Let’s Destroy Bitcoin | Hacker News Reader

Let’s Destroy Bitcoin

medium.com

44 pointsnoch43 comments

Better link (no sign-in): https://www.technologyreview.com/2018/04/24/143425/lets-dest...

The article presents some long term and optemistic possibilities of making bitcoin irrelevant but I'm not that convinced.

I'm by no means a bitcoin maximalist but the trust and network effect of bitcoin is something every other cryptocurrency struggles with and difficult to replicate.

I believe a government run blockchain is a likely outcome, but this doesn't share many similarities with Bitcoin. The currently common rhetoric is that Bitcoin's 21 million limit is a hedge against inflation and a government blockchain won't satisfy people looking for this.

The Facebook sneak attack it proposes has two avenues - it's own token (and we saw how that went) or taking over. The possibility of an more insidious takeover is already a possibility now that vendors like PayPal allow you to buy cryptocurrency. PayPal's version of Bitcoin is already nothing like Bitcoin - only PayPal can hold it for you, you cannot sent it to others or spend it and others cannot send you more. The problem is that even if PayPal or Facebook could capture a large user market that doesn't give miners much incentive to switch to their fork. When they forked it would be world news and I'm sure they'll have lawsuits and people demanding their private keys.

I find the third option the most likely, but who knows how long it will take and if it will happen. As a new digital currency bitcoin has had the fairest distribution, currently the greatest hashrate and if you include wBTC and it's variants it still finalises a lot of value. For an already complex idea it has name recognition and is relatively simple and understandable in design. It's hard cap of 21 million and stage in the distribution cycle can only be replicated in time. Even now by many metrics Ethereum has an advantage over Bitcoin but as the article point out is only a third of its market cap. Even if Bitcoin's market dominance dropped in the short term it's unlikely that the distribution would favour a single project.

To conclude I'm not convinced that Bitcoin's stop of greatest cryptocurrency market cap will be replaced in the short to medium term of 10-20 years.

"As a new digital currency bitcoin has had the fairest distribution,"

Not sure I agree with this, but I guess it depends what your definition of fair is. I think something like 2%-5% of BTC wallets control like 90% of BTC.

It's even worse. According to [1], around 2.18% of addresses control around 94.75% of BTC.

[1] https://bitinfocharts.com/top-100-richest-bitcoin-addresses....

that includes exchange addresses though. each exchange address can hold BTC for millions of people

You're right, it includes exchange addresses. Although my stance is, other people's address == other people's BTC. They might temporarily be a custodian for you but sooner or later something bad will happen.

With half of all bitcoin emitted in just the first 4 years of existence, it's not very evenly distributed. Yet most other cryptocurrencies have an even worse distribution, especially those with premines.

I think in the long-term the first scenario will happen more so than the third, and the second, even if it does, will just be how Bitcoin is treated now. It's unlikely national governments will allow an unregulated digital currency to not be under their control, both from regulatory and economic angles.

This market cycle we've already seen countries place more regulations on cryptocurrencies as an asset class yet Bitcoin's market cap continues to grow.

People that follow the narritive of bitcoin as a hedge against inflation will still tend towards bitcoin.

(2018) -- so bits of this already seem dated, for instance the idea of a profusion of coins being a feature, not a bug, of the crypto space:

> “Think of this as an incredibly efficient barter system,” says Harvey. “Barter is generally inefficient, but if you have a network and you tokenize the goods and services and enable it with a blockchain, it can become very efficient.”

Maybe I just lack imagination, but I am missing where the benefit of this lies. When Apple has a coin and Toyota has a coin and Kodak has a coin (as seemed would be the case at the time this was published) why is it any better to hold any one of these, barter through a set of intermediaries to get to ConvenienceStoreCoin, and settle your candy purchase with that, versus holding investments, selling them for cash, and using that everywhere? Even putting transaction costs aside, my sense is that a natural, shared unit of account / means of exchange will end up predominating and this "incredibly efficient barter system" won't be the paradigm shift the article implies.

