Summary
- The central unallocated pool reached its final-five-/8 trigger on 3 February 2011. RIPE NCC crossed a separate regional threshold on 14 September 2012, when it reported that allocations from 185/8 had begun under a one-/22-per-LIR arrangement.
- Three evidentiary states must remain separate. Version 1 of proposal 2010-02 records the proposed ceiling and rationale. The versioned final adopted text is absent from the material considered here. RIPE NCC’s announcement records what the institution reported implementing on the threshold date.
- The cap changed the role of documented need. Need remained an eligibility condition, but a larger validated requirement could no longer produce a larger allocation from the protected remainder.
- Equal maximum allocations per LIR used a recognised institutional unit and avoided the classifications demanded by alternative rules. That equality did not account for differences in prior holdings, corporate structure, customer base or access to technical and commercial substitutes.
- IPv6 was connected to eligibility and transition purpose. The public record considered here supplies no basis for attributing measured IPv6 deployment, market entry, routing or conservation outcomes to the rule.
- “Founding assumption” is analytical shorthand for an abundance-era operating condition in which justified demand could ordinarily be served from expanding unallocated stock. It is not a claim about a documented belief held by RIPE NCC’s founders.
The number that framed, but did not settle, the decision
A /8 contains exactly 16,777,216 IPv4 address values. A /22 contains exactly 1,024. Dividing the former by the latter gives 16,384 theoretical /22-sized units in a perfectly aligned /8. That calculation precedes reservations, fragmentation, prior commitments, operational exclusions and later adjustments. It is neither a count of practically awardable blocks nor a count of applicants, allocations, assignments, routed prefixes, independent companies or beneficiary networks.
The calculation nevertheless identifies the scale of the constraint. Each capped allocation consumed a small, fixed fraction of a finite block. Larger allocations would have consumed that stock in larger increments. A uniform ceiling therefore had the mechanical potential to preserve opportunities for more future allocations than continued awards at larger sizes.
Mechanical potential is narrower than observed conservation. The practically available inventory may have differed from the full prefix because not every address was necessarily available for the same purpose at the same time. Application volume also mattered: a cap preserves options only while qualified demand arrives to exercise them. The duration of the stock and the number of recipients are empirical properties of administration over time, not consequences that can be calculated from prefix length alone.
The arithmetic also leaves the entitlement question unanswered. Once quantity stopped scaling with demand, policy had to specify the unit to which the fixed maximum attached. RIPE NCC used the Local Internet Registry. A qualifying LIR with demand far above the ceiling faced the same maximum as one with a much smaller requirement. Differences above that ceiling ceased to affect the quantity awarded from the protected remainder.
That choice preserved an option for later claimants by restricting present ones. It also defined which distinctions the registry would recognise. LIR status remained visible and administrable. Corporate affiliation, accumulated inventory, customer population and the availability of substitute arrangements did not form part of the equal-block formula.
The institutional issue was therefore more specific than shortage. Scarcity supplied the need for restraint. It did not determine whether equality should follow accounts, corporate groups, entrants, previous holdings or some graduated measure of unmet demand. Those were alternative ways of locating the entitlement within the same finite stock.
Two exhaustion events and three evidentiary states
The global and regional events occurred at different administrative levels.
On 3 February 2011, the rule governing the last five unallocated /8 blocks in the central pool was triggered. One block went to each of the five Regional Internet Registries. The contemporaneous central-pool record establishes that central allocation and its stated global rationale.
The recipients of that event were regional registries, not providers, enterprises or households. Each registry still held and administered regional inventory after the central pool reached its trigger. Differences in existing stock, demand and regional policy meant that the five registries did not necessarily reach their own final phases together.
RIPE NCC reported its regional threshold more than nineteen months later. Its 14 September 2012 announcement said that it had begun allocating from 185/8. It described an arrangement under which each LIR could receive one /22 and said that an applicant needed an IPv6 allocation and demonstrated IPv4 need. The same announcement placed RIPE NCC’s institutional scale at approximately 8,000 members in more than 75 countries.
The policy history available here has a deliberate boundary. Version 1 of proposal 2010-02 records a proposed maximum allocation no larger than /22 and explains the transition rationale advanced for that design. It shows what was put forward at that stage. It is not the versioned final adopted text.
