[arin-ppml] ARIN 58 Table Topic - If the Waitlist didn't exist, what would?
Eric C. Landgraf
echarlie at vt.edu
Mon Nov 2 11:39:11 EST 2026
Hi Fernando,
One observation made during the policy experience report at ARIN 58 is
that operators requesting a /22 from the waitlist often do not get
served at all, or at least not in a timely manner. ARIN is freeing up
lots of /24s in tracking down ownership, but current policy requires
them to issue aggregates---they cannot fulfil a request for a /22 with 4
/24s.
For this to be workable (and I will echo Mike Burns' request that you
draft a proposal), any policy wanting to hand out more than a /24 from a
successor to the waitlist must amend 4.1.6.
Regards,
Eric C. Landgraf
Virginia Tech
On Oct 30 13:17, Fernando Frediani wrote:
> /24 is nothing as it is /22, so at least give a maximum of a /22 so they can
> get sizeable minimum enough to get some revenue and transfer additional
> space in the future if needed. Equally important is not to grandfathering
> those currently in the list who have more than a /22
>
> Fernando
>
> On 10/30/2026 12:58 PM, Mike Burns wrote:
> >
> > Hi Fernando,
> >
> > “I once again urge the community to consider giving out addresses from
> > the waiting list only to newcomers, those who have no IPv4 at all, up to
> > a /22 and *not* grandfathering the current list.”
> >
> > I agree, for the most part, and also proposed this. Give a /24 to those
> > without addresses, one time only.
> >
> > But it failed to reach consensus, and I supported abandoning it.
> >
> > It was Prop-327, filed three years ago.
> >
> > Now, do you think it might be worth revisiting it?
> > If so, I encourage you to write a new proposal.
> >
> > Regards,
> > Mike
> >
> > *From:*ARIN-PPML <arin-ppml-bounces at arin.net> *On Behalf Of *Fernando
> > Frediani
> > *Sent:* Friday, October 30, 2026 9:58 AM
> > *To:* arin-ppml at arin.net
> > *Subject:* Re: [arin-ppml] ARIN 58 Table Topic - If the Waitlist didn't
> > exist, what would?
> >
> > This is a great report. Thanks for sharing it Mark
> >
> > I once again urge the community to consider giving out addresses from
> > the waiting list only to newcomers, those who have no IPv4 at all, up to
> > a /22 and *not* grandfathering the current list.
> > Those who have something adapt how to live with what they have and when
> > necessary go to market to transfer. Lowering expectations to receive any
> > extra allocation from the waiting list helps as well.
> >
> > Also agree with the point that making it intentionally unreasonably
> > difficult to obtain IPv4 has never been a good idea for IPv6 adoption.
> >
> > Fernando
> >
> > On 10/30/2026 1:21 AM, Mark Van Daele via ARIN-PPML wrote:
> >
> > On Oct 28, Eric C. Landgraf wrote:
> > > It is unfortunate that none of the people actually *using* the
> > waitlist have chimed in on the mailing list about what getting rid
> > of the waitlist would mean to them.
> >
> > Hey Eric/All,
> >
> > I'll take that invitation. For disclosure: I operate MARKVD-NET
> > (AS402766), a small independent network in the Chicago suburbs
> > that manages an ARIN /24 recently issued via the waitlist. I would
> > go through the process again if needed. I've been a lurker for
> > some time, but excited to get involved in the community.
> >
> > Most of the organizations I serve run on a small budget or no
> > budget. They are mostly volunteer community groups in my area,
> > whose networks I often support on my own time. Having allocated
> > space lets me deliver services to them with greater flexibility at
> > lower costs.
> >
> > For networks like these, equipment is not the difficult part. You
> > certainly don't need a power hungry Cisco router to speak BGP
> > anymore. Our edge routers are small linux servers running BIRD,
> > and they handle full tables without issue. When the hardware is
> > cheap, ipv4 space becomes one of the largest barriers of getting
> > started, and certainly not a trivial one.
