[arin-ppml] Revised - ARIN-2023-8: Reduce 4.1.8 Maximum Allocation

Wed Feb 21 10:46:23 EST 2024

On Wed, 21 Feb 2024, 12:30 Owen DeLong, <owen at delong.com> wrote:

>
>
> On Feb 21, 2024, at 07:20, Fernando Frediani <fhfrediani at gmail.com> wrote:
>
> 
>
> Hi
>
> This rather seems to be a vague assumption as you didn't provide anything
> substantial for it to be a blocker to have a policy adjusted in order to
> contemplate only new entrants.
> Why is it bad ? Do you think it is still rational to keep supplying IP
> addresses to those who already have some in detriment to those who have
> nothing ?
>
> I think any legitimate use of IPv4 addresses is no more or less worthy
> than any other. I see no reason to elevate theoretical new entrants to the
> point of depriving existing legitimate users.
>

Oh yes, those who have already can never make a better usage of what they
already have and really justify for more IP addresses (that don't exist) in
detriment of newcomers ?
There are countless ways to always better use of what one already has and
it sounds very unreasonable to continue assigning more addresses to these
organizations in times of exahustion. Need to balance things correctly,
face reality and be reasonable given the current scenario.

>
> IPv4 is an obsolete technology. Preserving an IPv4 free pool against
> legitimate demand to facilitate latecomers and laggards failure to deploy
> IPv6 is simply not in the overall best interests of the internet.
>

Well, that's another discussion. Newcomers don't have any and cannot do
anything without a minimal IPv4 even if they prefectly deploy IPv6.
Trying to force things only towards IPv6 ignoring the practical side sounds
more like ideology.

> This is not unenforceable and just a supposition unsupported by real data.
> ARIN has means to develop ways to check these newer organizations and
> separate the possible fraudsters from the legit ones. Just before there it
> serves to inhibit a lot of organization to even request IPs under the
> waitlist making it much cleaner and fair. LACNIC has been doing it for
> years and it has proven to be successful in terms of fairness and
> possibility to check these organization requests correctly. Are we going to
> avoid having a policy which is the right thing to do just on the
> supposition that there will be fraud ?
>
>
> While I stayed that the process in question was morally equivalent to
> fraud, it is 100% legal and utterly indistinguishable from a legitimate new
> entrant.
>
> The policy you are proposing is not only the wrong thing to do (see
> above), it is also quite trivially worked around. One can legitimately spin
> up an organization for a few hundred dollars and a few hours of work.
>

That's a theory you have without knowing ARIN tools and possibilities.
Again, it doesn't matter much the theoretical possibilities. There will
always be unlimited. What matters most is to have as a policy what is
correct, fair and in the interest of community.

It is the interest of community that more organizarions and consequently
users can connect to the Internet, develop new business and make up new
technology and it is fair to think to make things to keep fitting newcomers
to this industry.

