One thing I get surprised often is the amount effort some people prefer
to put in repealing IPv6 than to deploy it.
Although I keep thinking that this proposal doesn't force anyone
automatically to do something, I fully agree with what was said by Owen
that "Business has a responsibility to the community(ies) in which it
operates and It has an obligation to function as a supportive member of
the community providing a general benefit to the community and not act
as a parasite consuming the community in question."
Also worth quoting John's comments that "ARIN's enforces your ability to
make productive use of any address space is predicated upon cooperation
with the same community, you might find it difficult to argue that you
wish the benefits of cooperation minus whatever obligations that
community collectively establishes."
Therefore I don't believe the question about deploying IPv6 or not is
merely a private question of any business for them to decide whenever
they like without consequences. If still they choose not to do then they
should have no right to complain of any obligations that community
wishes to establish.
Best regards
Fernando
On 15/11/2019 15:53, [email protected] wrote:
My own entry into the IPv6 world began with a mandate issued by the
Executive Office of the President, which mandated that after a magic
date that all Federal networks, and therefore those of their connected
contractors have the ability to use IPv6.
Back in 2008, this was not as easy as it seems today. While most of
the basic services like DNS, SMTP and HTTP(s) servers had the needed
support in the most recent versions of the software, a whole lot of
stuff ran on earlier versions, which began the rush to first get to
the latest version so that we can turn IPv6 on to meet the
requirement. Even Windows XP, the most recent version at that time of
Microsoft Workstation Software had to have IPv6 added, as it was not
enabled by default.
The people I worked with had put IPv6 in their RFPs for a few years,
so there was not as much hardware requiring a forklift upgrade.
Today, almost anything more than a few years old has IPv6 enabled by
default. It is almost unwise to have it turned off, as it can be used
as a bypass around your security that you do not even think of,
because of sayings like "This is an IPv4 ONLY shop", does not mean it
is not present, even just locally on your lan bypassing any
restrictions that you may have.
At some point in the hopefully not too distant future, IPv6 will
become the main protocol on the Internet. In a lot of residences it is
already there with many major sites and large ISPs already having it
in place.
I see this draft policy as more of a nudge than a push in the right
direction. As pointed out earlier, one of the main uses of this
policy is to direct enterprises that wish to expand their IPv4
inventory to adopt IPv6, at least in the limited extent of those who
interact with ARIN. Often this is an IS/IT group who should be able to
have IPv6 working in less than a day, or can assign an intern to do
it. Without such a requirement, as pointed out by others, they will
NEVER move.
Even with this policy, businesses that see no need to have IPv6 will
still not have to face this policy unless they are obtaining more IPv4
addresses.
Maybe we would have been better off had China had used up the
remaining address space much earlier than 2011 for its academic
networks, instead of going to IPv6. It would have forced these issues
years earlier.
It is the right thing for anyone part of the Internet to have IPv6.
CIDR, NAT, CGnat and whatever the next band aid is not going to
stretch IPv4 forever, and the complications of all these devices,
instead of an end to end connection is going to make things harder for
those that remain on IPv4 and all the extra hardware to share the
limited address space.
I think the time is right now to step forward and take a stand. If you
want to increse your percentage of your IPv4 holdings, you need to be
taking steps to also be ready for the future. That future is IPv6.
The policy is needed, and even the ARIN Board has agreed that anyone
with a need for numbers need to consider a move to IPv6.
Albert Erdmann
Network Administrator
Paradise On Line Inc.
On Fri, 15 Nov 2019, Owen DeLong wrote:
On Nov 14, 2019, at 20:14 , Michel Py
<[email protected]> wrote:
Hi Owen,
Owen DeLong wrote :
You seem to be assuming he’s in the internet business. He made it
pretty clear he’s talking from
the enterprise perspective where the internet isn’t the revenue
generating portion of the business,
but merely one of the many tools used by the business to accomplish
its revenue goals.
Indeed. The Internet is not the same thing as the Internet business.
There is no Internet without customers. Some customers, such as the
mobile market and the low-end residential market can be forced into
IPv6 because they control nothing, but the enterprise market is not
subject to this. The enterprise adoption is a trickle, for reasons I
have explained publicly for years.
I'm sorry to say it bluntly, but the enterprise business is about
making money, not saving the world from an impending doom that has
not happened. I say it again : the problem of IPv6 is that it is a
solution to a problem that I do not have.
Well… More accurately, it is a solution to a problem that you feel it
is better to live with than to solve. In short, you feel that the
barriers to implementing the cure are worse than living with the
symptoms of the disease.
Like or to, the need for NAT is a problem at least most enterprises
have. The fact that we have an entire generation of engineers who
have grown up not understanding the advantages of end-to-end
addressing (and don’t understand that stateful inspection is a
dependency for NAT, but can be implemented without header mutilation)
further complicates the recognition of this problem, but you and I
are both old enough to remember an IPv4 internet with transparent
addressing and the benefits thereof.
