Re: I-D ACTION:draft-ietf-dnsop-inaddr-required-04.txt

[Dean Anderson <dean@av8.com>]

> > I don't seen an ad hominem in the statement you quote.
>
> Loaded terminology (eg, "dictate") while speculating presumptively
> about the motives and intentions of the draft's author.

It's hardly "presumtive speculation". It's in the title.

Whether RFC's dictate the policies of registries could be an open
question, I think. The RFC's that established the current policy predate
the current registries.  This draft modifies the RFC's that were the basis
for the current policies.  Clearly, it is the intent of the authors to
change this policy, as well as the behavior of implementors, and the
policies of the registries.  The question of changing this policy, which
is now embedded in the policies of multiple organizations, each with their
own procedures, is a valid question for this draft.

If RFC's can't change these policies, then this Draft (and the proposal
for removing Reverse DNS) is beyond the authority of this working group,
and perhaps beyond the authority of the IETF.  (Though I think, at least
with respect to IPV6, it isn't.)  If it is beyond the authority of the
working group, it is a mute question.

> See message dated 27 Mar 2003 18:40:05 (terms like "culprit", "mislead",
> and so forth while talking about another person posting to the same
> message thread) for a more obvious example.

As co-chair, you should know the question "Is reverse optional?", is
answerable by reference to RFCs, and it not vague. The question "Is
anything broken without reverse?" is also answerable by reference to RFCs,
because it follows from the fact that reverse is optional. These are
questions of fact, not opinion.

From the message you refer to:

  Dean Anderson writes:
  That would be fine, except that some people like Vixie and others tend
  to write software that puts reverse map entries in log files. Then
  consultants go in and try to figure out who rooted a server, and find
  bogus reverse entries, and no IP addresses.  The culprit gets away
  because someone told people (or implemented it directly) that they can
  use reverse as a way to determine the relationship between the network
  operator and the host operator.  They also mislead people about the use
  of reverse, so that your assertion (according to them, anyway) that "no
  one is forcing [me] to populate reverse maps" is false.  Vixie again
  repeats the assertion that not having reverse is a sign something is
  broken, in his last email on the subject.


This is a true statement. Paul Vixie and Rob Austein (Rob has made the
same claims in the past), have asserted that something is broken if
reverse doesn't work. Paul recently posted a message dated 27 Mar 2003
18:28:52 +0000 to DNSOP stating this misinformation again.  Rob Austein
posted to this same effect 1 Oct 1986 (yes, a long time ago).  Both
yourself and Paul Vixie are senior enough to know that in fact Reverse DNS
is optional, and that (also in fact) nothing is broken (according to RFC)
if reverse doesn't work.

If you state as fact something that is false, you are misleading people.
It is misleading, even if you think it is true, but it is not. My use of
the word "misleading" is accurate.

Furthermore, stating as fact something that you know or can be expected to
know is false, is "intentionally misleading". I did not use the phrase
"intentionally misleading", in order to soften the effect. However, this
phrase is clearly also accurate and applicable in this case. Vixie's
recent claim is "intentionally misleading".

Not only did I avoid exaggerated "loaded terms", I have softened the terms
I am using.

The word "culprit" above refers to someone who has committed some abuse
where logs are used to identify the abuser. It is an appropriate use of
the term.  Perhaps you have misread the paragraph.

Also, "loaded terms" may be exaggerations, but they are not "ad hominems".
You are misusing the term, if that's your objection to my message. I think
I've shown that I have not exaggerated anything.

> > I presume I am allowed to disagree with others. Especially, since I
> > am not alone in the position.
>
> Yes, you're allowed to disagree.  What you're not allowed to do is to
> post essentially the same comment over and over again, or attack the
> people proposing an idea that you happen to disagree with.

Good. I have done neither of these things. Perhaps you could chastise
those that have: Randy Bush, and Brad Knowles, in just the last few days.

> > Further, it seems to be an attempt to renew a debate that we've been
> > having for 10 days. You should chastise the author who brought up
> > this old topic, not the person who is pointing out that it is an old
> > topic.
>
> Yes, the fact that you seem to be construing this as a continuation of
> the last ten days worth of in-addr discussion is part of the reason
> that I asked you to cease and desist.  Several people (including both
> WG co-chairs) asked the draft's author to bring the draft back from
> oblivion for discussion.  Part of the point of the draft is to say
> that applications that break without the reverse tree should be fixed.

The previous discussion has been about discontinuing reverse. It has been
argued and that nothing breaks without reverse, and that there are
alternatives for the convenience functions.  You (as co-chair) privately
asked that this topic stop.  Subsequently asking the for the revival of
this draft "from oblivion" is troubling.  And you asked my to stop
pointing out the flaws in this draft, which are unique to the draft. The
draft is, at best, a new angle to an old discussion. That is an
inappropriate use of your authority to bias the discussion.  Apparently,
there is no point in discussing this draft. It sounds like the co-chairs
have already approved it, despite many senior persons arguing convincingly
for the removal of Reverse.  Where is the consensus?

Indeed, I seem to be justified in my construing it as part of the old
topic:

  George Michelson writes: Wed, 19 Mar 2003 10:48:00 +1000
  ok. so you mean "any thread with draft-ietf-dnsop-inaddr-required-00.txt"

  I'd say its hung jury. Time for a re-trial.

I count a number of people for discontinuing Reverse, and making
substantial technical contributions:

  George Michelson
  Kevin Darcy
  Shan Kerr
  JINMEI Tatuya

I think others can be convinced with a demonstration of traceroute that
doesn't use reverse, but uses ICMP node identification or other
alternatives.