No versioned final policy record is included in the evidence considered for this article. That absence prevents a textual comparison between the initial proposal and the operative wording after deliberation. It leaves open such matters as final drafting, qualifications and any intervening revision.
The announcement supplies a different fact: RIPE NCC said that the one-/22-per-LIR arrangement was being applied when it entered 185/8 on 14 September 2012. The article therefore attributes the design to proposal version 1 and the reported implementation to the contemporaneous announcement. It does not use either source as a substitute for the missing final versioned text.
This separation matters because institutional documents answer different questions. A proposal records an advocated design and rationale. Adopted text fixes the rule authorised through the policy process. An implementation announcement reports institutional action at a specified time. Later operational evidence would show how individual cases were handled and what recipients did with their allocations.
The regional threshold was consequently a change in inventory administration rather than the end of every form of IPv4 activity. Existing allocations remained in use, downstream assignments could continue, registered resources could change control under applicable transfer arrangements, and returned or recovered space could be treated under later rules. The defining change concerned new distribution from the protected regional remainder: a larger justified claim no longer produced a correspondingly larger award.
When need became an eligibility test rather than a quantity rule
Need-based allocation links the size of an award to a validated operational requirement, subject to the stock and limits then in force. The registry examines expected use, rejects unsupported demand and allocates an amount related to the accepted claim. That model works most naturally when inventory is sufficiently elastic for one approved allocation to leave the general service available to later applicants.
A final protected block changes the intertemporal problem. Meeting a large claim today may eliminate the possibility of meeting even a small claim tomorrow. A registry must then consider the interests of people and organisations that have not yet applied and may not yet exist. Their claims cannot be verified in advance, but the policy can reserve an opportunity for them.
The reported /22 ceiling performed that reservation mechanically. An applicant still had to show IPv4 need, yet demand above the ceiling no longer affected quantity. The registry verified entry into a standard entitlement rather than deciding how much of the remaining stock a large forecast deserved.
This reduced the stakes of demand forecasts. An applicant had less to gain from projecting a requirement far above the fixed maximum, and the registry did not have to compare large requests against one another to divide 185/8. It could focus on the conditions for receiving the standard block.
The design also used a recognised unit. RIPE NCC already dealt with LIRs as accountable recipients responsible for registration and resource administration. Attaching the cap to that relationship avoided the need to classify independent companies, customer populations or degrees of economic disadvantage before each award. That advantage was institutional legibility, not proof that the rule imposed the smallest administrative burden among all possible designs.
The same legibility created the policy’s central abstraction. An LIR account could represent a new independent provider, an established operator with extensive earlier resources, a subsidiary within a wider group or an organisation serving many downstream networks. Equal treatment of accounts did not make those underlying positions equal.
The IPv6 condition fitted the transition logic. Requiring an IPv6 allocation linked access to the remaining IPv4 stock with a formal step towards the larger address space. A small IPv4 block could support dual-stack infrastructure, translation services or legacy reachability while the recipient developed IPv6 capability.
Holding an IPv6 allocation and deploying IPv6 are different operational states. The condition showed eligibility and policy purpose. It left actual routing, traffic, service availability and customer adoption to be demonstrated through operational observation. The article accordingly treats IPv6 as a transition linkage rather than a measured behavioural result of the last-/8 allocation.
Conservation deserves the same precision. A smaller award uses less inventory than a larger one under otherwise equal conditions. That relationship establishes the cap’s capacity to spread allocations. It supplies no observed duration for the stock and no measured increase in successful entry.
Strategic responses also belong in the analysis as scenarios. When entitlement follows an LIR, a related organisation might consider whether separate eligible accounts could produce separate entitlements. The public material used here provides no finding that this occurred. The scenario identifies the kind of boundary that an account-based rule places outside ordinary need review.
The policy was therefore defensible without being self-validating. It preserved future options, made the maximum predictable and connected eligibility to transition readiness. Its fairness depended on whether LIR equality was an acceptable proxy for the interests the remaining stock was intended to serve.
Allocation, assignment, transfer registration and routing
Several distinct acts sit between central inventory and operational use.
