[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Camellia for OpenPGP
-----BEGIN PGP SIGNED MESSAGE-----
On Apr 23, 2007, at 10:11 AM, Simon Josefsson wrote:
> Jon Callas <jon@xxxxxxxxxx> writes:
>> Personally, I don't think that Camellia's (or any other symmetric
>> cipher's) patent status matters for us. An implementer can safely
>> ignore an algorithm with little impact.
> The proposal was to make Camellia a normative reference in the base
> OpenPGP document. I see a problem with that.
> Having the reference be informational would be OK, as long as there
> are no requirements to implement it. Only then can implementers that
> wish to conform to the base spec safely ignore Camellia.
> Given what people have said about NTT's intentions in this area, it
> may be worthwhile to pursue and ask them to provide a better patent
> license that would guarantee that Camellia can be used freely. But
> that is a separate problem, and not related to OpenPGP. Until there
> is a good license available, I would be against anything more than a
> informational reference in the OpenPGP document.
There are not going to be (substantial) changes to 2440bis. We're far
too late for that -- and that's a process comment. We're waiting for
the IESG to decide 2440bis can go out the door. This would be a
substantial change. The 2440bis train has left the station. I
genuinely thought that went without saying, and offer that as my
excuse for not saying it.
However, if someone were to write a document for Camillia in OpenPGP,
that would be a reasonable thing to do -- *after* 2440bis comes out
of the pipe. It is in fact the rough consensus of this group that
that's presently the best way to add algorithms.
"Normative" and "informational" aren't precisely the right terms, but
I know what you mean. I think you meant to say "mandatory-to-
implement" (or MUST) versus "optional-to-implement" (or MAY). I agree
with you completely that making Camillia a MUST would not be good,
but I didn't read it. Perhaps I just assume that any new algorithm
would be a MAY. Certainly I interpreted this request this way.
The reason I think that patents aren't an issue is that we have two
algorithms that were problematically patented, and still have one.
They are IDEA and RSA. If you remember, there were OpenPGP
implementations that did not implement RSA until the patent expired,
and some that were a bit schizo about it. IDEA is still problematic
from an intellectual property standpoint, but there's an easy way
around it, which is just don't implement it. Since compatibility with
PGP 2 is now deprecated, there's no need to implement IDEA, and so
the patent on it only hurts IDEA.
I think the same is true with Camillia (and SEED, GOST, etc.) --
whatever the IP status is, it doesn't matter because we don't *need*
these algorithms. If the IP is so onerous that they aren't
ubiquitously deployed, that's their problem, not ours. Note that
assumes they'd be MAYs.
-----BEGIN PGP SIGNATURE-----
Version: PGP Universal 2.6.1
-----END PGP SIGNATURE-----