Fernando

>
> ARIN can prevent the recording of that organization’s subsequent
> acquisition in the ARIN database, but that’s about all that ARIN can do.
>
> Owen
>
> Fernando
> On 21/02/2024 04:13, Owen DeLong via ARIN-PPML wrote:
>
> Anyone using IP to conduct business should recognize that IPv4 is out and
> they’ll need IPv6 to do business going forward.
>
> I oppose Fernando’s idea that the waitlist should be limited to new
> entrants. In addition to being bad policy, this is completely unenforceable
> and only leads to widespread workarounds (which are morally equivalent to
> fraud but probably don’t quite fit the legal definition of the term). (The
> cost to spin up an organization to acquire resources and then acquire the
> organization is trivial compared to the value of the IPv4 resources
> obtained).
>
> Owen
>
> On Feb 20, 2024, at 19:28, Denis Motova <dmotova at brcrude.com>
> <dmotova at brcrude.com> wrote:
>
> Owen:
>
> I appreciate your thoughtful and constructive suggestion.
>
> There are a couple of factors at play here that I'd like to address
> directly, if possible:
>
> Regarding the Existing Waiting List - I'm uncertain about the rationale
> behind altering the current waiting list and applying new criteria to
> members who have already been approved. I believe any new policy should not
> retroactively affect those who have already undergone approval. Approved
> members should continue to receive the resources they were initially
> granted based on their justification until such point as new users are
> added under the new policy (after its approval) and its updated
> distribution methods are implemented.
>
> As for the New Policy for Future Applicants - Future applicants may be
> required to select from a /22, /23, or /24 allocation, with the decision
> weighted based on the considerations Owen has mentioned regarding the
> allocation of new resources.
>
> I support the sentiments expressed by Fernando Frediani; there should be a
> reasonable approach that balances the need to avoid impacting the size of
> routing tables while still providing users with the flexibility they
> require to conduct business rather than treating IPs as a hobby.
>
> Thanks again,
> Denis
>
>
> On 20 Feb 2024, at 21:53, Owen DeLong <owen at delong.com> <owen at delong.com>
> wrote:
>
> How about this:
>
> Each waitlist recipient specifies a desired block size and a minimum
> acceptable block size. Wait list recipients can change their minimum
> acceptable block size at any time so long as it is no shorter than their
> originally approved block size.
>
> When ARIN receives a block to fulfill a waitlist request, the first
> waitlister in line with a minimum acceptable block size ≥ the available
> block size gets it.
>
> In other words, let’s say we have the following waitlist:
>
> Party Approved Minimum acceptable
> A /23 /23
> B /22 /23
> C /22 /24
> D /24 /24
> E /22 /23
> F /22 /24
>
>
> Let’s say ARIN receives a /24. The first /24 would go to party C.
> If ARIN then received another /24, it would go to party D.
> If ARIN then received a /22, Parties A and B would receive a /23 each.
>
> Owen
>
>
> On Feb 16, 2024, at 17:01, Denis Motova <dmotova at brcrude.com>
> <dmotova at brcrude.com> wrote:
>
> Dear Scott,
>
> I appreciate the innovative perspective and thorough thought process
> you've articulated in your email.
>
> There are a couple of points I'd like to highlight:
>
> The new policy shouldn’t be retroactive, it should be only a policy going
> forward. I mention it only because I think it’s important to make that
> distinction clear.
>
> Secondly, I find your proposed approach in the second paragraph
> intriguing. It's far more nuanced than simply restricting everyone to a
> maximum of a /24. I believe you're onto something promising here, and it
> could serve as a sensible strategy moving forward.
>
> Regarding the issue of "time," it's important to acknowledge the existence
> of a secondary market for IPs. If there's significant pressure, purchasing
> IPs should be considered a viable option rather than solely relying on
> expedited access through the waiting list. Maintaining a balance is key;
> those with urgent needs can acquire IPs through purchase, while others can
> join the waiting list and adhere to the traditional process. Personally, I
> believe this approach strikes a fair and equitable balance.
>
> -Denis
>
>
>
> On 16 Feb 2024, at 21:14, Scott Leibrand <scottleibrand at gmail.com>
> <scottleibrand at gmail.com> wrote:
>
> The point isn't to "improve the visual appearance of the waiting list
> numbers". Everyone knows the free pool is empty except for the reclaimed
> dregs, and we're deciding who should get how much of the dregs. The point
> of this proposal, limiting the maximum allocation to /24, is to allocate
> smaller netblocks to organizations that have been waiting a shorter amount
> of time, instead of making everyone wait longer while those with a
> non-time-sensitive justification for a larger block can get one and those
> who only need a smaller block wait in line longer.
>
> Another alternative to limiting everyone to a /24 would be to prioritize
> the waitlist such that everyone's place in line is determined by how long
> they've been waiting divided by how many /24s they're requesting. So at any
> given time, we might be fulfilling /24 requests that have been waiting 6
> months, /23 requests that have been waiting a year, and /22 requests that
> have been waiting 2 years. (Or 1, 2, and 4 years, respectively.) That way
> no one is penalized for accepting a smaller block, and an organization who
> can usefully use a /24 now and a /24 later gets a /23 worth of space in the
> same amount of time as someone holding out for a contiguous /23.
>
> -Scott
>
> On Fri, Feb 16, 2024 at 12:56 PM Denis Motova <dmotova at brcrude.com> wrote:
>
>> Dear William,
>>
>> I appreciate your message and your input.
>>
>> I have some reservations about agreeing with the statement you made, and
>> I'll explain my reasoning below:
>>
>> I strongly believe that there are numerous legitimate businesses
>> currently on the waiting list seeking IP space allocations of /22, /23, and
>> /24. By removing the option for these allocations, we essentially transform
>> the waiting list into what you described in a previous post as catering to
>> "hobbyists and speculators." It's unlikely that any serious company would
>> require only 256 IPs within a network; that's essentially a micro-network.
>>
>> As you are aware, there are multiple avenues for obtaining IP space,
>> including the waiting list and authorized purchase methods. From my
>> perspective, if a business urgently needs IP space, they would likely
>> follow the example of AWS and invest in acquiring the necessary resources
>> rather than wait through the waiting list.
>>
>> For instance, one of our customers acquired a /17 by purchasing it from
>> the market after providing justifications to ARIN for the IP space. While
>> this involved a significant financial investment, it demonstrated the
>> seriousness of their business needs.
>>
>> I fail to see the value in limiting everyone's network size solely to
>> improve the visual appearance of the waiting list numbers.
>>
>> Thank you once again for your collaborative spirit and feedback.
>>
>> Sincerely,
>> Denis
>>
>>
>> On 16 Feb 2024, at 15:52, William Herrin <bill at herrin.us> wrote:
>>
>> On Fri, Feb 16, 2024 at 8:52 AM Denis Motova <dmotova at brcrude.com> wrote:
>>
>> A. Decreasing the allocation to a /24 means that new allocation
>> holders would receive a minuscule network, hardly sufficient for
>> small to mid-sized deployments.
>>
>>
>> Hi Denis,
>>
>> At this point, the wait list is for hobbyists and speculators: people
>> who can afford to wait, which a serious business cannot.
>>
>> Tell me I'm wrong.
>>
>> Regards,
>> Bill Herrin
>>
>>
>> --
>> William Herrin
>> bill at herrin.us
>> https://bill.herrin.us/
>>
>>
>> _______________________________________________
> ARIN-PPML
> You are receiving this message because you are subscribed to
> the ARIN Public Policy Mailing List (ARIN-PPML at arin.net).
> Unsubscribe or manage your mailing list subscription at:
> https://lists.arin.net/mailman/listinfo/arin-ppml
> Please contact info at arin.net if you experience any issues.
>
>
>
>
>
>
> _______________________________________________
> ARIN-PPML
> You are receiving this message because you are subscribed to
> the ARIN Public Policy Mailing List (ARIN-PPML at arin.net).
> Unsubscribe or manage your mailing list subscription at:https://lists.arin.net/mailman/listinfo/arin-ppml
> Please contact info at arin.net if you experience any issues.
>
> _______________________________________________
> ARIN-PPML
> You are receiving this message because you are subscribed to
> the ARIN Public Policy Mailing List (ARIN-PPML at arin.net).
> Unsubscribe or manage your mailing list subscription at:
> https://lists.arin.net/mailman/listinfo/arin-ppml
> Please contact info at arin.net if you experience any issues.
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.arin.net/pipermail/arin-ppml/attachments/20240221/03569241/attachment.htm>