[Analemma_]

Two things:

1) Bitcoin isn't new, it has been around for years now, and millions in venture capital has been poured into it. It's still a pain in the ass to use at all and basically impossible to use securely. I'm starting to think that if it could be fixed, it would be by now.

2) Not all those problems are UX faults. A lot of us happen to consider transaction irreversibility to be a bug and not a feature, for example.

[darawk]

> 1) Bitcoin isn't new, it has been around for years now, and millions in venture capital has been poured into it. It's still a pain in the ass to use at all and basically impossible to use securely. I'm starting to think that if it could be fixed, it would be by now.

I think it's improved quite a bit in that time. Especially so if you consider some of the newer cryptos. Is it there yet? No. But these things take time. Computer interfaces took a long time to get where they are today. It takes a while to find the right abstractions, and people are constantly iterating on these things in the crypto space.

> 2) Not all those problems are UX faults. A lot of us happen to consider transaction irreversibility to be a bug and not a feature, for example.

Well, I respectfully disagree. You may want to build reversibility as a layer on top of Bitcoin. Perhaps something mediated by trusted third parties. But that should be opt in. You want the underlying financial infrastructure to have immutable, irreversible transactions. On top of that base, you can and likely should build support for reversibility in certain contexts.

[yorwba]

> You may want to build reversibility as a layer on top of Bitcoin.

How is that supposed to work? Once the money has been sent to someone's private key, you rely on their cooperation to get it reversed.

To effectively force them to give it up, a trusted third party would have to hold an even larger amount in escrow. But then what if that third party refuses to cooperate? You would need a larger escrow service to hold the escrow service accountable ... or they would have to make their identity public and be personally liable for any losses, putting government regulation back in the loop.

And this is assuming the transaction was made voluntarily in the first place. If someone steals your money they are not going to make the transaction reversible if they can avoid it. Unless you have the clout to convince a majority of miners to do a hard fork, good luck getting your money back.

[darawk]

In Ethereum, reversibility could have strict constraints. For instance, you could devise a contract in which the funds are held in escrow for 90 days, during which time a neutral third party may reverse the transaction using their key.

The important point is that the actual parties to the transaction may choose how they want to mediate this, whether they want reversibility, what conditions they want it to have, and have it all enforced by code and cryptography.

Of course, this burden would only really take place on high value transactions. For most transactions, there'd be a standard configuration people and merchants would use.

[mirimir]

A major innovation by Silk Road was its escrow system.