Documentation

[survived]

Just a list of things not to forget to mention:

• Signers indexes
  • Signers are identified by their indexes i: $0 \le i < n$
  • Signers must occupy exactly the same indexes at keygen, key refresh, and aux data (re)gen. \ E.g. if signer-abc occupied index 2 at keygen, it must have exactly the same index at key refresh and aux data (re)gen.
  • At signing, caller specifies a map that defines which index i-th signer had at keygen.

[maurges]

At signing, caller specifies a map that defines which index i-th signer had at keygen

You mean the VssSetup::I? I thought it's not an index map, but a map of points used for SSS. In that sense we still need to preserve indexes, otherwise lookups will give wrong points.

[survived]

Nope, I mean parties_indexes_at_keygen or S in the protocol and in the specs

[maurges]

Ok, I think we need something similar for key refresh then

[survived]

Imo it makes less sense to do that in key refresh as all n parties must participate in key refresh

On Mon, Apr 24, 2023, at 11:52 AM, d86leader wrote:

Ok, I think we need something similar for key refresh then

— Reply to this email directly, view it on GitHub https://github.com/dfns-labs/cggmp21/issues/23#issuecomment-1519780094, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADRTINCHNR3FNBB5WHOQMOLXCZEOTANCNFSM6AAAAAAXHBEKOA. You are receiving this because you authored the thread.Message ID: @.***>