> Maybe I just lack imagination, but I am missing where the benefit of this lies. When Apple has a coin and Toyota has a coin and Kodak has a coin (as seemed would be the case at the time this was published) why is it any better to hold any one of these, barter through a set of intermediaries to get to ConvenienceStoreCoin, and settle your candy purchase with that, versus holding investments, selling them for cash, and using that everywhere?

These corporate coins, are a form of company scrip.

" In the United States, mining and logging camps were typically created, owned and operated by a single company. These locations, some quite remote, were often cash poor; even in ones that were not, workers paid in scrip had little choice but to purchase goods at a company store, as exchange into currency, if even available, would exhaust some of the value via the exchange fee.

With this economic monopoly, the employer could place large markups on goods, making workers dependent on the company, thus enforcing employee "loyalty". "

https://en.m.wikipedia.org/wiki/Company_scrip

You are correct. A good definition of money is "the most liquid good." People want to use the single most liquid good in transactions, for ease of transaction.

In other words, good money drives out the bad.

While I hold a modest amount of Bitcoin, it’s pretty clear that it’s just digital gold at this point, and not really a currency in a practical sense. And the mining situation is ridiculous, both from a power usage perspective and a miner centralization perspective.

I’ve been digging into this project lately and it’s pretty exciting: https://www.chia.net especially considering Bram Cohen is the brain behind it (the BitTorrent guy).

Really good real time community on Keybase, I also launched a forum that I’m trying to get some traction with to preserve knowledge for the project at https://chiaforum.com

It’s a lot of fun building and running the chia farming rigs, feels way more accessible than the mining stuff. Also, since it’s barely just launched, and it’s on good tech foundation, it’s exciting to be in “early” on a new coin.

Yeah, I think a lot of altcoins try to create the 'early in' vibe. Not sure how many and which of them avoid becoming irelevant in the long term.

Perhaps one that avoids the 'early in' vibe by having an even distribution over time.

My biggest concern with Chia, is that Proof of Space doesn't actually "use up" any space. In particular, this means that when a "miner" is faced with a choice of two competing blocks (or transactions) their best option is to simply "mine" on both of them. This is doable because, by it's very nature, proof of space has little to no opportunity cost.

As a result, it seems that proof of space doesn't really come to any consensus. In particular the Chia software will track up to 5 simultaneous chain tips as a consequence of this.

P.S. My understanding could be a few years out of date, so do correct me if I'm wrong.

Is this something I could get into with 2TB disks? Say 12-15 2TB disks. I have a bunch that are too old to trust and are sitting on a shelf. Any idea how disk failures affect things?

You need a temporary space of 312GiB to "plot" which is RAM intensive(~4GB min, more if you parallelize) and disk I/O bottlenecked(serious Chia farmers put their temp on an SSD), but finished plots are 101.4GiB each and are just files accessed occasionally, so you can fit a dozen or so on a 2TB. If the drive goes, so do the plots, but the cost is just the time to replot so you probably don't need a backup strategy.

Plotting tends to be a bottleneck - hours to finish one in my experience, though that is with using the storage drive for the temp space. But even with 26 plots(2.5 TiB) I have already earned 4 Chia in a few weeks.

Yes, you could do that. 24-30TB of plots would win a block roughly every day or so with the current netspace.

I'm using a bunch of used drives and not worried about disk failure because worst case scenario is replacing the drive and re-generating new plots.

My only complaint about Bitcoin is the energy use. And I think it would be reasonable to whitelist things that are allowed to use so much electricity (or make their prices rise as a function of use)

So you are ok with the low tps, the lack of programmability, the lack of anonymity, the high fees, the slow transactions, the centralized mining, the tainted community and the lack of development?