An IANA-to-RIR allocation places a block under a regional registry’s administration. That was the level of the February 2011 event. It conveyed no direct award to a network operator or end site.
RIR inventory consists of resources held and administered under rules applicable at a particular time. Original prefix size is only the starting point. Commitments, exclusions, returns and other constraints can change what is available for ordinary allocation.
An RIR-to-LIR allocation recognises an LIR as the recipient of a block. Under the arrangement reported in September 2012, the relevant entitlement was capped per LIR. The allocation record identifies a registry counterpart; it does not enumerate devices, customers or economically independent networks.
A downstream assignment places some portion of an allocation with infrastructure, a customer or an end site. The same prefix can support different combinations of direct addressing, customer assignments and shared-address systems. Allocation size alone reveals none of those arrangements.
Transfer registration records a recognised change in control under an applicable policy. It differs from allocation out of previously unallocated inventory. A registered change might accompany an arm’s-length transaction, a merger, a reorganisation or another permitted event, so a registration record should not automatically be treated as a sale.
A BGP observation concerns reachability. A separately visible announcement can show that a network presents a route for a prefix, but it does not measure dense utilisation, beneficial ownership, customer count or transaction price. A smaller block may also appear inside an aggregate rather than as its own route.
These distinctions prevent claims from travelling farther than their evidence. A registry allocation establishes receipt at one administrative layer. Entry, transition, market value and competitive effect arise later, through the recipient’s circumstances and decisions.
Equal treatment at the LIR boundary
The rule equalised a maximum at a point RIPE NCC could recognise. That boundary had practical value. LIRs were institutional counterparts with established responsibilities, while a direct allocation formula based on every customer, device or affected member of the public would have been unworkable.
The equality was narrow by design. One LIR was not necessarily one company, one network or one country. LIRs differed in prior holdings, customer numbers and demand. Some organisations could consider direct registry status, while others relied on assignments from a provider. The same nominal allocation could therefore relieve very different constraints.
An incumbent with substantial earlier inventory might use the additional block as a marginal reserve or compatibility resource. A new LIR without comparable holdings might depend on it for initial IPv4 reachability. A downstream network could benefit indirectly through its provider without controlling the address resource. The common cap treated none of these differences as grounds for a different quantity.
Approximately 8,000 members across more than 75 countries illustrated RIPE NCC’s reach, but those counts represented neither applicants nor beneficiaries. They also revealed nothing about how participation in the policy discussion was distributed. Institutional scale and allocative incidence are separate questions.
The fairness test must therefore ask what equality is meant to achieve. If the aim was to prevent a small number of large immediate claims from consuming the protected remainder, the fixed cap directly addressed that risk. If the aim was to equalise adaptation opportunity among independent organisations, prior holdings and corporate affiliation became relevant. If the aim was to favour future entry, a rule neutral between established and new LIRs only partially targeted that objective.
Six constituencies through three incidence channels
The cap reached six constituencies through prior inventory, direct registry access and secondary exchange. Incumbent LIRs and new LIRs encountered different starting positions; networks outside the LIR channel depended on intermediaries; prospective transfer sellers and buyers faced a market shaped by scarcity; and RIPE NCC administered the boundaries connecting all three channels.
A shared evidentiary boundary governs the incidence analysis. There is no complete linked account here of practical inventory, applications and decisions, affiliations and prior holdings, downstream assignments, routing, IPv6 activity, transfers or prices. The mechanisms below are consequently grounded comparisons rather than estimates of aggregate winners, losses or causal effects.
Prior inventory separated incumbents from entrants
An incumbent LIR entered the final-/8 phase with organisational experience and, in many cases, some stock obtained under earlier conditions. Its formal entitlement to the new block was the same as that of another qualifying LIR. Its material position could be very different.
Earlier holdings offered options. An incumbent could make more intensive use of existing assignments, introduce address sharing, redesign services, expand IPv6 or seek additional resources through transfer arrangements. Each response shifted costs into engineering, capital, customer management or dependence on counterparties. The earlier stock did not eliminate scarcity, but it could provide room in which to adapt.
