Good Morning,
please excuse my belated reply to the comments of the the last days. I
will also reply to the contents of the other two messages later today.

However, I also believe that we need to talk about an elephant in the
room concerning the two concerns voiced about missing consideration of
Geoff's two reviews. Please understand that I make below comments as an
individual to further contextualize these concerns. Opinions and
conclusions stated are my own and have not been coordinated with my co-
editors, chairs, or our area director.

> I am not a DNS expert, but I have to agree with Paul on his DISCUSS
> on seeming contradictions in the recommendations. Much like Paul, I
> am intrigued why there has been no response beyond acknowledgement of
> Geoff Houston’s two DNSDIR reviews.

After WGLC, in which Geoff did not participate, Geoff was assigned as a
reviewer by himself in his role as DNSDIR director and conducted a
DNSDIR review of the document:

https://mailarchive.ietf.org/arch/msg/dnsop/25qhHg96-Sw3otx-B6ANM7xAPTY/

This first review received a detailed response by us, including five
PRs and several follow up questions, especially concerning points where
text contributed to the discussion in response to the reviewers earlier
requests in the WG was concerned:

https://mailarchive.ietf.org/arch/msg/dnsop/uwk2YrprOAv_5HU8vso5smsHr14/

There was no further feedback on the document. However, on only
acknowledged points, a WG member expressed their agreement to the
feedback provided to the review, noting again that this is the result
of WG consensus.

https://mailarchive.ietf.org/arch/msg/dnsop/R8QDGT1P_GH4ORKQPqgUtVqCfYI/

Geoff, in return, noted that he would not engage in the conversation on
the contents of the provided review. 

https://mailarchive.ietf.org/arch/msg/dnsop/BZfdSHnpF-DSTeTA-Zd29yOtxCM/

He also notes that:

"""
It appears to me that _as an Informational RFC_, this draft is as ready
as it will ever be, and its a fine document. As a BCP, however, it is
Not Ready in my view, as it strays beyond the role of a BCP.
"""

This, at least to me, creates the impression that the contents are, at
that point, ready. Instead, only a meta-question on the nature of the
document remains, which needs to be discussed and resolved.

These concerns were then acknowledged, clarifying the role of BCPs:

https://mailarchive.ietf.org/arch/msg/dnsop/e8j4ZUkXQMO7jEWzM20IUo0m7d0/

However, this clarification seems to not have been convincing in the
context of the discussion:

https://mailarchive.ietf.org/arch/msg/dnsop/4eduaYxsV89Y78z7DTk92PxBozU/

Which, in turn lead to further clarification:

https://mailarchive.ietf.org/arch/msg/dnsop/pXfyuyJ-7NrM9X-nbVI8L97IC9k/

After this clarification, the discussion on this first review stopped.
As no further feedback from the reviewer was received on the PRs, these
have been merged and are present in -10.



When the document progressed further, a telechat review was added:

https://mailarchive.ietf.org/arch/msg/dnsop/50QWxwtDE2fG6yygbs5ROy36GDw/

This review reiterated some points from the earlier DNSDIR IETF-LC
review, i.e., apparently points were not addressed to the reviewer's
satisfaction. Again, a response, along with three PRs, was provided:

https://mailarchive.ietf.org/arch/msg/dnsop/sp57Z4CBzcOkBHbgCcFaqvFzhRk/

Engagement to this response was, again, limited, in the form of a
request to post the diffs to the mailinglist as well:

https://mailarchive.ietf.org/arch/msg/dnsop/JFU_CCpjyD5vHsI0EI8preNQb6Q/

After posting the diffs, no further engagement was seen, and as such,
the PRs have not yet been merged.



With this turn of events, it personally strikes me as somewhat odd, to
now--for the third time--hear that there was no engagement with the
provided reviews from our side.



Given the history of the document, I am concerned that there may be
more than technical contents at play here, leaving me uncertain whether
the discussion always runs within the bounds of the note-well. While I
personally prefer not to stray to those matters, at this point, I think
that it might be beneficial to the process if we at least openly
discuss them.

