What if two sidechains are implemented at once? What if people get excited about one sidechain today, but a second even-better one is published the very next week? What if the original mainchain decides to integrate the features of the sidechain that you just one-way pegged to?

In these cases, the user looses money, whereas in the two-way peg they would not lose a thing.

While the one-way peg is interesting, it really doesn't compare.

Paul

On Oct 10, 2017 4:19 PM, "Lucas Clemente Vella" <lvella@gmail.com> wrote:
2017-10-09 22:39 GMT-03:00 Paul Sztorc via bitcoin-dev <bitcoin-dev@lists.linuxfoundation.org>:
That is only a one-way peg, not a two-way.

In fact, that is exactly what drivechain does, if one chooses parameters for the drivechain that make it impossible for any side-to-main transfer to succeed.

One-way pegs have strong first-mover disadvantages.
 
I understand the first-mover disadvantages, but I keep thinking that if the new chain is Pareto optimal, i.e. is in all aspects at least good as the original chain, but in some so much better to justify the change, the initial resistance is an unstable equilibrium. Like a herd of buffaloes attacking a lion: the first buffalo to attack is in awful disadvantage, but if a critical mass of the herd follows, the movement succeeds beyond turning back, and every buffalo benefited, both those who attacked the lion and those that didn't (because the lion was chased away or killed).
 
--
Lucas Clemente Vella
lvella@gmail.com