The cap constrained incumbents by refusing to scale the final allocation with their larger demands. This was part of its intertemporal function: limiting present awards preserved the possibility of future ones. An established provider with rapid growth could regard the standard block as small relative to its customer base, yet the same constraint prevented such a provider from consuming a much larger portion of the protected remainder through one claim.
Scarcity could also increase the usefulness or exchange value of earlier holdings. Contemporary scholarship on the emerging IPv4 market treated dormant and underused resources as part of an institutional-economics problem. Its early estimates neither establish a complete regional market nor attribute a particular gain to the /22 policy. The important point for incidence is that historical inventory affected the alternatives available when new registry supply was capped.
A new LIR received the policy’s clearest formal benefit. Continued large allocations to earlier applicants might have left a later entrant with no direct space from 185/8. The fixed maximum reserved the possibility of a small independent allocation after the regional threshold.
A /22 could support core services, a limited customer population, translation infrastructure or dual-stack operation. It could reduce immediate dependence on an upstream provider or a transfer purchase. That option was valuable even where the block fell short of total projected demand.
Formal entry and competitive parity were different outcomes, however. A new provider requiring more capacity had to combine the allocation with technical substitution, provider space or a commercial acquisition. An incumbent could face the same cap while drawing on a larger inherited base. The rule broadened the opportunity to receive something without equalising the full resource position of recipients.
The resulting incidence turned on starting conditions. Equal additions to unequal stocks leave the stocks unequal. Correcting that history would have required a policy aimed at previous holdings, reclamation or redistribution, none of which was inherent in a cap on new allocations.
Direct registry access separated LIRs from mediated networks
A network outside the LIR channel encountered scarcity through a provider or through the decision whether to become an LIR. A small operator, enterprise or community network could seek provider-assigned space, redesign services around IPv6 and translation, or accept the responsibilities associated with direct registry status.
Mediated access was not necessarily deficient. Provider assignments could supply addresses efficiently without requiring every network to maintain a direct relationship with RIPE NCC. The arrangement also placed registration and resource-management duties with an accountable intermediary.
Dependence carried trade-offs. Provider-assigned space could make renumbering necessary when changing providers and could constrain independent routing arrangements. Shared addressing could reduce the need for public addresses while complicating inbound connectivity, logging and troubleshooting. Direct membership could offer greater autonomy but introduce financial and administrative obligations.
The equal allocation stopped at the LIR boundary. Differences among downstream networks entered the registry’s view through their providers rather than through separate entitlements. A large LIR serving many customers and a small LIR serving its own infrastructure could receive the same maximum, while the customers behind them had no independent claim under that formula.
RIPE NCC benefited from using a counterpart it already recognised. Eligibility and compliance could be evaluated against an existing institutional relationship. The registry avoided having to decide which downstream activity deserved its own entitlement.
At the same time, rationing enlarged the burden of public explanation. Consistency could show that qualifying LIRs received the stated maximum. It could not alone justify why the LIR, rather than an independent corporate group or a class of entrants, was the proper unit of equality. That justification belonged to the design of the rule, not to the accuracy of individual application review.
The registry also faced questions of remedy. An applicant could seek correction of factual or procedural errors concerning its eligibility. Broader objections—such as whether the entitlement unit ignored a relevant difference—required policy review rather than a larger award in one case. Under scarcity, an individual remedy that consumes more inventory can affect later claimants who are absent from the proceeding.
Secondary exchange separated potential sellers from buyers
Once a qualifying LIR could no longer obtain a larger allocation from the protected remainder, transfer arrangements became more important to networks with additional IPv4 demand. A holder with surplus resources could become a seller; a network facing the ceiling could seek space as a buyer.
For a potential seller, scarcity created the possibility of realising value from resources no longer required for current operations. The holder might instead retain the block for future use, reorganise existing services, return it or use it within an address-enabled offering. Renumbering, proof of authority and the loss of future flexibility could all affect the decision to transfer.
A registered transfer did not necessarily identify an arm’s-length market exchange. Corporate reorganisations and other changes in recognised control could produce similar records. The transfer channel must therefore be understood as a means of reallocating control, only some of which may reveal a market price.
