roconnor-blockstream

Indeed small payload expansions seem to be completely unproblematic. Allowing messages of any lengths between 0 and 55 bytes is basically free, and there seems little reason not to support...

Is it accurate to say that if we only support 32-byte "messages" we should strongly recommend or require that those messages be SHA256 tagged hash values, and that if we...

unification-fd is now broken in nixpkgs-unstable for the same reason. Updating the upper bounds for logict seems to be all that is needed.

the use of `stdio.h` is probably a holdover from when I was using the `FILE` interface. Please make a PR to replace it with `stddef.h`

As far as the C FFI goes, jets are not allowed to do `malloc`, so there should be no problem with compling the jets to WASM?

I intended to have the C Makefile produce a static library for the jets, but I didn't have any actual consumers for such an artefact so I haven't bothered implementing...

The Tech Report seems fine to me.

Additionally `bip0340-schnorr-verify` is written in the tech report; however the code uses `bip_0340_verify`

And `check-sig-verify` is missing from the tech report.

illegal HIDDEN children is formally defined in https://github.com/BlockstreamResearch/simplicity/blob/5e7976326156ac9db9b5f8695dd597e8e370e130/C/dag.h#L308-L314 (technically that defines legal HIDDEN children). I suppose I can break the various conditions of well-formedness into their own names. But I'm...