[imtringued]

Yes Bitcoin Cash wouldn't have to exist if the original bitcoin just accepted the 2x blocksize increase. That would have given Bitcoin one or two years more time until scaling becomes a problem again. That's enough time for Lightning Network to be completed.

[gruez]

>Yes Bitcoin Cash wouldn't have to exist if the original bitcoin just accepted the 2x blocksize increase.

the segwit2x fork was planned for AFTER the bcash/bch fork. also, there were 2 issues at the time: block size increase and segwit implementation. the segwit2x fork didn't occur, so everybody is pigeonholed into either [yes blocksize increase, no segwit] or [no blocksize increase, yes segwit]

[xorcist]

Scaling is what limits capacity increase. A block size increase, taken in isolation, does not change scaling. It only means capacity is increased.

And while it would be welcome to increase capacity, there are many ways to do that. Coinbase fixing their broken wallet would single handedly give more spare capacity than any other measure right now. If their customers would instead choose a service where you pay low or no fees we wouldn't have the fee spikes we have right now. To argue that a globally coordinated rule change is somehow the easy way out requires some very good explanation.

That said, it might be prudent to mention in this context that we activated a blocksize increase only a few months ago. The opt-in, non-coordinated kind, which you are free to use at any time. (Unless you are with Coinbase, of course. In which case, see above.)

[tedunangst]

How robust can a blockchain be if a single wayward service can cripple the network?

[xorcist]

That's a very good question! It's robust in the sense that there is a lot of work produced with no downtime, but not very in the sense that most people gets less useful work out of it. Consider the case there most users would like to tumble their coins before each transaction, as was the case with certain markets, preferably a full 16 rounds and how that would affect usability of the system as a whole. This problem affects dark net markets just as well as exchanges.

There are also other real life examples too, of single wayward services that made the blockchain harder to use for everyone, such as the casino which recorded a transaction in the global ledger for all eternity for every dice roll. Those are all priced out now, together with some of the more excessive tumblers, but that solution could be considered crippling the network.

Are there "too big to fail" services within the Bitcoin ecosystem? Matters are complicated further by the fact that the exchange in question foot the bill to their customers, one customer at a time, so the economic impact for them is negligible.

Why "fail", then? Well, it turns out that somebody looked at their hot wallets recently and noted that the mostly inappropriate choices of transactions to make had resulted in a non trivial amount of non-spendable dust collecting. In a bank run situation, that would result in them not being able to fully empty their wallets. The further down we investigate, the more there is to learn.