What is that based on? What are the probabilities?


I think the burden is on you to show why it's so important to free up that storage space. It's not a major problem for scaling. Why is it a good trade-off? What will we gain?


A checkpoint makes all transactions prior to the checkpoint irreversible. If that's not what you mean, maybe you should clarify.

I think pruning is great. Anyone who doesn't have the storage space for the entire blockchain should do it. However, since storage space is not a critical bottleneck in Bitcoin scaling and isn't a major threat to node decentralization, I think it would be a net loss to implement it at the consensus level. The tradeoff isn't worth it.

I get what you're saying with checkpoints from a practical view. The problem is, once you give developers that control, it's like Pandora's Box. It might begin an endless political controversy. How much proof of work is enough before Bitcoin is ultimately decided by human consensus? Maybe not two days worth of blocks, but how about a month or a year -- is that enough? Does this open up new selfish mining attack vectors where the attacker would release their chain reorg directly before a checkpoint is hard coded?

I think it's a niche of duped customers who hopes to recover their BTC through a token pump. Maybe it's inspired by Binance's BNB. I think this is the general idea: Leverage the brand (people at least have heard of it) and customer database, pay for a state-of-the-art trading back end since the bankruptcy trust has much more assets than liabilities, and pay victims with a pro rata share of a dividend-paying ICO token that could be traded on other exchanges.

As I was saying, the people promoting it are probably just hoping to sell into a GOXCoin pump and walk away. It's less about launching a working exchange and recovering real money for victims. What are the chances of that happening?

I definitely think so. There's a perfect storm brewing. It's tough to say when full-fledged adoption will happen, but if trust in banks and governments continues to decline while people increasingly embrace technology, I don't see any way around it.

That goes against the entire philosophy of proof of work, though.

The point of proof of work is that it is cumulative -- there is a chain of work proven back to the genesis block. If one block is invalid, every block mined on top of it is invalid. This is a basic mechanism to prevent dishonest mining.

What you're suggesting are called "checkpoints", and then pruning the blockchain below the last checkpoint. The early versions of Bitcoin actually included checkpoints to ensure reliability of confirmations. The use of checkpoints has been phased out -- there hasn't been one coded into the protocol since 2014. This is because of what I mentioned earlier: Checkpoints give developers power over Bitcoin's security model because developers can make blocks irreversible when they otherwise would have been reorganized by miners.

Implementing Bitcoin support absolutely has costs.

They need to dedicate developer and server resources on an ongoing basis. They need to develop and maintain a robust wallet and custody solution, as well as a robust solution for currency conversion. They also need to account for custody risks. Or if they hire a processor like Bitpay, that's a 1% cost right there, which is only a tiny bit better than Visa. I also assume processors like Visa give giants like Amazon favorable fee arrangements as well, so Bitpay may in fact be worse.

I also strongly disagree that it'll "very likely result in a wave of customers they otherwise wouldn't have." What businesses implemented Bitcoin support and then saw massive growth? I remember several cases where businesses added Bitcoin then later dropped it because there was no interest. Adding Bitcoin as a payment option has never driven consumers to Bitcoin.

Besides, Amazon doesn't need to achieve growth that way. They are basically taking over the world by expanding their core logistics model and consolidating neighboring industries. As long as they offer all the major payment options consumers want, this is an afterthought to them at best.


I'm pretty sure that Bitcoin holders intent on buying stuff on Amazon already buy gift cards with bitcoins -- Bitrefill, Bitpay, other gift card malls. If they use it directly on Amazon instead (or on Amazon through Bitpay), I just don't see it making a big difference.

The formula itself was never part of the protocol. That's why miners stopped supporting it once they realized they could increase their revenues with fees.

Core only removed priority from its fee and confirmation estimations because it was making those estimates inaccurate, as miners had completely stopped supporting it.

What miners did had everything to do with the formula: Miners are the ones who made sure transaction priority ceased to exist. For that, you can thank miners for being rational. It had absolutely nothing to do with Core.


You're confused about what "Bitcoin rules" are. Transaction priority was never a consensus rule. It was a client side rule. Not required by the protocol. That's why there was no fork when miners (and later Core) removed it from their clients.

That's interesting. I'm not sure why that never dawned on me. When Bitcoin Cash forked, people started sweeping keys to recover and sell the coins. Unless those sweeps were done extremely carefully, the hard forks probably handed Chainalysis an incredible amount of address clusters and transaction history on a silver platter!

On top of risking malware, people take big privacy risks to recover hard forked coins.

I remember Jameson Lopp (of statoshi.info) said he did a study on this and came to similar numbers as Chainalysis.


