Re: [Ntp] [EXT] Need guidance on interpreting NISTIR 8366

Ulrich Windl <> Wed, 12 May 2021 22:49 UTC

Return-Path: <>
Received: from localhost (localhost []) by (Postfix) with ESMTP id 6AA753A1786 for <>; Wed, 12 May 2021 15:49:12 -0700 (PDT)
X-Virus-Scanned: amavisd-new at
X-Spam-Flag: NO
X-Spam-Score: -1.9
X-Spam-Status: No, score=-1.9 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, NICE_REPLY_A=-0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, URIBL_BLOCKED=0.001] autolearn=ham autolearn_force=no
Received: from ([]) by localhost ( []) (amavisd-new, port 10024) with ESMTP id EoLMuw2lZRHZ for <>; Wed, 12 May 2021 15:49:08 -0700 (PDT)
Received: from ( [IPv6:2001:638:a05:137:165:0:3:bdf7]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by (Postfix) with ESMTPS id EC9A83A1780 for <>; Wed, 12 May 2021 15:49:07 -0700 (PDT)
Received: from (localhost []) by localhost (Postfix) with SMTP id BA2906000050 for <>; Thu, 13 May 2021 00:49:01 +0200 (CEST)
Received: from ( []) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (Client CN "", Issuer "DFN-Verein Global Issuing CA" (not verified)) by (Postfix) with ESMTPS id 846BA600004D for <>; Thu, 13 May 2021 00:49:01 +0200 (CEST)
References: <>
From: Ulrich Windl <>
Message-ID: <>
Date: Thu, 13 May 2021 00:49:01 +0200
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101 Thunderbird/78.10.0
MIME-Version: 1.0
In-Reply-To: <>
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Language: en-US
Content-Transfer-Encoding: 8bit
Archived-At: <>
Subject: Re: [Ntp] [EXT] Need guidance on interpreting NISTIR 8366
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: <>
List-Unsubscribe: <>, <>
List-Archive: <>
List-Post: <>
List-Help: <>
List-Subscribe: <>, <>
X-List-Received-Date: Wed, 12 May 2021 22:49:13 -0000

On 5/12/21 7:58 PM, Daniel Franke wrote:
> (CCed recipients, see 
> < <>> 
> for context)
> I'm writing for guidance on interpreting NISTIR 8366 as it pertains to
> my work and that of the NTP working group, and in particular on
> several matters surrounding the following language from section 4.1:
>      "Avoid terms that perpetuate negative stereotypes or unequal
>       power relationships
>           Examples: master/slave; smart/dumb; right/left"

I'm afraid any "dumbphone" wouldn't make it to the market (just as 
D.U.M.B. monitoring for harddrives;-)

> with the footnote,
>      "In this example, avoid using right to mean 'good' or 'normal,'
>      and left to mean 'bad' or 'abnormal.'"
> The NTP working group is presently discussing extending RFC 8915
> ("Network Time Security for the Network Time Protocol") to support
> PTP (the Precision Time Protocol, IEEE 1588). IEEE 1588 makes
> pervasive use of master/slave terminology to describe fundamental
> concepts of the protocol. (As a historical aside, the terms and
> concepts of "master clock" and "slave clock" are much older than PTP;
> see < 
> <>> for some
> examples from as early as 1838). What should we consider to be best
> practice for dealing with this sort of situation, where we're building
> on work that originates from a non-IETF standards body and uses
> terminology that is plainly non-compliant with NISTIR 8366?

I wonder: Will the movie "Masters of the Universe" be renamed?
Is "Superman" still allowed? Blackholes? White dwarfs?

> Any work that builds on IEEE 1588 can't reasonably avoid using the
> technical concepts that IEEE 1588 uses "master/slave" to represent.
> It could continue using the same terminology, either with or without
> an acknowledgement that it is violating NISTIR 8366 in doing so. Or,
> it could define its own synonyms, but would still need to use the
> original terms at least one when defining the new ones in order to
> make the synonymy clear to the reader. What's the correct approach?

I think the real question is: How many non-master people were ever hurt 
by a master PTP clock? How many people are afraid of a PTP master clock?

> Taking this a little further: "unequal power relationships" is an
> awfully broad concept and I don't think any time synchronization
> protocol can ever get away from it no matter what euphemisms it
> chooses. I happen to have written Byztime
> < 
> <>> which is a truly
> peer-to-peer protocol, but NTP and PTP are inherently

I also wonder: Have there been any claims against the phrase "black power"?

> hierarchical. They have to be, because they exist to support a
> hierarchical social function. The answer to "what time is it?", in the
> context of TAI or any timescale based on it (such as UTC) ultimately
> derives from a central authority, BIPM, and the network of reference
> clocks coordinated under it. And even with Byztime, there's still
> power inequality between the set of configured peers, which are
> authorized to participate in consensus, and the rest of the world
> which is unauthorized.
> I think we can find some principled middle ground here. I take for
> granted that everyone reading this agrees that slavery is evil, but
> only a tiny minority, mostly the intellectual heirs of Proudhon, favor
> the complete abolition of *all* power hierarchies. I would be very
> surprised if NIST, a bureau of the United States government, intended
> to officially endorse anarchism, so I think a more moderate reading is
> justified. I propose that the text be understood as being limited in
> scope to power relationships in which participation is non-consensual
> (FSVO, recognizing that different political philosophies have
> different notions of what consent means). Hence, "master/slave" is
> still out but we can use "client/server", "controller/agent", and
> other such terms that still imply unequal power relationships but lack
> connotations of coercion.

Are a "brown out" and "blackouts" still legal, an is "I had a blackout" 
no longer an excuse?

> Finally, as to that footnote about "left/right". This seems to be
> telling me that RFC 8280 ought not have spoken approvingly about
> "human rights" and that we ought to rename our "intellectual property
> rights" disclosures. "Right" as in "right-handed", "right" as in
> "correct", and "right" as in "just entitlement" are not merely
> homonyms; they're the same word with the same etymology, tracing to
> the Proto-Indo-European "h₃reǵtós" which had all the same meanings.
> The same root, with all its meanings preserved, shows up in other
> modern Indo-European languages as well, such as the Spanish word
> "derechos". Now, this is obviously ridiculous and not an intended
> reading. I've never encountered anyone who sincerely thinks that
> phrases like "human rights" are problematic, and surely I've gotten
> too literal. But I'm struggling to come up with any more reasonable
> interpretation. How am I to take this footnote seriously and give it
> meaning, while avoiding this sort of inference?

I think (as we know from politicians): As words don't make the world any 
better, words also don't make it worse.  Isn't this all just a great 
waste of time and energy?


> _______________________________________________
> ntp mailing list