On 9/21/12 12:39 PM, Michael Richardson wrote:
"Sean" == Sean Turner <[email protected]> writes:
Sean> that requested the points and that the "notes" column would
Sean> contain "not for IKEv1" - and then we'd talk about it. Dan
...
Sean> In this unfortunate situation, I'd like everyone to consider
Sean> the (third surgically attached) hand that shares the burden:
Sean> reserve the code points for 802.11 SAE in the Group
Sean> Description registry, be very explicit about it in the
As I was reading up just before this, my thought was to write, rather
than "not for IKEv1", instead write:
"only for 802.11 SAE"
but it would actually say what they were.
Sean> more). The burden is then shared by the IETF assigning code
Sean> points for something some despise/dislike and the IEEE
Sean> implementers following an additional link from the registry
Sean> they've already got to consult (they have to follow the link
Sean> because the registry values aren't copied to their spec). The
I understand that you are hacking on the dumb customer reading registry
is also too dumb to follow the breadcrumbs over to IEEE in order to find
out what the groups are, in order to properly demand things.
<TCPDUMP HAT ON>
I don't like this.
I want to know what the code point is for. tcpdump is not implementing
the protocol, just decoding it. I don't know if IEEE 802.11 will let
me see the assignments without jumping through hoops, of it they
will get mad if someone who has access to documents, submits a patch.
</TCPDUMP HAT OFF>
I claim that customers are actually too dumb to read the IANA registry
anyway, at most they grep rfc-index.txt for "IPSEC" and list that. So
actually it doesn't matter what we say in the IANA registry, as long
as implementers get it, it's okay.
if: "Reserved for 802.11 SAE Brainpool Group 14"
will fit, I say go for that. It means that tcpdump will print something
like that out if it sees things (and if this is used in the parent SA,
it will be visible).
Michael,
I can see your point. I can see that the additional layer of
redirection I added could be considered useless. From a procedural
point of view though, I think we should discourage dual use of registries.
spt
_______________________________________________
IPsec mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/ipsec
