That is for adding trading support. Any forks which have worked out the replay protection problem - which includes BCH - are trivial to add withdraw support for, and many exchanges have offered this from day one (demonstrating they're likely above board as far as handling users' funds go).

Coinbase and others saying they'll "wait and see" is really just another bump on a long road of screwing users out of funds. They once again profit off of later price information - and in this case can even decide to argue that they held the assets and so own the BCH - essentially just earning another piece of their users' pies.

I'm amazed people still stick with Coinbase. Any exchange that even nominally 'puts users first' (i.e. isn't trying to screw them for profit) has offered withdrawal of BCH from the get-go. Though OTOH anyone who has seen how Coinbase operates with price fluctuations and so on would have guessed they'd end up doing this for the fork.

Pick your partners wisely folks. :)

It would take a lot of computing power to reverse (at least) 6 blocks worth of transactions.

No, it just takes a miner with >51% mining power. Any of the big pools could do it without flinching.

Not really, if someone had the insane power of 60% mining power, the probability of mining 6 consecutive blocks would still be 0.60^6 only ~4.67%, but this falls exponentially, so if they wait a few more minutes and have to reverse one more block, that's 2.8%

It's just very unlikely this could work. Possible, but very unlikely.

Edit: Also, if you have a very large amount of coins, you probably want to sell it off slowly, instead of tanking the market in one go. This makes it even more difficult since if you're caught reversing the first few transactions, your account will be flagged.

That's not how the 51% attack works.

If you have >51% of the mining power, you can overpower the rest of the network. Simply pick a block in the past and start mining from there, ignoring everyone else. With >51% of the hashrate you _will_ overcome the rest of the network and enforce your new chain.

They can do it in secret. Build a secret parallel chain while they transact, then release the chain after they've got their BTC.

And 51% is not "insane". It's absolutely tiny. As I said, _any_ of the large pools could do it, right now.

OK I must be missing something here: how can any of the large pools have 51% of the mining power?

How can you have more than one pool, each with more than half the power?

EDIT: OK I see where you're coming from I think... you mean any of the pools could 51% the forked branch, right?

Yes 51% of the branch you are attacking. They have hard forked, any bitcoin mining pool could easily control 51% (like 80%) of the bitcash mining worldwide.

I assume it means any of the larger pools in Bitcoin could switch over to one of the smaller altcoins and overwhelm the much smaller pool of miners there.

I don't follow this stuff closely, so I'm clearly missing an important detail here: how can several different entities simultaneously have the ability to "easily" control 51% of the hashrate?

I kind of think it is a little bit like this.

Imagine a new sport called zoccer. It's just like regular soccer, but if a team scores 5 goals within the 90 minutes, they automatically win the match.

Right now there are only a few small teams playing. But Nike has said they are willing to sponsor a tournament, if there is enough interest.

At any point, Barcelona or Manchester United or Real Madrid or Chelsea or AC Milan or any other large soccer team team can just say "we are now a zoccer club" and swoop in take Nike's money.

They all have the power to switch and dominate the new scene.

Bitcoin (BTC) and Bitcoin-cash (BCH) share the same Proof-of-Work (PoW) algorithm. Thus, mining power is split between these two networks. Most mining power is currently on the BTC, which means that it is at least possible that multiple miners on BTC have more than 51% of the mining power currently deployed on BCH.

Having 60% of the mining power means you mine 1.5 blocks on the "false" chain for every block on the "real" chain.

You don't need to mine 6 blocks in a row against the other miners, you just need to have a longer chain. So you simply fork the chain, and then submit that as the "real" chain.

With 60% of the mining power and a 6 block deficit, you'd overtake the main chain in 120 minutes (approx; 12 blocks), since you're making up half a block of deficit per block they mine.

If you really had 60% of the mining power, you could undo 3 days of transactions in about a week of mining.

In what world is 2.8% sufficiently small to be ignored?

6) says that coins with smaller mining pools are more susceptible to double spending attacks correct?

Yes, since it is easier to gain 51% of the available mining power. that's why coins live or die by miner adoption.

You are making a way bigger deal out of this than it is.

Exchanges can quite easily protect themselves.

All they have to do is increase the confirmation time before they "accept" coins.

The longer you wait, the more secure your coins are. I would suspect that a day is easily enough time to be confident that a double spend won't happen.

A 51% attack can reliably undo any number of confirmations.

It is not as easy as that would suggest.

Sure, you can build a longer blockchain, but nothing forces other people to accept that blockchain.

IE, people can checkpoint the chain.

Also, if 51% attacks consistently started to happen, that would cause the price if that chain to plummet to 0.

There is no point stealing money that is worth 0.

A 51% attack could be done by someone who hopes to dump their gains before the market reacts, or it could be done by someone who wants to see the value go to zero. Bitcoin Cash could start checkpointing itself, but at that point it'd be centralized and you're right that its value would probably go to zero.