Re: Author and attendance measurements [Was: Re: Thought experiment [Re: Quality of Directorate reviews]]

John C Klensin <> Fri, 08 November 2019 13:56 UTC

Return-Path: <>
Received: from localhost (localhost []) by (Postfix) with ESMTP id EAE3B1200F9 for <>; Fri, 8 Nov 2019 05:56:25 -0800 (PST)
X-Virus-Scanned: amavisd-new at
X-Spam-Flag: NO
X-Spam-Score: -1.898
X-Spam-Status: No, score=-1.898 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, SPF_HELO_NONE=0.001, SPF_NONE=0.001] autolearn=ham autolearn_force=no
Received: from ([]) by localhost ( []) (amavisd-new, port 10024) with ESMTP id Ln_TnJCsGfe0 for <>; Fri, 8 Nov 2019 05:56:24 -0800 (PST)
Received: from ( []) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by (Postfix) with ESMTPS id C73D212003F for <>; Fri, 8 Nov 2019 05:56:23 -0800 (PST)
Received: from [] (helo=PSB) by with esmtp (Exim 4.82 (FreeBSD)) (envelope-from <>) id 1iT4kP-000Cz6-BH; Fri, 08 Nov 2019 08:56:21 -0500
Date: Fri, 08 Nov 2019 08:56:15 -0500
From: John C Klensin <>
To: Keith Moore <>,
Subject: Re: Author and attendance measurements [Was: Re: Thought experiment [Re: Quality of Directorate reviews]]
Message-ID: <335EAA17262BD4585E4E8964@PSB>
In-Reply-To: <>
References: <> <> <> <> <> <> <> <26819.1572990657@localhost> <> <> <> <> <> <> <> <11796.1573157656@localhost> <> <>
X-Mailer: Mulberry/4.0.8 (Win32)
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline
X-SA-Exim-Scanned: No (on; SAEximRunCond expanded to false
Archived-At: <>
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: IETF-Discussion <>
List-Unsubscribe: <>, <>
List-Archive: <>
List-Post: <>
List-Help: <>
List-Subscribe: <>, <>
X-List-Received-Date: Fri, 08 Nov 2019 13:56:26 -0000

--On Friday, November 8, 2019 02:31 -0500 Keith Moore
<> wrote:

> On 11/8/19 2:11 AM, Jari Arkko wrote:
>> Pretty healthy numbers I'd say, even, if 600-700 people
>> manage to  publish RFCs and the meetings draw 900-1100 people.
> To me it sounds unhealthy.   If meeting attendance is an
> indicator of the number of active participants, it sounds like
> about half of the participants are publishing an RFC every
> year. With that many authors I have to wonder how many people
> have time to participate in discussion of others'
> documents.   Of course a person can author one document
> while discussing others, but it's easy for an author's
> attention to others' drafts to diminish while revising his own
> draft.


As Jari cautioned, those numbers have to be interpreted with
care.  For example, the number of documents with many authors
has has risen over the years since Jon Postel aggressively
pushed back against the practice.  We also have issued documents
that are revisions of older ones.  In general, original author
names have been retained even when newer authors have been added
on and the actual involvement of the original ones has been
minimal.   There may also be issues with listed contributors
because, at the discretion of a WG or whomever is holding the
pen, that category may include someone who contributed a concept
or a few critical paragraph, but who was not otherwise
significantly involved in the production of the document.

Even Jari's comment that you quote is problematic because, while
600-700 people may have their names on RFCs that are published,
sometimes it is accurate to say that they published them and
sometimes some of them are just names.  Even the definition of
"author" may differ from one document to the next.  We also
aren't careful about distinguishing between authors and editors.
For example, with a document that is really developed by a WG,
the "author" may be mostly a compiler of WG comments and text
rather than a contributor of and advocate for original ideas.
Either may involve a great deal of time and effort, but the kind
and level of involvement is different.  I even know of at least
a couple of often-cited RFCs in which there are several listed
authors but none of them, independent of their intellectual or
other contributions, actually provided any significant fraction
of the published text.

Similarly, comparing author counts to f2f meeting registrations
does not quite work and is probably getting worse over time as
meeting costs rise and active remote participation increases.
We should have at least approximate numbers by now for total
registered participants, but probably not reaching back to the
point that Jari is comfortable with the authorship numbers. 

I can imagine ways to estimate numbers and control, in the
statistical sense, for some (but not all) of those issues
without, e.g., interviewing authors or doing a careful
document-by-document evaluation.   They all would require a lot
of work.  Maybe others disagree, but I'd rather see the people
who would probably need to do that work spending their time on
substantive technical contributions instead.  But, without doing
that work (at least), inferences about how many people are
writing documents and therefore not evaluating those of others
seems fairly close to pure speculation.