As stated before, there seems to be a very specific perspective on the
merits of the draft. This has been accompanying the document from the
first time it has been presented, where it was met with an ad hominem
to start:

https://youtu.be/AhLrHoFL2lw?t=6025

The individual nature of the objection was also recognized by others,
see, e.g.:

"""
Geoff Houston was in not favour of the proposal, but found himself in
the minority. After IETF 118, Geoff therefore wrote a blog on the
subject.
"""

https://www.sidnlabs.nl/downloads/iMiufCFWoTVkikn1NrUSk/35fd7436c731e2bd9f9f0e7e62e5b015/CENTR_Report_on_IETF118.pd

Furthermore, there was a set of blog posts engaging with the topic, and
why the editors are wrong, while the document was seemingly not
included in the DNSOP summary of IETF118 posted in the same series:

https://www.potaroo.net/ispcol/2023-11/dns-ipv6.html
https://www.potaroo.net/ispcol/2023-11/dns-ietf118.html
https://www.potaroo.net/ispcol/2024-02/truncation.html
https://www.potaroo.net/ispcol/2024-03/truncation-v6.html
https://www.potaroo.net/ispcol/2024-07/truncation.html

Beyond that, during that time, there were also several talks focusing
on, in my opinion, ridiculing the editors of this document, see for
example:

https://www.youtube.com/watch?v=7vhm1ZD3xPY

00:09: "by now we've thrown facts out of the window and gone for 
        religious zealotry"
00:46: "[...] 'someone' is proposing in the IETF to actually reverse 
       that and actually say 'here is some v6 guidelines use it take it
       seriously we recommend that you put v6 absolutely everywhere and
       the real question in my mind is 'is this sane advice or are you 
       demonstrating a fundamental ignorance of how shocking is v6 over
       actually doing the bits of DNS that matter' because the problem 
       is v6 really does have a problem."
07:32: " 'Let's go for an experiment and do v6 only says some Zealot 
       out there' and it's kind of 'do you really hate your 
       customers?' "
14:04: "Let's change the DNS completely. I like your drugs. It's not 
       reality. But i love your drugs."

While this person-focused discussion was ongoing, there was a brief
flare of discussion on the WG mailing list, leading to, among other
things, the inclusion of text initially requested by Geoff, to which he
then subsequently objected in the review, see, e.g.,:

https://mailarchive.ietf.org/arch/msg/dnsop/R8QDGT1P_GH4ORKQPqgUtVqCfYI/

After that, the document was effectively stale, and it did not get to a
point where a call for working group adoption was scheduled.

In the meantime, the editors designed and executed experiments to
better understand the voiced claims on challenges with IPv6 and the
negative impact of fragmentation.

This work has, by now, been published: 

https://dl.acm.org/doi/10.1145/3730567.3764439

After data from this work was presented to the WG, the discussion
around the draft also finally progressed, and a call for adoption was
scheduled.

During the adoption call, Geoff took a strong stance against the
document on principle again. However, the chairs ultimately concluded
on rough consensus having been achieved, leading to adoption.

During the WG discussion phase, Geoff then did not deeply engage with
the document and in the discussions around it.



With all of this said, I find myself in a difficult position to always
find charitable and constructive interpretations for observed events. I
understand that this is likely easier from a different perspective. 

However, I am, personally, convinced that the changes we implemented go
beyond merely acknowledging feedback from the two reviews, i.e., we
already merged five PRs for -10 based on the first review, with three
more pending--again missing--feedback.

Of course, we would still be very much happy to integrate further
feedback, if we receive it in response to on our posted proposed
changes.



I hope that this summary sufficiently contextualizes the circumstances
around the reviews, and why--despite us having implemented significant
changes--an impression of us not engaging with them may have emerged.

If you have any further questions on this, I would be very much happy
to further discuss them.

With best regards,
Tobias

-- 
Dr.-Ing. Tobias Fiebig
T +31 616 80 98 99
M [email protected]

_______________________________________________
DNSOP mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to