> But if, as part of testing new PQC algorithsm, one really does want
> to exercise the full DNSSEC state-machine, including use of DS RRs
> for downgrade-resistant signalling of support for particular
> algorithms, then Mark's proposal seems to make sense, and does not
> appear to impose any  burden on implementations that don't support
> PRIVATE algorithms.

I think in case of PQC experiments the following holds true:

- DNSOP is considered overloaded.
- A draft for a standards track RFC should have a strong argument why this
  change is needed.
- There are 2 code points for private DS algorithms that could be used to
  implemented what is described in this draft.

So I think is on the PQC experiments to describe why using one of those
code points for private hash functions is not enough to conduct the
experiment. Just saying that new standard hash functions (which are
completely different from the behavior of any of the current hash
functions) do not 'appear to impose any burden on implementation [...]'
is not enough.

In my opinion, PQC experiments should use SHA-256. However that discussion
is less relevant at the moment. Any experiment that wants to use a different
hash function can just use a code point for private hash functions.

_______________________________________________
DNSOP mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to