Buyers gained access to capacity beyond the registry ceiling. They also faced search, negotiation, block reputation, integration and financing questions. A small buyer could find those frictions significant relative to the desired quantity, while a larger organisation might have more capacity to evaluate and complete a transaction.
Willingness to pay and operational need were related but distinct. A market could move addresses towards organisations placing high private value on them, yet purchasing power did not necessarily track transition difficulty or wider public benefit. Conversely, a standard registry allocation could broaden formal access while leaving resources with recipients whose realised need was modest.
Later empirical work comparing reported transfers with changes inferred from registration and routing cautions against treating official lists as a complete census. That work spans several regions and later years, so it serves as a measurement warning rather than a finding about price or motive at RIPE NCC’s threshold.
The transfer channel completed the incidence triangle. Prior stock influenced whether an organisation could sell or avoid buying. Direct registry access supplied a capped starting block. Secondary exchange offered additional flexibility while shifting allocation from an eligibility rule towards negotiation and capital.
The transition block and the costs of substitution
The strongest defence of the reported arrangement describes the /22 as a compatibility resource rather than a complete answer to IPv4 demand. That interpretation fits both the small size of the allocation and the IPv6-allocation condition.
IPv6 expanded the address space available for new endpoints and services. Its usefulness still depended on the reachability of customers, equipment and counterparties. During transition, a network might have to operate both protocols rather than replacing IPv4 immediately. The remaining IPv4 block could support services that required legacy reachability while IPv6 capability developed alongside them.
Carrier-grade address translation allowed many users or devices to share a smaller pool of public IPv4 addresses. This reduced direct address consumption by changing network architecture. It also moved work into translation equipment, port management, logging, abuse response and application support. Services needing inbound connections could be particularly difficult to operate through shared addressing.
Provider-assigned space offered another route. It could be efficient for a network that valued connectivity more than independent resource control. The cost was potential dependence on the provider, including renumbering and limits on routing autonomy when the commercial relationship changed.
Transfers offered larger or additional blocks where counterparties could agree. They introduced capital and transaction considerations that the uniform registry allocation avoided. They also enabled reallocation from holders with lower current use without requiring the registry to reclaim every apparently quiet block.
Reclamation and more intensive use of earlier allocations could release capacity, but operational use was not always visible from a public route. A routed block might be lightly occupied, while an address range absent as a separate announcement might operate inside an aggregate. Renumbering an established service could impose costs even when the address count suggested spare capacity.
These alternatives show why the /22 could have option value. It gave a qualifying LIR some directly registered IPv4 capacity with which to maintain compatibility or operate transition infrastructure. It did not eliminate the need to choose among IPv6, sharing, provider dependence and commercial acquisition when demand exceeded the cap.
The eligibility linkage should not be mistaken for an observed deployment effect. Establishing that the allocation accelerated IPv6 would require recipient-specific histories of technical change. The defensible interpretation is institutional: RIPE NCC connected access to the protected IPv4 remainder with possession of an IPv6 allocation and presented the small block as part of transition.
Participation, authority and the meaning of “founding assumption”
A registry distributing comparatively elastic inventory can ground much of its authority in competence. It keeps accurate records, applies common criteria, evaluates need and makes resources available predictably. Rationing retains those responsibilities while adding a question that technical diligence alone cannot answer: which differences among valid claimants should affect their share?
The figures reported in September 2012 describe a large and geographically broad institution. They are not a participation measure. A meaningful account of deliberation would distinguish unique contributors from repeated interventions and identify the organisational interests represented. It would also place active entities against the wider membership that did not enter the discussion.
Such information would clarify representativeness without turning technical policy into a simple vote. Address policy depends on expertise, and a relatively small group may identify consequences that a broad plebiscite would overlook. Expertise, however, does not remove differences in material position among established holders, new entrants, service providers and registry administrators.
The missing final versioned text also matters here. Proposal version 1 reveals the design and rationale advanced at that point. The announcement reveals how RIPE NCC described implementation. Neither provides a complete view of revisions, objections or accommodations made between those stages.
Scarcity changed the entity of justification. Under a need-scaled model, an applicant could contest whether its demand had been assessed accurately. Under the cap, the larger dispute concerned the rule that made additional need irrelevant to quantity. Review had to address not only facts about the applicant but also the classification chosen for entitlement.
“Founding assumption” names this institutional transition in analytical terms. It refers to an operating environment in which additional justified need could ordinarily be answered from expanding unallocated stock. No founding-era statement in the evidence considered here attributes that premise to particular founders.
The assumption ended at the margin rather than across every registry function. RIPE NCC continued allocating, maintaining records and registering changes. What disappeared was the expectation that a larger accepted need would normally lead to a larger new allocation from regional inventory.
A bounded comparison of four designs
The observed equal-/22-per-LIR rule and three alternatives expose different answers to the same problem. None removes scarcity. Each identifies an entitlement unit, demands particular information, invites possible strategic responses and determines which decisions can be reviewed.
The observed rule: equality at the account
The reported design attached one standard maximum to each qualifying LIR. Its central advantage was that RIPE NCC already recognised the recipient. The rule avoided the corporate, historical and comparative-need classifications required by the alternatives.
That reduced the number of questions decided within each allocation. Once eligibility and demonstrated need for the standard block were established, the registry did not have to weigh one applicant’s forecast against another’s or reconstruct the applicant’s entire resource position. Applicants also knew that presenting greater demand could not increase the protected allocation.
The strategic concern followed the boundary of the rule. If related organisations could hold separate qualifying LIRs, account structure might affect the number of available entitlements. That is a design scenario rather than a finding about recipient conduct.
The transition fit was broad but shallow. More qualifying LIRs could potentially obtain a small compatibility block, while a network with greater immediate demand had to rely more heavily on substitutes. A uniform set of allocations might be announced separately or carried within aggregates; any change in routing volume depended on recipient topology and announcement choices.
Remedy under this rule was clearest for mistakes about eligibility, need for the standard block or account treatment. Complaints that the formula ignored prior holdings or common ownership challenged the policy’s unit of equality. Comparing realised treatment would require allocation-level records connected to the identities behind the LIRs.
A needs-based taper: quantity and timing remain contested
A taper would preserve a relationship between validated demand and quantity while lowering the maximum as inventory declined. It could give a larger block to a network facing substantial near-term compatibility requirements without returning to unrestricted need-based allocation.
Its entitlement would be a share of accepted demand subject to a time-varying ceiling. That design makes timing more consequential. An applicant arriving before a reduction could qualify for more than an otherwise similar applicant arriving later. Forecasting, urgency and utilisation would remain central to the award.
Applicants might respond by filing earlier or presenting aggressive demand projections. Registry staff would have to judge forecasts under conditions in which every larger award reduced the remainder. These are plausible incentives, not observed outcomes.
A taper could ease transition for a high-demand network by supplying a larger contiguous block. That block might support aggregation within the recipient’s network. The same earlier consumption could leave later networks dependent on provider assignments or transferred fragments, with uncertain routing consequences.
Review would focus on demand findings, the applicable ceiling and the timing of a request. A successful appeal increasing one award would reduce the stock left for people not represented in that case. Any retrospective remedy after depletion would therefore be difficult even if the original classification proved wrong.
Testing a taper would require reconstructing how alternative ceilings interacted with the demand claims presented over time. The result would remain sensitive to changed applicant behaviour, since organisations anticipating a taper might not file as they did under a fixed cap.
Corporate-group entitlement: the account becomes a classification question
A corporate-group rule would attach one standard entitlement to an independently controlled group rather than to each LIR account. It responds directly to the possibility that formally separate registry relationships may sit under common control.
The design would need ownership and control evidence. The burden of obtaining and evaluating that evidence is unmeasured. Corporate control could also differ from operational integration: subsidiaries within one group might run separate networks across different jurisdictions, while formally independent organisations might coordinate infrastructure closely.
Strategic response could take the form of restructuring, altered ownership arrangements or disputes over the point at which influence becomes control. Again, these are scenarios created by the classification, not claims about what organisations would actually do.
Transition compatibility would depend on how the group used the single block. A connected organisation might distribute or aggregate it efficiently. Geographically or operationally separate subsidiaries might find one shared allocation awkward, particularly where each needed its own external reachability. Routing effects would turn on topology and any need to divide or deaggregate the resource.