There are more people on Namedroppers, which has also discussed removing
Reverse.  Articulate arguments have been made, making the case that
Reverse should be removed from IPV6.  Many of the technical arguments
presented haven't been addressed by the proponents of reverse.  There is
much work to do in IPV6 before Reverse can really be used, and technical
obstacles to overcome.  There are substantial, credible problems for IPV6
that have been ignored by the proponents, and apparently, the co-chairs.
Not only do applications have to be fixed, protocols have to be fixed, to
keep reverse in IPV6.

It seems easier to fix the few applications like traceroute, to use
alternatives, than to make the changes for IPV6.

> > However, if it were a new issue, I should be allowed to state my
> > disagreements to it, and to help clarify its meaning and implications.
>
> As I said in my previous note: if you have something genuinely new to
> say, go ahead, but please don't repeat the same "we should just get
> rid of the reverse tree" comment you've been making for the last week
> and a half, and please let other people have their say without
> stomping all over them the way that you and Brad did to each other the
> last time around.

I have not made any repeating comments.  People have raised a number of
unique objections to the removal of Reverse.  I have pointed out flaws in
each of those objections.  In other cases, people have posted questions
about our asserted harms and other claims, and our position has been
clarified.

However, certain other people (for example, Brad Knowles) have made
repeatedly the same post. For example, Randy Bush has made several
comments that have no technical merit whatsoever. In fact, he has not made
any technical contribution to the discussion whatsoever. All of his posts
on the topic have been totally inappropriate ad hominems, sometimes
containing foul language.

The term 'Ad hominem" means an attack on the person, as opposed to the
arguments made by the person. I have been subjected to ad hominem attacks
by Randy Bush, and by Brad Knowles.  By contrast, I have not made any
personal attacks, except to complain about their ad hominem attacks.
Pointing out the flaw in a argument is an attack on an idea, not an "ad
hominem".  Attacking an idea is not a personal attack.

This is an example of an Ad hominem:

  Brad Knowles writes:  Mon, 24 Mar 2003 23:02:34 +0100
        Tell me something.  How many decades have you been doing DNS?
  How many decades have you been doing e-mail?  Of those decades, how
  much of that work has been at the highest volume sites in the world,
  where millions and billions of spams are dealt with on a daily basis?

        Unless you have more experience doing the DNS than anyone in the
  world, and more experience doing e-mail and dealing with spam than
  anyone else in the world, and you can call on this experience in
  making proclamations about the kind of spam that other people might
  possibly be seeing and how the DNS might or might not be involved,
  you really don't have a leg to stand on.

        Personally, since your first name isn't Paul, or Eric, or some of
  the various other names I am familiar with, I don't think there's
  much chance of this.

This is another example of an ad hominem:

  Randy Bush writes:  Mon, 24 Mar 2003 17:47:41 -0800
  i will send you each a bona fide "You Are Right" certificate if you
  will both shut the hell up.  you are one message from my .procmailrc.
  dean is a long-time entry.

  randy

This is another example of an ad hominem:

  Randy Bush writes: Tue, 1 Apr 2003 08:39:00 -0800
  cool!  i did not realize rfcs had such power.  i will now write
  some rfcs to
    o end world hunger,
    o end all war, and
    o let everyone eat cake

  <sheesh>!

  randy

They are ad hominems because they attack me personally. They have nothing
to do with any of the arguments I have made. They offer no
counter-examples, or otherwise refute anything I said. They are attacks on
my person: to ridicule me, or suggest I don't have enough experience to be
able to read RFC's or analyze problems. They don't point out any flaws in
facts or analysis I present. Instead, they try to point out personal
flaws.

> > Do we make judgements now without debating the issues? Or am I only
> > allowed to comment on one topic?
>
> No, and no, but please let other people have their say too, without
> chasing folks away from the list.

I have not chased anyone away. I have not ridiculed anyone, or told anyone
they shouldn't post.  You seem to be chasing me away, or at least telling
me and others that we shouldn't point out the flaws in proposals promoting
reverse, and changing current policy on reverse that dates back to 1986.
And also that we should ignore the problems presented by Reverse to IPV6.
Ignoring these problems threaten IPV6 deployment.

> I'm not that fond of the "debate" model for WG operations, it's much
> too prone to nonterminating loops in which each participant feels that
> he must have the last word.

Having the last word isn't debate.  But not having debate, as you propose,
is worse.  Only biased proposals are discussed by their proponents.
Everyone who disagrees has been told to "shut up".  This is not
productive.

> What seems to work better is to write a draft, give everybody a chance
> to state his or her view of whether the draft is on the right track and
> how it might be improved (preferably based on experience with running
> code), then see whether there's rough consensus on what to do next with
> the draft.

That is a debate. To make progress, it is important the chairs organize
the debate so that facts are presented, and decisions are made logically,
rather than in reference to personalities.  Otherwise the organization
just stamps the positions of the chair. This eventually leads to a lack of
participation, and to poor decision-making.  As the chair, you have
responsibilities to make the decisions of the organization robust. Failing
these responsibilities sabotages the goals of the working group.

Instead of using your authority to bias discussion around proposals you
personally support, you should chastise those making ad hominem attacks,
or otherwise unprofessional posts, to keep discussion on a professional
level, and work to clarify arguments and objections, so that rational
decisions can be made.

		--Dean

#----------------------------------------------------------------------
# To unsubscribe, send a message to <dnsop-request@cafax.se>.