MIME-Version: 1.0
References: <yblwm4259bs.fsf@wx.hardakers.net>
 <3B7B6340-D218-4992-A11A-335B0DAF6007@sury.org>
 <ybla50x62zj.fsf@wx.hardakers.net>
 <42051fc6-570f-47fe-a28e-480417a370ec@isc.org>
In-Reply-To: <42051fc6-570f-47fe-a28e-480417a370ec@isc.org>
From: Shumon Huque <shuque@gmail.com>
Date: Fri, 7 Nov 2025 16:15:58 -0500
Message-ID: 
 <CAHPuVdXZtj-oUcbtZCgxxCr5zN0On4NEXPv43SjphrYgN4STpQ@mail.gmail.com>
To: =?UTF-8?B?UGV0ciDFoHBhxI1law==?= <pspacek@isc.org>
Content-Type: multipart/alternative; boundary="000000000000bf47df064307b0f2"
Message-ID-Hash: BEUJBL436FYEV7LHPDXPR2BPWQWJDNNT
CC: Wes Hardaker <wjhns1@hardakers.net>,
 =?UTF-8?B?T25kxZllaiBTdXLDvQ==?= <ondrej@sury.org>, dns-at-ietf@ietf.org
Precedence: list
Subject: =?utf-8?q?=5Bdns-at-ietf=5D_Re=3A_managing_workload_within_and_across_DNS_WG?=
	=?utf-8?q?s?=
Archived-At: 
 <https://mailarchive.ietf.org/arch/msg/dns-at-ietf/cyYcWh5edTLjt5LHCdP1Z2kv8Pc>

--000000000000bf47df064307b0f2
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Fri, Nov 7, 2025 at 3:54=E2=80=AFPM Petr =C5=A0pa=C4=8Dek <pspacek@isc.o=
rg> wrote:

> On 07. 11. 25 14:45, Wes Hardaker wrote:
> > Ond=C5=99ej Sur=C3=BD <ondrej@sury.org> writes:
> >> With groups large in content and people like dnsop, one of the problem
> >> that I have as a chair are the following indicators:
> >>
> >> - Is there enough interest in the draft?
> >> - Has there been enough reviews?
> >> - Has the consensus been reached?
> >
> > I think those are good helpful questions, thank you.  Do they work for
> > satisfying whether or not a draft is ready for last call though?  One o=
f
> > the load issues, IMHO, is not just can we take it on (interest and
> > enthusiasm are common) but also "was it actively worked on"?  A common
> > pattern with overloaded groups can be that Last Calls bring up a ton of
> > issues because people ignored it until that point.  How do you work
> > around this sort of problem?
>
> I think requiring implementations before it is allowed to progress
> through a suitable point in the process is a good filter. If authors are
> not committed they will not spend time fiddling with code.
>
> Ad the argument 'but that would create a group of gatekeepers':
>

On this point, I think I heard Warren already make the observation
in the OPSARA meeting  that in some other WGs, a small group of
influential implementers have been perceived to gate keep accepted
work. While on balance, I agree that we need implementations, we
need to avoid the situations that Warren cites. We need to balance
implementer interest, with operator interest, and IETF principles (e.g.
what is good for the Internet, not what is good for the business of a
specific set of implementers).

For me, one of the most important benefits of implementation work is
to prove out protocol designs, identify gaps in our understanding of
they would work in the field, and thus inform our work to develop
technically correct solutions.

Shumon.

--000000000000bf47df064307b0f2
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div dir=3D"ltr">On Fri, Nov 7, 2025 at 3:54=E2=80=AFPM Pe=
tr =C5=A0pa=C4=8Dek &lt;<a href=3D"mailto:pspacek@isc.org">pspacek@isc.org<=
/a>&gt; wrote:</div><div class=3D"gmail_quote gmail_quote_container"><block=
quote class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-left:1=
px solid rgb(204,204,204);padding-left:1ex">On 07. 11. 25 14:45, Wes Hardak=
er wrote:<br>
&gt; Ond=C5=99ej Sur=C3=BD &lt;<a href=3D"mailto:ondrej@sury.org" target=3D=
"_blank">ondrej@sury.org</a>&gt; writes:<br>
&gt;&gt; With groups large in content and people like dnsop, one of the pro=
blem<br>
&gt;&gt; that I have as a chair are the following indicators:<br>
&gt;&gt;<br>
&gt;&gt; - Is there enough interest in the draft?<br>
&gt;&gt; - Has there been enough reviews?<br>
&gt;&gt; - Has the consensus been reached?<br>
&gt; <br>
&gt; I think those are good helpful questions, thank you.=C2=A0 Do they wor=
k for<br>
&gt; satisfying whether or not a draft is ready for last call though?=C2=A0=
 One of<br>
&gt; the load issues, IMHO, is not just can we take it on (interest and<br>
&gt; enthusiasm are common) but also &quot;was it actively worked on&quot;?=
=C2=A0 A common<br>
&gt; pattern with overloaded groups can be that Last Calls bring up a ton o=
f<br>
&gt; issues because people ignored it until that point.=C2=A0 How do you wo=
rk<br>
&gt; around this sort of problem?<br>
<br>
I think requiring implementations before it is allowed to progress <br>
through a suitable point in the process is a good filter. If authors are <b=
r>
not committed they will not spend time fiddling with code.<br>
<br>
Ad the argument &#39;but that would create a group of gatekeepers&#39;:<br>=
</blockquote><div><br></div><div>On this point, I think I heard Warren alre=
ady make the observation</div><div>in the OPSARA meeting=C2=A0 that in some=
 other WGs, a small group of</div><div>influential implementers have been p=
erceived to gate keep=C2=A0accepted</div><div>work. While on balance, I agr=
ee that we need implementations, we=C2=A0</div><div>need to avoid the situa=
tions that Warren cites. We need to balance</div><div>implementer interest,=
 with operator interest, and IETF principles (e.g.</div><div>what is good f=
or the Internet, not what is good for the business of a</div><div>specific =
set of implementers).</div><div><br></div><div>For me, one of the most impo=
rtant benefits of implementation work is</div><div>to prove out protocol de=
signs, identify gaps in our understanding of</div><div>they would work in t=
he field, and thus inform our work to develop=C2=A0</div><div>technically c=
orrect solutions.</div><div><br></div><div>Shumon.</div><div><br></div><div=
><br></div></div></div>

--000000000000bf47df064307b0f2--