> >
> > I don't want to eliminate the waitlist, I'd much rather eliminate
> > the waiting. ARIN should have a pool of addresses ready for
> > organizations that can justify the need via the existing NRPM. In
> > addition to defunct holders as it is today, that pool should come
> > primarily from holders who have space they cannot justify, often
> > because they will never *ever* use it all.
> >
> > Where I land:
> >
> > 1. Leave the current waitlist in 4.1.8 largely as is. The October
> > distribution shows the waitlist works when space comes back to
> > ARIN. The problem is supply and not the method of delivery.
> >
> > 2. Increase supply through a community appeal and ARIN outreach
> > encouraging voluntary returns, especially from public
> > institutions, legacy holders, and government. Many of the largest
> > ipv4 holdings were issued before ARIN existed to organizations
> > that would not qualify for as much space under today's rules.
> > Space that was given freely to public bodies should be returned
> > freely when it isn't needed or would never realistically be used.
> >
> > 3. Give waitlist priority to organizations with no prior ipv4, or
> > holding a /22 or less. They have the fewest alternatives and might
> > bring the greatest benefit.
> >
> > 4. Don't redirect recovered space to 4.10, it certainly doesn't
> > need it now and won't for a very long time. At ARIN 54, staff
> > reported 13,938 /24s remaining in the 4.10 pool, with less than
> > 400 issued per year. That is more than 34 *years* of supply at the
> > current rate. The same report noted it would take about 1,727 /24s
> > to clear the waitlist, against quarterly distributions averaging
> > around 95 /24s. Recovered space sent to 4.10 would sit idle for
> > literal *decades* while new operators wait. If anything set a
> > smaller 4.10 reservation pool (3 years?) and refill it vi 4.1.7 as
> > blocks arrive. That frees up space to be issued *now* to fill need.
> >
> > Two more points, both IMHO:
> >
> > - ARIN should not monetize recovered addresses, directly or
> > through a third party. It directly goes against ARIN's role as a
> > steward, it turns ipv4 into a walled garden of those who have and
> > those who can pay, and it rewards holding space rather than using it.
> >
> > - ipv6 is the future, and new networks *should* be ipv6 first. My
> > networks run dual stack where possible and I'm working to move
> > every service I deliver so it is reachable over ipv6. That said,
> > nearly every network still needs ipv4 to reach the rest of the
> > Internet. Encouraging ipv6 adoption is right; penalizing ipv4,
> > making it intentionally unreasonably difficult to obtain ipv4, or
> > making ipv6 the only way to get ipv4 shouldn't be the goal and
> > makes ARIN poor stewards of the resource.
> >
> > In conclusion, my question and plea to every large and legacy
> > holder on this list: Can you reasonably justify your allocated
> > ipv4 based on current or planned future use? If not, returning the
> > excess would do more for new operators than any change to the
> > waitlist. Please be a good steward.
> >
> > Thank You,
> >
> > Mark Van Daele
> >
> > MARKVD-NET (AS402766)
> >
> >
> > On 10/29/2026 2:26 PM, Eric C. Landgraf wrote:
> >
> > Hi Martin,
> >
> > "Fairness" appears incidental to RFC 7020 and the NRPM; in the
> > discussion of ARIN-2026-4 I argued against taking the
> > principle of
> > fairness out of Section 6 and putting it in Section 1, because
> > some
> > policies like §4.10 are inherently unfair (for certain
> > definitions of
> > fairness).
> >
> > The PDP §1.4.1 does call for "Fair and Impartial Number Resource
> > Administration", but reading the context, it seems clear to me
> > that
> > policy proposals should limit ARIN's discretion in how to enforce
> > policy---i.e. it says the rules must be imposed fairly, rather
> > than the
> > rules must be fair.
