[ppml] IPv6 flawed?

On Mon, 17 Sep 2007, Kevin Kargel wrote:

> Ah, but every rule and law we have in our society is more important than
> the human.  The highway patrol officer doesn't care that there was
> nobody else on the road when you were going 85mph completely under
> control in a safe vehicle.. even if you did have a pressing need because
> you didn't organize your schedule efficiently..  Granted there are
> extenuating circumstances like life or limb, but I doubt being late for
> your job or working more efficiently would be accepted.

I think you misunderstand. These laws are created to protect the
human(s): The person and occupants driving at unsafely high speeds that
put themselves and others at risk of serious injury.  In the midwest,
highway speed speed limits _are_ higher. SD has 75mph, with a $5 fine
and no points for the first 5 mph over.  Montana has 80mph. These areas
of the midwest are flat, straight, are well maintained, have good
signage, and largely unpopulated with large safety margins. High speeds
there are not very dangerous to the driver or other occupants. By
contrast, the roads in say, New England, hardly have a straight section,
are often poorly maintained, have poor signage, and don't have safety
margins to contain accidents to the roadway.  New England roads are
barely safe at any speed.  Putting race tires on a BMW or Corvette
changes nothing.

The highway patrolman does not care that the _speeder_ thought it was
safe: The speeder isn't qualified to pass judgment on the safety of the
highway.  Others who were qualified, decided it was unsafe at those
speeds. And of course, speed limits were reduced in the 70's not for
safety (at least not primarilly for safety), but for fuel efficiency so
to reduce dependence on foreign oil. This is also a social policy that
benefits society.

Your example provides a good metaphor for the differences between
craft/operations and engineering for ISPs. Operations people sometimes
like to "speed", and because they don't immediately "crash", they assert
that it was therefore "safe", just like the guy in the 'vette.  But an
engineering person comes along, 'does the math', and like the civil
engineer for the highway, calculates that the practice is "unsafe", and
that a rule against that practice should be adopted.

It would be ridiculous for the speeder to say:

  "I'm an experienced driver, and I think it is safe at 85mph. I don't
  care what the civil engineer calculated, or policy required".

It would also be ridiculous for the operations people to say,
essentially:

  "I'm an experienced administrator and I believe differently, and I 
  don't need any facts to justify my belief. I'm going to tell other
  people it is safe, and hide the reasons that it isn't safe (with
  hardball tactics)."

It seems entirely ridiculous, but I have actually experienced operations
people really saying such things (at the IETF even!---Well, I should add
those were NANOG/Vixie-associated operations/craftspeople. That group
has a greater propensity for such statements.)

Of course, such incredible statements make more sense when you realize
they will make a great deal of money on the unreliable practice, or will
ingratiate themselves to their cronys (who can then act later in
conflict of interest).  But just like the speeder, they are still often
not qualified to make those decisions, regardless of how many "cars they
have operated". Belief and faith does not refute science and math.  
Belief motivated by money and cronyism is especially dubious.

Good social policy requires honesty and integrity, and solid technical
analysis of the impacts and cost/benefits.  People play nice until some
person or group figures out a way to screw the others and make a lot of
money for doing so.  No one plays "hardball" for the hell of it;  they
have motivation: money and cronyism.  Tort law, the Law of Agency, the
Law of Corporations, Associations and Groups, the Law of Trusts stands
against this anti-social behavior, and provides definitions of
obligations, proper conduct, misconduct, and legal procedures for
reparation and redress.  But there are often simpler ways to fix things.
That things need to be fixed is certain once you can identify people
playing hardball, or behaving dishonestly, or benefiting from a conflict
of interest. Cronyism by itself may not be that bad, but it is certain
to be bad when it is mixed with these other things, especially money.

> It would be great if we could make a policy that said "just be good and
> play nice with others" and everyone would do that..  but I don't really
> expect that to work.

You can work against those who don't have honesty and integrity; those
who play hardball, behave dishonestly, and benefit from conflict of
interest.  Resist corruption, even if you benefit from it; _especially_
if you benefit from it.  Ignoring corruption just allows things to get
worse. Be patient; don't expect results overnight. Shining a light onto
corruption has always, and will always, prevail eventually.

But policy implies policing, and that requires some introduction to law
and justice. Civil society can't work without some kind of coercion and
penalty. Laws are not suggestions.  But coercion and penalty must be
consistent with the laws and constitutions.  Coercion that is
inconsistent with laws and constitutions is unjust and is simply a form
of warlordism or gang rule; things that civil society must oppose as a
first order of business.

> BTW, I personally think we owe Legacy owners something.  Those were by
> and large the folks that jumped in and did the early experimenting,
> they spent the big chunks of R&D money when it was expensive and paved
> the way for the rest of us.  We are living in the results of their
> work and dreams and adventurousness.  I think they actually deserve a
> little extra consideration. So (imho) if they want to continue using
> the IP space that was granted them back when nobody else cared or
> wanted it I say "More power to em"..  If they are not using the space
> and see their way to return it I will think nicer things about them,
> but I won't think bad things if they want to keep what they have.

Thanks for the sentiment.  I think legacy holders always appreciate
recognition of their efforts. However, more significant than recognition
is that legacy's already have a license or lease agreement with the
government that ARIN isn't authorized to break. ARIN is just the latest
agent of the government, maintaining the government records that SRI and
Network Solutions previously maintained.

Let me give you another example: Many states have leased toll roads to
private companies. The private companies get to set and collect tolls,
and in return must maintain the roads. But they do not _own_ the roads,
and they cannot interrupt legacy buried or overhead cables which have a
previous permanent or unexpired right-of-way agreement.  The state
governments can put someone else in charge, or do it themselves, again.
Likewise, ARIN can be replaced, just like SRI and Netsol were previously
replaced.

		--Dean

-- 
Av8 Internet   Prepared to pay a premium for better service?
www.av8.net         faster, more reliable, better service
617 344 9000