encrypted unlock time reminds me of the mitch hedberg joke [extension] ".. i'll meet you when that guy's eating a hamburger... well, which guy? ... you'll know..." .. also HNY, MRL!

I feel like it would be helpful to have all functions related to the protocol organized together. Maybe a namespace.. For example, the Blockchain:: namespace seems to contain both protocol and non-protocol functions (e.g. validate_miner_transaction() is protocol, get_current_blockchain_height() is not)

There is a layer above the base protocol, best/standard practices in constructing transactions and blocks, that belong together (e.g. mixin selection algorithm, optimal block construction, RCTTypeFull for only one input, zero lock time, how to use subaddresses and integrated addresses [including default dummy encrypted payment ID])

cryptonote_config seems to contain both protocol and non-protocol global constants

idk anything about the node/network system but I'm sure there are similar boundaries (ex. nodes won't relay tx with fee below the minimum)

Happy 2020 to all!

koe seems to have left, but such reorganization is probably a better topic for -dev

I'm sure there would be many different opinions on it

Thanks to suraeNoether for doing preprint review; I'll start incorporating his suggestions in preparation for IACR archive posting

A great way to start out the new year!

Hny Scooby. Koe and endogenic should talk about the codebase together

Sarang: still reading the multi version; expect comments later

Then onto clsag

Thanks! For multi, I suggest reading the soundness portion to see if the witness extraction and proof relation make sense

The rest isn't as polished as the single-index IACR version

I added line numbers to multi.tex to make it easier to reference

k