> >
> > Now, transfers and the waitlist both require a needs-based
> > justification---yes we know Mike thinks very little of this,
> > at least
> > in part because it complicates the business of trading
> > IPs---but the
> > material change for end users of sending reclaimed addresses
> > to the 4.10
> > pool is that small operators must convincingly deploy IPv6 and we
> > prevent transfers of new allocations directly from ARIN. And for
> > operators who cannot justify under 4.10 AND hold less than a
> > /20 in
> > aggregate, they are still left with 3 options:
> >
> > 1) purchase on the transfer market, justifying need to ARIN;
> > 2) lease from their ISP or some other organization; or
> > 3) figure out how to qualify for 4.10 space.
> >
> > At present, I suspect that (1) and (2) will actually be much
> > cheaper for
> > organizations than (3): a /24 leases for about $2000/a and
> > sells for
> > about $7000, based on the best data I can readily get. This is an
> > absolutely *trivial* cost compared to the capital investment in
> > equipment and connectivity involved in running even a small
> > network, and
> > substantially less than the annual rent and power for a rack
> > at an IXP.
> > The straw man of a "little guy" isn't running on a 6509 he
> > found on the
> > side of the road, and if he is, IPs cost less than his power!
> > The needs
> > justification requirement presumably helps deflate prices of the
> > transfer market, by raising the bar required for speculative
> > address
> > purchases, and the organizations actually able to purchase
> > addresses
> > speculatively are almost certainly aware that doing so hurts
> > their
> > customers' ability to use their services.
> >
> > It is unfortunate that none of the people actually *using* the
> > waitlist
> > have chimed in on the mailing list about what getting rid of the
> > waitlist would mean to them. I do not know what that means,
> > but it does
> > not give me a lot of confidence in there even being a
> > non-strawman
> > argument for keeping the waiting list.
> >
> > Eric C. Landgraf
> > Virginia Tech
> >
> > On Oct 28 19:21, Martin Hannigan wrote:
> >
> > Mike,
> >
> > I don’t need AI to make post slop for me on this issue.
> > “Uniqueness and
> > conservation" isn't the whole mission. RFC 2050 and the
> > NRPM introduce fair
> > distribution alongside conservation and aggregation as
> > goals, and needs
> > based assessment has always been a “rule” for fairness.
> > Not just an
> > efficiency rule. (rathole alert) If ARIN were only a
> > directory it wouldn’t
> > be issuing resources.
> >
> > Warm regards,
> >
> > -M<
> >
> >
> >
> > On Wed, Oct 28, 2026 at 18:09 Mike Burns <mike at sum.net>
> > <mailto:mike at sum.net> wrote:
> >
> >
> > Hi Marty,
> >
> > Forgive me for using your post as my example. I got
> > the answer below my
> > signature from AI, as I am interested in a larger
> > discussion about the
> > inevitable ingression of AI into policy development.
> >
> > Nonetheless, I do stand by its argument. Giving out
> > free /22s is not going
> > to change any economic situation, and the IPv4 market
> > has moved the debate
> > into a different realm.
> >
> > RIRs are for uniqueness and conservation. Not for
> > equalizing little guys
> > and big guys. And these days, price offers more
> > conservation than anything
> > else. But as I also said, the waiting list is small
> > potatoes, and it is
> > actually functioning quite well.
> >
> > Although there was that subject for the lunch table…
> >
> > Regards,
> >
> > Mike
> >
> >
> >
> > Hi Martin,
> >
> > Yes, I think that became an important rationale for
> > it: giving smaller
> > organizations access to IPv4 without requiring them to
> > compete in the
> > transfer market with much larger buyers.
> >
> > But I don't think that necessarily argues for
> > maintaining the waiting list
> > indefinitely.
> >
> > We already exclude organizations holding more than a
> > /20 and limit
> > waiting-list distributions to a /22. If the objective
> > now is to make the
> > remaining recovered IPv4 serve a public-policy purpose
> > rather than simply
> > enter the market, 4.10 seems to me the better
> > destination.
> >
> > It preserves non-market access to scarce IPv4, but
> > ties that access
> > directly to IPv6 deployment.
> >
> > So I agree with the equity concern. I am questioning
> > whether the waiting
> > list remains the best mechanism for addressing it.
