A problem with CA

[william.sanders]

@mthcl

Assuming that

1. we have (decentralized) checkpoints, and
2. with the checkpoints, we can design the rules in such a way that the (signed) timestamps are not too much off with very high probability (as we recently discussed in the “protocol” group),

would it be easier to solve the above problems?

After I now actually understand how you want to set checkpoints, this maybe could work, because we do 1 single colored voting for a milestone.

However, I have no idea how colored voting would work for CA.