It's worth noting that Chainalysis put the estimate at 2.78 million - 3.79 million. "Nearly 4 million" is what made the headlines. They also included the early "Satoshi" coins in those numbers, which they estimate = about 1 million bitcoins.

If we take the low end and table the Satoshi discussion, that's around 10%.

To keep your channels open, you'd want to remain online, or at least come online periodically to make sure no one tries to steal your coins by broadcasting an old channel state. Otherwise you'd want to hire a watchtower service to handle this for you. You can't really "back up" your channels and go offline.

With Lightning, your private keys are always online. It's always safer to keep your coins in cold storage. I would only use a limited amount in payment channels, an amount I were willing to lose -- same as an online desktop or mobile wallet.

Fee priority didn't solve this problem. It was actually just a way for users to pay less fees!

It was miners who took it upon themselves to remove transaction priority because it was node policy, not part of the protocol. Transaction priority was losing them revenue so they stopped enforcing it. It made no sense to keep using priority in Core's fee calculations if no miners were actually using it. There's some good discussion of the issue here, including this bit:

That's just the thing. Bitcoin isn't widely accepted. This is still the early adopter phase.

Amazon would love to accept BTC if customers really wanted it -- it's a merchant's dream. I assume they don't bother because there isn't actually strong demand for it, and if there were, there may not be enough market liquidity to liquidate their coins without losses. It's just an unnecessary risk for them.

That's obvious, though. Wouldn't you rather get something for free rather than paying the actual cost? Neither Bitcoin nor Lightning are perfect, but it's the closest we can get to having our cake and eating it too.

If you keep giving users more block space for free, they'll keep taking it, regardless of the long term costs. Raising the ceiling higher and higher means there is no mechanism to enforce fees. How's that going to work a few halvings down the road when block rewards are amounting to little more than 1.5 BTC? "Free transactions for all" sounds like a nice socialist paradise, but miners aren't in this for charity.

Every one of us has incentive to avoid paying fees. So, why do people like me still oppose bigger blocks and prefer the settlement layer approach? Because I also hold bitcoins, and I know that the block size limit is integral to guaranteeing the security of Bitcoin and thus the value of my coins.

Sounds like sharding, which is planned for Ethereum in the future. Sharding doesn't seem compatible with proof-of-work because the vast majority of nodes won't ever fully validate the blockchain, only the shard to which they belong to. An SPV proof would be possible, and that would provide a decent level of security. Downloading the whole blockchain and then pruning it defeats the purpose, though, so most people would never do it.


That seems to be the biggest risk -- like a torrent with no seeders and only leechers.

The bankruptcy filing states that Cotten is the only one with access to the cold storage wallets. Presumably, there are other wallets the company can still access. They may be consolidating those funds as we speak. The Reddit sleuthing -- while convincing on a preliminary read -- doesn't seem to conclusively identify any cold storage wallets.


I think bringing him home alive would be much more effective at solving the problem.

There's a distinction that the Bitcoin Wiki makes between "full node" and "archival node" which I think is pretty useful:



Archival nodes help the network by bootstrapping full nodes who need to download and validate the entire blockchain. You can prune the chain and even turn listening off and still be a full node -- it just means fully validating.

Lightning uses bidirectional channels -- there are two parties per channel. The capacity limit for each channel is the sum total committed by both parties. If two people commit 0.5 BTC to a channel, the total capacity of the channel is 1 BTC -- that's the total that can be settled when the channel is closed. Neither of you can settle the channel with more BTC than the two of you put in. Any more than that and you'd be creating bitcoins out of thin air. The Bitcoin protocol doesn't like that.

Not quite. Light clients in Ethereum are not capable of fully validating a block by themselves. It's the SPV model -- looking up a transaction and validating headers but not validating entire blocks. What the OP is talking about sounds more like fully validating nodes that prune their blockchain to save storage space.

How do you know? Any evidence or just more empty conjecture?

Both transaction fees and price steadily rose from 2015 until the end of the run in December, 2017. Users were obviously willing to pay higher and higher fees. When Bitcoin has even more utility because of increasing adoption, a robust LN, sidechains, smart contract ecosystem, etc. I think that would justify higher fees yet.


That's the best case scenario, and it's with a base block size of 1MB. Witness data is not a problem for the quadratic sighashing problem, but increasing the base block size is. Some contrast to your "data":

If someone sent coins to both you and the designated addresses, you weren't transacting with the sanctioned individuals and your coins didn't come from them. If you receive coins from the designated addresses, you're obligated to freeze the funds and contact the Treasury Department within 10 days. That's all.

If you've looked into how unlikely address collisions in Bitcoin are, then you know this scenario is ridiculously unlikely. No terrorists are going to troll you by sending real money to your wallet. Worrying about this is ridiculous.