[Ntp] draft-ietf-ntp-roughtime

Erik Kline <ek.ietf@gmail.com> Mon, 27 September 2021 22:13 UTC

MIME-Version: 1.0
From: Erik Kline <ek.ietf@gmail.com>
Date: Mon, 27 Sep 2021 15:13:38 -0700
Message-ID: <CAMGpriVBpDUf2E4k_QmnyQuJ8k6BW9MLGVxXenAb4TGUQSG3cg@mail.gmail.com>
To: draft-ietf-ntp-roughtime.authors@ietf.org
Cc: NTP WG <ntp@ietf.org>, maik.riechert@arcor.de
Content-Type: multipart/alternative; boundary="00000000000049af5b05cd01690b"
Archived-At: <https://mailarchive.ietf.org/arch/msg/ntp/XPFSYCPjshtSMUWi-589fpU-zfg>
Subject: [Ntp] draft-ietf-ntp-roughtime
Precedence: list

Authors, all,

Per this message [1] on the chromium.org proto-roughtime mailing list, Maik
(cc'd) had some trouble posting his comments to this mailing list (not sure
why yet).

I'm including the comments inline here for everyone's consideration:

"""
Hi,

In the Roughtime draft at
https://www.ietf.org/archive/id/draft-ietf-ntp-roughtime-05.html there is a
comparison with other NTP-like protocols, but I'm missing a comparison with
RFC 3161. I know that both protocols serve different purposes (time server
vs. time stamping) but they still seem to have a lot in common.

In my opinion, Roughtime could be seen as a more efficient alternative to
RFC 3161 by supporting batch signing through Merkle trees. RFC 3161 also
supports Roughtime's intended usage model of asking multiple servers
sequentially since this just relies on nonces.

Of course, Roughtime doesn't directly support time stamping (binding) data,
since that's not needed for time servers, but you could abuse the nonce for
that to achieve the same effect.

In summary, I think a comparison with RFC 3161 would clear up uncertainty
and confusion, as I've seen both mentioned together in various places on
the web.

Regards
Maik
"""

Please pardon the unusual cut'n'paste of comments from other mailing lists.

Thanks,
-Erik

[1]
https://groups.google.com/a/chromium.org/g/proto-roughtime/c/2zNTyBE5b1M/m/-myxssANAQAJ

[Ntp] draft-ietf-ntp-roughtime Erik Kline