Re: [tcpm] poll for adopting draft-gont-tcp-security

[Joe Touch <[e-mail]>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1



Fernando Gont wrote:
> Joe Touch wrote:
> 
>>> So...if your opinion has now changed, and the only problem you have
>>> with this I-D is the section on "ICMP processing", would you
>>> support draft-gont-tcp-security (or at least not object its
>>> adoption as a wg item) were I to, e.g., remove the section on ICMP
>>> processing, replace it with a summary of what's in
>>> draft-ietf-icmp-attacks, or e.g., note that the provided advice is
>>> only for TCP implementations that desire good resiliency?
>> It may be useful to explain how to efficiently implement reordering
>> and reassembly, but not in an RFC.
> 
> Is this a joke?
> 
> RFC 816: "MODULARITY AND EFFICIENCY IN PROTOCOL IMPLEMENTATION"

That's 817, FWIW.

> RFC 815: "IP DATAGRAM REASSEMBLY ALGORITHMS" (see Section 4)
> 
> and,
> 
> RFC 1936: "Implementing the Internet Checksum in Hardware" (of which you
> are on of the co-authors)
> 
> You argue against an I-D that gives implementation advice, yet you have
> co-authored such an I-D? Should we take your opinion seriously?

I didn't give advice, I gave an example, as do many RFCs. I don't agree
that efficiency is an appropriate focus of an RFC, and no, that doesn't
stop them from being written by others. There is a difference which is
relevant to some (granted, not all) recommendations in this document.

It is important to deciding whether you're describing (informationally)
or recommending (BCP-ish).

Joe
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkpRgtoACgkQE5f5cImnZrtfNwCeL6XSS17VVEyV+/8jTTJxsWrU
E/EAn1lw40/a4rCCYTXEZLpGBnY5NM7U
=zV7c
-----END PGP SIGNATURE-----