Return-Path: <jmh@joelhalpern.com>
X-Original-To: gendispatch@mail2.ietf.org
Delivered-To: gendispatch@mail2.ietf.org
Received: from localhost (localhost [127.0.0.1])
	by mail2.ietf.org (Postfix) with ESMTP id E11A06E9E92C
	for <gendispatch@mail2.ietf.org>; Tue,  7 Oct 2025 06:34:34 -0700 (PDT)
X-Virus-Scanned: amavisd-new at ietf.org
X-Spam-Flag: NO
X-Spam-Score: -2.796
X-Spam-Level: 
X-Spam-Status: No, score=-2.796 tagged_above=-999 required=5
	tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1,
	DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, HTML_MESSAGE=0.001,
	RCVD_IN_DNSWL_LOW=-0.7, RCVD_IN_MSPIKE_H3=0.001,
	RCVD_IN_MSPIKE_WL=0.001, RCVD_IN_VALIDITY_RPBL_BLOCKED=0.001,
	RCVD_IN_VALIDITY_SAFE_BLOCKED=0.001, SPF_PASS=-0.001]
	autolearn=ham autolearn_force=no
Authentication-Results: mail2.ietf.org (amavisd-new); dkim=pass (1024-bit key)
	header.d=joelhalpern.com
Received: from mail2.ietf.org ([166.84.6.31])
	by localhost (mail2.ietf.org [127.0.0.1]) (amavisd-new, port 10024)
	with ESMTP id hDltU3PXkFRD for <gendispatch@mail2.ietf.org>;
	Tue,  7 Oct 2025 06:34:34 -0700 (PDT)
Received: from mailb2.tigertech.net (mailb2.tigertech.net [208.80.4.154])
	(using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits)
	 key-exchange X25519 server-signature ECDSA (P-256))
	(No client certificate requested)
	by mail2.ietf.org (Postfix) with ESMTPS id 4090C6E9E925
	for <gendispatch@ietf.org>; Tue,  7 Oct 2025 06:34:34 -0700 (PDT)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=joelhalpern.com;
	s=2.tigertech; t=1759844073;
	bh=3paBgu94hvmfyilc0zSGk1IAbqUrXwhAIU/oonq36Lo=;
	h=Date:Subject:To:Cc:References:From:In-Reply-To:From;
	b=CLqMzB0HbwlR9nW1vrJB4vfQIZLsi6G/Ss6CXukOlQmEWHM6U/BPQR0Mu5dvmxY6j
	 u8ih6tmQSgJ5ddqHMMDeAD+CpS1ZbBZGjXv3a+TcLPdBpt4wBs9tc/Uqr5jf1kDshw
	 Dzqegt8/CTRaxXxxMB+slVzdF32aumjmd9BXxG8k=
Received: from localhost (localhost [127.0.0.1])
	by mailb2.tigertech.net (Postfix) with ESMTP id 4cgxvF3xGlz1nsBD;
	Tue,  7 Oct 2025 06:34:33 -0700 (PDT)
X-Quarantine-ID: <rtGaIO8JyOJM>
X-Virus-Scanned: Debian amavis at b2.tigertech.net
Received: from [IPV6:2600:8806:103:9500:e142:600:828f:e5c2] (unknown
 [IPv6:2600:8806:103:9500:e142:600:828f:e5c2])
	(using TLSv1.3 with cipher TLS_AES_128_GCM_SHA256 (128/128 bits)
	 key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest
 SHA256)
	(No client certificate requested)
	by mailb2.tigertech.net (Postfix) with ESMTPSA id 4cgxvD5ZBzz1ns8V;
	Tue,  7 Oct 2025 06:34:32 -0700 (PDT)
Content-Type: multipart/alternative;
 boundary="------------yBdJYLZ6exlaO4zLRH4EHUdk"
Message-ID: <872c83bd-ecba-458f-a67a-9dbc9a888664@joelhalpern.com>
Date: Tue, 7 Oct 2025 09:34:28 -0400
MIME-Version: 1.0
User-Agent: Mozilla Thunderbird
To: Kyle Rose <krose@krose.org>
References: 
 <CABcZeBMcY8EnzBM9OfQSH-sd5WZPdx66sMB37CC6q=X8j6nQsg@mail.gmail.com>
 <87587424-efe8-48e8-bfcf-6a1d3e542a58@joelhalpern.com>
 <CAJU8_nXwNBuANX3BHuLS6nXrpSC-752EX1A-b557PGYA7+J+Kw@mail.gmail.com>
Content-Language: en-US
From: Joel Halpern <jmh@joelhalpern.com>
In-Reply-To: 
 <CAJU8_nXwNBuANX3BHuLS6nXrpSC-752EX1A-b557PGYA7+J+Kw@mail.gmail.com>
Message-ID-Hash: UI5RIBSPMYRRODGXZALNAIRUDBEWV24Q
X-Message-ID-Hash: UI5RIBSPMYRRODGXZALNAIRUDBEWV24Q
X-MailFrom: jmh@joelhalpern.com
X-Mailman-Rule-Misses: dmarc-mitigation; no-senders; approved; emergency;
 loop; banned-address; member-moderation; nonmember-moderation; administrivia;
 implicit-dest; max-recipients; max-size; news-moderation; no-subject;
 digests; suspicious-header
CC: Eric Rescorla <ekr@rtfm.com>, GENDISPATCH List <gendispatch@ietf.org>
X-Mailman-Version: 3.3.9rc6
Precedence: list
Subject: =?utf-8?q?=5BGendispatch=5D_Re=3A_New_draft=3A_draft-rescorla-auto-minutes-0?=
	=?utf-8?q?0?=
List-Id: General Area Dispatch <gendispatch.ietf.org>
Archived-At: 
 <https://mailarchive.ietf.org/arch/msg/gendispatch/z7laoJ3DHMHKKkLqCpOeZKbn_OA>
List-Archive: <https://mailarchive.ietf.org/arch/browse/gendispatch>
List-Help: <mailto:gendispatch-request@ietf.org?subject=help>
List-Owner: <mailto:gendispatch-owner@ietf.org>
List-Post: <mailto:gendispatch@ietf.org>
List-Subscribe: <mailto:gendispatch-join@ietf.org>
List-Unsubscribe: <mailto:gendispatch-leave@ietf.org>