Making matters worse, enterprises failing to deploy IPv6 enjoy all of
the advantages of the toxic polluter business model. The costs of
their refusal to move forward with the rest of the internet are borne
not by those making said refusal, but pushed off on those sharing the
internet with them who cannot complete their transitions so long as
there is a critical mass of enterprises holding back progress.
6 years ago, you thought that I was full of it. We had a couple
beers and you respectfully dismissed me as an IPv4-only crackpot.
Oh, I still think you’re full of it to some extent. I don’t think I
dismissed you as a crackpot so much as we respectfully agreed to
disagree on several areas. I think little has changed in the
intervening years.
With you, I do not seek revenge. You are a formidable opponent and I
respect you as such, but look back in the past.
From my perspective, there is nothing to seek revenge for. I don’t
see you as an opponent so much as just someone with a differing
opinion and operating on a different time line.
You feel that the self-serving advantages of delaying IPv6 deployment
in your environment outweigh the broader public interest advantages
of proceeding to a point where IPv4 deprecation can begin. From a
purely Ayn Rand/Gordon Gecko oriented perspective, this is one
available philosophy. It’s the same mentality that will likely lead
to human extinction through global climate change… It’s the attitude
that a business should first and foremost maximize profit above any
other concern.
It’s not a philosophy I embrace. Does a business have an obligation
to make a profit? Certainly. Does a business have other duties
besides maximizing profit? IMHO, yes. IMHO, a business has a
responsibility to the community(ies) in which it operates. It has an
obligation not to dump toxins into the local rivers for those living
downstream to deal with. It has an obligation not to partially
offload the payment of its employees onto the taxpayers (a la a
certain large well known chain of stores). It has an obligation to
function as a supportive member of the community providing a general
benefit to the community and not act as a parasite consuming the
community in question.
6 years ago, when we shared a couple beers on stage. I told you so.
You did not listen. You were wrong.
This will likely not surprise you, but I disagree. Even then, I
agreed that enterprises would likely be the last class of laggards
procrastinating the deployment of IPv6.
You say that this procrastination will likely continue indefinitely.
I feel that its days are numbered. Not as short as I’d like to see,
but I believe sooner than you expect.
I disagree with Michel in a number of areas. He and I have had
frank discussions about this.
However, the points he raises are legitimate and we ignore or
dismiss them at our peril.
I am glad you realize the peril part of it. 6 years ago, you never
thought I would be challenging you publicly on this. 6 years ago,
you would not even have considered the possibility that we would
have this talk on this mailing list.
Actually, I am not at all surprised to see you still publicly
challenging me on this. I may not have predicted 6 years ago that it
would be you, but I fully expected some contingent of IPv6 opponents
would still exist in the enterprise realm and that this discussion
would still be ongoing.
There is one more thing you should realize about enterprise business
: they like people who have been steady in predicting the future.
I'm on track.
Meh… I’m fond of the saying that Prior Performance does NOT guarantee
Future Results.
The nice thing about the enterprise world is that no enterprise is
for ever and new ones come to life every day. There is a time coming
in the not too distant future where deploying a new enterprise
without IPv6 will seem as silly as deploying one without IPv4 today.
At that point, then it’s just a matter of time before a combination
of ever increasing quantities of new enterprises combined with
attrition of old ones shifts the dynamic.
Things may move slower that many of us would like because of the drag
induced by people who share your mindset, but, nonetheless, time is
on the side of those of us who believe IPv6 will eventually replace
most of the current IPv4 utilization on the internet.
Eventually (assuming we manage not to go extinct due to climate
change in the meantime), we will get there. The question is will we
ever realize the wisdom of ripping off the bandage, or, will we
continue to peel at the edges making a slightly lower level of pain
last for a much much longer time period. Personally, I prefer a
shorter period of slightly more significant disruption. You obviously
prefer to endure a prolonged period of pain (or denial about pain in
your case).
Owen
_______________________________________________
ARIN-PPML
You are receiving this message because you are subscribed to
the ARIN Public Policy Mailing List ([email protected]).
Unsubscribe or manage your mailing list subscription at:
https://lists.arin.net/mailman/listinfo/arin-ppml
Please contact [email protected] if you experience any issues.
_______________________________________________
ARIN-PPML
You are receiving this message because you are subscribed to
the ARIN Public Policy Mailing List ([email protected]).
Unsubscribe or manage your mailing list subscription at:
https://lists.arin.net/mailman/listinfo/arin-ppml
Please contact [email protected] if you experience any issues.
_______________________________________________
ARIN-PPML
You are receiving this message because you are subscribed to
the ARIN Public Policy Mailing List ([email protected]).
Unsubscribe or manage your mailing list subscription at:
https://lists.arin.net/mailman/listinfo/arin-ppml
Please contact [email protected] if you experience any issues.