> >
> > Regards,
> >
> > Mike
> >
> >
> >
> >
> >
> > *From:* Martin Hannigan <hannigan at gmail.com>
> > <mailto:hannigan at gmail.com>
> > *Sent:* Wednesday, October 28, 2026 5:55 PM
> > *To:* Mike Burns <mike at iptrading.com>
> > <mailto:mike at iptrading.com>
> > *Cc:* Eric C. Landgraf <echarlie at vt.edu>
> > <mailto:echarlie at vt.edu>; arin-ppml at arin.net
> > *Subject:* Re: [arin-ppml] ARIN 58 Table Topic - If
> > the Waitlist didn't
> > exist, what would?
> >
> >
> >
> > Mike,
> >
> >
> >
> > Wasn’t the point of the waitlist equity with the big
> > $corp big $$budgets?
> > Competition and fairness?
> >
> >
> >
> > Thanks,
> >
> >
> >
> > -M<
> >
> >
> >
> >
> >
> > On Wed, Oct 28, 2026 at 16:43 Mike Burns via ARIN-PPML
> > <arin-ppml at arin.net> <mailto:arin-ppml at arin.net>
> > wrote:
> >
> > Hi Eric,
> >
> > I think we agree on the cleanest solution.
> >
> > I encourage you to draft a proposal to deal with
> > returned addresses in a
> > manner that doesn't perpetuate the waiting list.
> > It is an opportune moment, with the list having
> > recently received a
> > substantial allocation and the decks somewhat cleared.
> >
> > In the long run, this is small potatoes.
> > I think establishing a cutoff date, grandfathering the
> > current list, and
> > allowing it to wither on the vine would be the best
> > approach.
> >
> > I don't think the failure to reach consensus in the
> > past is much of a
> > barrier.
> >
> > Regards,
> >
> > Mike
> >
> > -----Original Message-----
> > From: ARIN-PPML <arin-ppml-bounces at arin.net>
> > <mailto:arin-ppml-bounces at arin.net> On Behalf Of 'Eric C.
> > Landgraf'
> > Sent: Wednesday, October 28, 2026 1:37 PM
> > To: arin-ppml at arin.net
> > Subject: Re: [arin-ppml] ARIN 58 Table Topic - If the
> > Waitlist didn't
> > exist,
> > what would?
> >
> > On Oct 28 13:12, Mike Burns wrote:
> >
> > ARIN is not a charity; that would not be consonant
> > with its purpose.
> >
> > Moving the proceeds one step away does not
> > eliminate the conflict. If
> > the police direct ticket revenue to a foundation
> > they created or
> > selected, the incentive problem has not
> > disappeared; it has merely
> > been moved.
> >
> > I do not support the blurring of incentives
> > between institutional
> > self-enrichment and enforcing the law equally. So
> > why do this with
> > ARIN, even at the one remove you propose?
> >
> > An alternate framing: this is equivalent to when your
> > county or town
> > auctions off property (thru a 3rd party) to pay
> > back-tax on it. In many,
> > perhaps most cases, we can't find the deeded "owner"
> > of the resource, there
> > is no entail, and someone forgot to pay their RSP fees
> > for some length of
> > time: as ARIN, we want to recover those fees.
> >
> > Under that framing, do we have a real conflict of
> > interest? Does the
> > facilitator take all the proceeds after the RSP fees
> > and transfer fees are
> > paid, or does ARIN get a fixed return if there's a
> > profit?
> >
> > I tend to agree with you that getting rid of
> > everything that isn't
> > effectively 4.10 space is probably the cleanest
> > solution, since needs
> > justification under 4.10 is still extant with the
> > extra provisio that you
> > have to be deploying IPv6. But as you noted, that did
> > not reach consensus
> > at
> > the time it was last proposed, so there is value in
> > exploring other
> > potential policies.
> >
> > Eric C. Landgraf
> > Virginia Tech
> > _______________________________________________
> > 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.
> >
> > _______________________________________________
> > 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.
More information about the ARIN-PPML
mailing list