility value (CUV). Then, since markets price assets based on future expectations, one must discount a future utility value back to the present to derive a rational market price for any given year".

“Cryptoasset Valuations” @cburniske https://medium.com/@cburniske/cryptoasset-valuations-ac83479...


I recommend you check out https://raiblocks.net if you're looking for a currency that could compete with Bitcoin as a means of payment. The tech is smart as hell.

No one ever mentions the insane amount of time it takes to sync their chain.. I'd say that is a big stopping point for a lot of people.. I've been trying for weeks to catch up. (Where other wallets sync in a reasonable time)

True. From what I read, they are working on a way to fix that. It is a major issue at the moment.

The surrounding developer community is getting more and more active. I recently joined the #development chatter on Discord(https://chat.raiblocks.net)

Feeless, near-instant transactions could displace a lot of the coins out there, if they can solve all attack vectors.


And who cares? Nobody is accepting raiblocks as a payment. There are more engineers working on Bitcoin and far more infrastructure around Bitcoin. All these altcoins are pump and dump schemes.

Network effect is real, but that can not defend bitcoin if tx continue to stay prohibitively high.

Nobody was accepting Bitcoin as a payment in 2009. I wish I'd cared about it more then.

This is what most people don't get, these technologies are all open source. So even if some random coin comes up with a great new feature, bitcoin can just add it or at worst someone can create a fork of bitcoin with the new tech. The value is not the tech, its the network.

Bitcoin has good liquidity and been tested though many drops and spikes. Each time it makes the network more antifragile. No one wants to trust a new currency where a large percent of holders have never been tested and might sell off their entire share at any time.


> bitcoin can just add it

Bitcoin is pretty much impossible to change now. It's just too big for that.

Example: Segwit voting was available in October 2016 but didn't manage to get significant miner support, finally got activated in August 2017 after a lot of controversy, and is still not seeing convincing adoption[1]. It seems to be stuck around 10%. If a soft fork is that difficult, then more drastic changes must be impossible.

[1] http://segwit.party/charts/#


The project is still early stage, but still it promises to be what bitcoin should have been from the start : a decentralized, feeless and instantaneous means of transferring value.

What value?

Very odd to recommend Raiblocks instead of at least Ethereum and Bitcoin Cash, which have 300x and 90x the volume respectively. They both also have low fees and near instant transactions, already work for their use cases, and can be bought with USD.

There are real competitors to Bitcoin Core today, and Raiblocks is not one of them. Your value proposition cannot be "like Ethereum but fractionally better because volume is two orders of magnitude less". There are even interesting coins like Aragon, Po.et, SALT, Aeron, district0x that actually do things Bitcoin doesn't do.

Competing for "payments on the internet" is a losing battle, of which you can find hundreds of other pumps on coinmarketcap. They all have nice landing pages about how the blockchain is the future and this is the new way for money and banking. That's a clear sign there's no future in a coin, because the founders can't come up with anything but "let's make Bitcoin faster".

Obligatory disclaimer that I don't invest in any of these things. I'd rather have people invest in curing disease, space travel, longevity research; things that affect the physical world and push humanity forwards. "Like Visa but fuck Visa" is not a compelling pitch.


Unproven tech. There's nothing but a whitepaper to analyze its claims.

My opinions:

There are not meaningful differences between different proof of work coins. All PoW is based on the same principles, has the same security costs, and the same scaling challenges. Bitcoin remains the top dog and will likely stay that way due to network effect.

PoS could be interesting, but it's largely untested in the way that Bitcoin is. We need to see how mineres/voters behave and whether large holders can exert undue influence. Ethereum is by far my favored coin to see this tested, and if successful, would actually have more security for a given network cost. This is potentially disruptive.

DAG-coins or whatever you'd like to call them seem to like they're getting 'free wins' and I suspect that their security model is not sound. They're interesting but extremely risky.

Ripple is simply not a cryptocurrency by any reasonable definition.


Good point, Ripple could only be a cryptocurrency if you define that as being "a digital asset designed to work as a medium of exchange that uses cryptography to secure its transactions, to control the creation of additional units, and to verify the transfer of assets" or some other crazy garbage.

Bitcoin is also one of the least interesting when it comes to innovation. Bitcoin has one of the worst max tx rate.

What bitcoin has is the network effect and security, but if the friction from tx fees becomes too high it could be dethroned.


Craigslist is no longer the only game in classifieds town.

I wish we could stop posting articles under a paywall

What you should actually wish it that payment to such publishers would be so easy that you wouldn't need to think about it.

If only there were some kind of internet-based money system that could integrate with these features...

As long as "easy" also means "cheap."

simply disable javascript

You might want to give https://github.com/nextgens/anti-paywall a shot.

Why?

no paywall for me.



Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | DMCA | Apply to YC | Contact

Search: