I think this edit helps, but is still a bit confusing in that the term “transaction validation” is overloaded. Later in the post, you define “Validation” and “Confirmability” as two separate things, i.e., “Validation” is all verification checks except “Confirmability” and “Confirmability” is only verifying prevouts exist and are unspent. I think the OP would be more clear if it explicitly mentions that for blocks under the milestone, both “Validation” and “Confirmability” checks are skipped. I think this also makes it more clear where the differences with Core’s architecture are: while this is conceptually a similar threat model, Core does need to check that inputs exist and are unspent because this is necessary for maintaining a “UTXO set” (in that we must know what inputs are being spent so that they can be deleted from the UTXO set before continuing).
1 Like