The thing is, Bitcoin doesn't have to use a ridiculous amount of energy. You could theoretically run the entire network from a single miner running on a 10 year old Android.

The problem is that the proof-of-work system creates an arms race and everyone starts building massive mining farms in order to get a bigger piece of the block reward pie. Some proponents will claim it's necessary for decentralization, and to a point, it's true, but we definitely don't need the amount of mining power in use today.

Is there yet good pruning solution for bitcoin? That is all of the old transactions are discarded... As size of chain itself is and will be an issue.

I wonder actually how many addresses with coins there are currently...

I know some localities turned to usage based rates, so a normal household (or maybe a normal house times two or three) was at a regular rate and then there was a sharp jump for usage after that (averaged over the month). Somehow they handled factories, etc. in a different system, but they wanted to only subsidize electricity for certain uses.

It wouldn’t be an issue if there were no externalities from generating and using electricity, right?

Yeah... but there are

Indeed, and those particular externalities are the same for each unit of electricity used (given some generation method), regardless of what the electricity is being used for. To me it makes a lot more sense for government policy to address these externalities directly, instead of trying to determine for each specific use case if it's okay to impose externalities or not.

I don’t think it’s necessary to think about all use cases... just those that use lots and lots of energy.

I’m not sure how that would work in practice. Does every use case get hit with a permanent ban the instant the combined electricity usage of that use case exceeds some limit? Was it okay for me to play around with the Bitcoin client on my laptop in 2011, but as soon as all Bitcoin clients combined to exceeded the limit, I can no longer run the client on my laptop? That strikes me as exceedingly odd and impractical.

I was envisioning if you use x% of the grid’s power, you get a visit from the epa

Edit: you’re right that this would not catch really decentralized mining. Which is definitely a hole. In those cases, it would be hard to regulate I agree

How much is trust worth?

Why was trust worth 12.5 BTC every 10 minutes at the beginning of May 11 2020, but only worth 6.25 BTC every 10 minutes by the end of the day?

How did trust get instantly cheaper without causing a disaster? Was Bitcoin paying too much for trust?

The hashrate decreased 25% over the few days on/around May 11 2020. So yes, the price paid for trust decreased 50% and in response the overall security of bitcoin dropped ~25% as a result.

But the security is still good enough, right?

As long as the mining expenditure exceeds the potential illicit rewards of a double-spend, it's not profitable to out-mine the network just to be able to double-spend. So the security of bitcoin should be an all-or-nothing thing: either a 51% attack is profitable, or it's not.

It seems to me if there was some way to determine the potential real world reward of a double-spend, Bitcoin could spend exactly the right amount on security: that number + 1. I don't know how to calculate that number.

I doubt that Bitcoin mining is matching it precisely because how would it? So it is probably overpaying or underpaying by some amount. So far, Bitcoin appears to have overpaid for security, given that it hasn't been 51%ed. Who knows by how much. Maybe after the next halving, or the one after that, it'll be underpaying?

It's really hard to price security investment because the primary signal that you've spent too little is often a very costly hack.

Flashloans and other derivatives greatly increase the risk since they now enable extreme leverage on only a small amount of initial capital.

Not related to my original point, but I feel like btc just substitutes trust in an existing banking system with trust in (your wallet provider, the exchange, your computer’s security, and the stability of btc’s price).

It’s not actually trust if you have to pay for it by destroying the planet.

How do you think the military industrial complex works to back the dollar?

Not by burning vast amounts of energy to do meaningless calculations whose results are discarded.

Correct. It's burning even more vast amounts of energy to do meaningless things that aren't calculations and demonstrably harm the world, especially oil-rich, poor countries.

Not definitive but at least the debt wiped out in the GFC including what is called "the lost decade". So at least quite a bit.

The most probable decline of bitcoin imo is regulation or societal collapse from ecosystem loss from climate change. Both are a 10-30year timeframe kind of thing imo.

lets

Reply on news.ycombinator.com