The principal remedy would be review of the grouping decision. Applicants would need notice of the ownership or control finding, a transparent definition and an opportunity to correct inaccurate information. A remedy should also address changes in control after allocation without treating every corporate event as proof of abuse.
Evidence for comparison would have to connect legal control with operational separation and actual use. The decisive question is whether group-level entitlement better represented independent demand than account-level entitlement without creating classifications too detached from network reality.
Entrant or prior-holdings priority: history becomes part of eligibility
A priority rule could retain the standard block while favouring independent entrants or applicants with limited earlier holdings. Instead of changing allocation size, it would change access order or eligibility when claims competed for the remaining stock.
This design targets adaptation capacity more directly than account equality. An applicant with no inherited inventory may depend more heavily on a small transition block than an established holder with substantial space. Prioritising that condition could protect a class with fewer alternatives.
The classification is not simple. A newly created subsidiary may belong to an established holder, while a long-standing LIR may be launching a distinct service with genuine compatibility needs. Measuring prior holdings also raises questions about affiliated organisations, transferred resources and the relevance of space committed to existing customers.
Applicants might reorganise or characterise activity to fit the preferred class. Such behaviour would depend on the exact definitions and benefits. The rule would need to distinguish legitimate changes in business structure from attempts to multiply priority.
For transition, priority could direct blocks towards networks with the least inherited IPv4 capacity. It might also deny space to an established operator building an IPv6-oriented service for which a small compatibility allocation had real value. More independent entrants could create more separately originated routes, while concentration in established networks might support aggregation; neither result follows from priority status alone.
Review would centre on entrant status, affiliation and the measurement of previous holdings. Reasons for an adverse classification would need to be specific enough to contest. Because priority affects queue position or eligibility rather than only quantity, delay itself could become part of the remedy.
A fair comparison would examine whether the chosen classifications identified organisations with materially fewer adaptation options. It would also need to consider how applicants changed their conduct in response to the priority rule.
Why the comparison remains bounded
The four designs distribute discretion differently. The equal-LIR rule places most discretion at the eligibility boundary and treats accounts alike. A taper preserves judgment about demand and makes timing part of quantity. Corporate-group entitlement moves scrutiny into ownership and control. Entrant or prior-holdings priority makes institutional history relevant to access.
Their routing and transition implications remain conditional. Larger allocations might aid aggregation for one network but consume stock sooner. More small allocations might broaden independent reachability while producing more separately announced prefixes. A block directed towards an entrant might have high compatibility value, or it might be less intensively used than one directed towards an established service. Those are propositions for testing, not results implied by the design.
No alternative can be ranked from prefix arithmetic or institutional description alone. The choice depends on the objective assigned the greatest weight: breadth of formal access, fit with near-term need, protection of independent entrants, neutrality between organisations, preservation of later options or restraint of administrative discretion. The designs make different compromises among those objectives rather than offering a single technically dictated answer.
Conclusion
Scarcity changed the marginal governance dispute. The central question was no longer only whether an applicant’s need was genuine. It became which unit was entitled to the remaining resource, when that entitlement should arise and which classifications could be reviewed when claimants differed.
The reported rule answered those questions with a recognised institutional unit and a fixed maximum. LIR status was legible to RIPE NCC, and the standard block limited the effect of large present claims on future availability. The same legibility excluded prior-holdings and corporate-structure differences from its equality test. Two qualifying accounts could be treated alike even when the organisations behind them had sharply different reserves and alternatives.
A needs-based taper would have retained greater sensitivity to operational demand while making timing and forecast review more consequential. Corporate-group entitlement would have addressed multiple accounts under common control while requiring judgments about ownership and operational independence. Entrant or prior-holdings priority would have targeted unequal starting positions while creating disputes over history, affiliation and classification.
The transition rationale remained credible within all four designs: scarce IPv4 space could support compatibility while IPv6 capability developed. What varied was who should receive that option, in what quantity and through which reviewable rule. Routing, strategic and administrative consequences depended on behaviour rather than following automatically from the entitlement formula.
The bounded trade-off was between a rule that was clear at the LIR boundary and alternatives that recognised more of the differences behind that boundary at the cost of additional judgment, information and contestable classification.