Re: [lamps] Proposed addition of hash-based signature algorithms for certificates to the LAMPS charter

[Stephen Farrell <[e-mail]>]

Hiya,

On 07/11/2018 03:50, Russ Housley wrote:
> The SECDISPATCH WG met on Tuesday afternoon, and they made this recommendation:
> 
>>  draft-vangeest-x509-hash-sigs-01 -- re-charter LAMPS WG to accept this draft
> 
> Three questions:
> 
> 1) Do you support the addition of this work to the LAMPS charter?

Not sure. (Sorry for being on the fence;-)

I believe that the hash based signature schemes are
sufficiently secure, if implemented and used properly,
so I don't have concerns about the basic crypto, though
we've ~zero experience in implementing and using these
schemes afaik.

I do think there will be applications for those schemes
but am concerned that it may be a bad plan to define
how to use those schemes for X.509. I'd much prefer if
there were interest in trying to see how to move beyond
X.509 entirely as we consider attempts to be secure
in a post-quantum sense. If there were sufficient interest
in that (moving beyond X.509) then the valid concerns that
Adam raised (about e.g. APIs and recovery) could, I think,
be better handled. (Sadly, I'm unaware of any credible
proposal to move beyong X.509, much as I'd like to see
one;-) I don't think LAMPS would be the right venue for
such work, were there to be interest in that.

If, OTOH, it is inevitable that X.509 CAs want to start
signing certificates using these schemes, then I guess
just defining OIDs in LAMPS would be reasonable, but I
figure that's the wrong direction to take, and I'm not
sure what relying parties would really be using them, so
defining those OIDs may be premature.

> 2) If it is added, would you review the document?

Maybe.

> 3) If it results in an RFC, would you implement?

Not I.

Cheers,
S.


> 
> Russ & Tim
> _______________________________________________
> Spasm mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/spasm
>