This is a multi-part message in MIME format.
--------------yBdJYLZ6exlaO4zLRH4EHUdk
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit

I have no objection to having AI transcripts (although experience in 
other fora indicates they have a ways to go.)  I simply object to 
treating those as a replacement for the minutes that highlight the key 
discussion items, action points, and decisions to be verified on the 
list.  (I can imagine that in a few years tools will be good enough to 
generate those with sufficient accuracy to actually save work, but we 
are not there yet.)

Yours,

Joel

On 10/7/2025 9:30 AM, Kyle Rose wrote:
> On Tue, Oct 7, 2025 at 9:07 AM Joel Halpern <jmh@joelhalpern.com> wrote:
>
>     As was discussed in the past, I believe this misses the point of the
>     minutes.  I do not object to having the recordings and
>     transcripts.  But
>     those do not highlight the decisions and action points, which is the
>     primary purpose of the minutes.
>
>
> While I and many others agree with you on this point, the last 
> discussion on the topic of how best to take minutes demonstrated 
> clearly that there is not consensus on this point.
>
> So given that some people really want a discussion transcript, an AI 
> transcript is *at least* a useful tool to those so interested. A 
> corrected transcript can then be fed back into the LLM to produce its 
> summary of any decisions reached, another thing that is *at least* a 
> useful tool to someone trying to compile such a list. These are tasks 
> that LLMs are actually pretty good at, and we should take advantage of 
> them, both to improve the output and to avoid the "Bueller?" problem 
> of begging for minute takers at the start of each session.
>
> Anything that reduces toil is a good thing, IMO. Let's not let the 
> perfect be the enemy of the good here.
>
> Kyle
>
--------------yBdJYLZ6exlaO4zLRH4EHUdk
Content-Type: text/html; charset=UTF-8
Content-Transfer-Encoding: 8bit

<!DOCTYPE html>
<html>
  <head>
    <meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
  </head>
  <body>
    <p>I have no objection to having AI transcripts (although experience
      in other fora indicates they have a ways to go.)  I simply object
      to treating those as a replacement for the minutes that highlight
      the key discussion items, action points, and decisions to be
      verified on the list.  (I can imagine that in a few years tools
      will be good enough to generate those with sufficient accuracy to
      actually save work, but we are not there yet.)</p>
    <p>Yours,</p>
    <p>Joel</p>
    <div class="moz-cite-prefix">On 10/7/2025 9:30 AM, Kyle Rose wrote:<br>
    </div>
    <blockquote type="cite"
cite="mid:CAJU8_nXwNBuANX3BHuLS6nXrpSC-752EX1A-b557PGYA7+J+Kw@mail.gmail.com">
      <meta http-equiv="content-type" content="text/html; charset=UTF-8">
      <div dir="ltr">
        <div class="gmail_quote gmail_quote_container">
          <div dir="ltr" class="gmail_attr">On Tue, Oct 7, 2025 at
            9:07 AM Joel Halpern &lt;<a
              href="mailto:jmh@joelhalpern.com" moz-do-not-send="true"
              class="moz-txt-link-freetext">jmh@joelhalpern.com</a>&gt;
            wrote:<br>
          </div>
          <blockquote class="gmail_quote"
style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">As
            was discussed in the past, I believe this misses the point
            of the <br>
            minutes.  I do not object to having the recordings and
            transcripts.  But <br>
            those do not highlight the decisions and action points,
            which is the <br>
            primary purpose of the minutes<span class="gmail_default"
              style="font-size:small">.</span></blockquote>
          <div><br>
          </div>
          <div><span class="gmail_default" style="font-size:small">While
              I and many others agree with you on this point, the last
              discussion on the topic of how best to take minutes
              demonstrated clearly that there is not consensus on this
              point.</span></div>
          <div><span class="gmail_default" style="font-size:small"><br>
            </span></div>
          <div><span class="gmail_default" style="font-size:small">So
              given that some people really want a discussion
              transcript, an AI transcript is *at least* a useful tool
              to those so interested. A corrected transcript can then be
              fed back into the LLM to produce its summary of any
              decisions reached, another thing that is *at least* a
              useful tool to someone trying to compile such a list.
              These are tasks that LLMs are actually pretty good at, and
              we should take advantage of them, both to improve the
              output and to avoid the "Bueller?" problem of begging for
              minute takers at the start of each session.</span></div>
          <div><span class="gmail_default" style="font-size:small"><br>
            </span></div>
          <div><span class="gmail_default" style="font-size:small">Anything
              that reduces toil is a good thing, IMO. Let's not let the
              perfect be the enemy of the good here.</span></div>
          <div><span class="gmail_default" style="font-size:small"><br>
            </span></div>
          <div><span class="gmail_default" style="font-size:small">Kyle</span></div>
          <div><span class="gmail_default" style="font-size:small"><br>
            </span></div>
        </div>
      </div>
    </blockquote>
  </body>
</html>

--------------yBdJYLZ6exlaO4zLRH4EHUdk--