The prefix is old enough to have outlived several business plans. In the diligence room it appears as a tidy row in a spreadsheet: block size, seller, proposed buyer, timetable for technical handover and a valuation large enough to interest both lawyers and finance departments. The engineers are not alarmed. They can announce the route, arrange upstream acceptance, move reverse DNS and clean the published contacts. If the only question were whether packets can be made to flow next month, the transaction would look almost complete.
The bank is asking a different question. So is the auditor. They do not merely want proof that the resource can be used today. They want to know whether the registration history will survive tomorrow's objections. How did the block first enter AFRINIC's records? Which organization held it? Did that organization change name, control or jurisdiction? Were old contacts still authorized when later updates were made? Was a dormant entity revived only for a sale? Has anyone asserted a claim? Did litigation, receivership, a court order or board discontinuity touch the process by which the registry recognized the holder? Could a future correction reopen the file after money has moved?
That is where the title-insurance analogy becomes useful. It is not a claim that IPv4 addresses are land, or that a regional Internet registry is a land office in the full property-law sense. It is not an argument for turning registry policy into a sales contract or for letting an insurer define Internet governance. It is an institutional-economics analogy about scarce assets whose value depends on a record, a chain of authority and confidence that old defects will not surface without a disciplined way to handle them.
Real-estate title systems developed because private promises were too thin. A seller could warrant that it owned a parcel, but the buyer also needed protection against old liens, forged deeds, missing releases, boundary mistakes, hidden heirs and prior claims that might appear years later. The answer was not faith in the seller. It was search, defect classification, disclosed exceptions, underwriting discipline and a claims process that could defend or cure title problems when they appeared. Title insurance did not make the seller an owner by rhetoric. It built a trust layer around the history of ownership and the reliability of the record.
AFRINIC now presents the analogous problem for IPv4. Scarcity has turned old address records into economic assets. Reported address-theft episodes showed how dormant records and weak identity evidence can become targets. The Cloud Innovation litigation showed that registry action over scarce address space can become an existential commercial dispute. Court-supervised recovery arrangements and board discontinuity showed that the registry's own continuity may be tested in public. Contested election processes showed that delegated authority can itself become an object of dispute. None of this means every AFRINIC prefix is defective. It means a serious market can no longer treat the ledger as a frictionless utility in the background.
The useful question is therefore narrow. When a scarce number resource is financed, audited, pledged, acquired or depended upon, what form of assurance lets buyers, sellers, lenders, auditors and future operators trust today's holder record through yesterday's changes and tomorrow's claims? The answer is neither registry ownership nor private comfort. It is a title-like assurance layer: a repeatable discipline of title plant, search, chain-of-registration evidence, defect disclosure, adverse-claim handling, underwriting standards and additive correction, adapted to number resources rather than copied mechanically from land law.
The analogy is about assurance, not ownership
The first misunderstanding should be removed at the start. A title-insurance analogy does not require ordinary real-estate title. It requires an asset whose value depends on the quality of a public or semi-public record and on the ability to reconstruct authority across time. IPv4 fits that description even if legal labels remain contested. A block is valuable because networks, counterparties and registries recognize a holder or operator as having a usable position. If that recognition becomes uncertain, the economic value weakens even while the numbers remain technically routable.
The analogy begins with function rather than doctrine. Title insurance in property markets does not create land. It does not make the county recorder the owner of the house. It does not guarantee that every commercial use of the land will succeed. It narrows a particular risk: that the record history contains a defect or adverse claim that undermines the buyer's position. The product exists because time is long, records are imperfect and a seller may not be present, solvent or honest when an old defect emerges.
For IPv4, the equivalent risk is not a boundary line in soil. It is a break in recognized authority. The break may involve an original allocation file, a registry update, a merger, a defunct company, a stale contact, a forged authorization, an unrecorded transfer, a disputed board resolution, a court order, an abuse-driven hold, a route history that points to control by another party, or a registry action later said to have exceeded authority. The market does not need to call these defects title defects in the land-law sense. It needs a vocabulary for impairments that can weaken confidence in a scarce identifier.
The analogy also clarifies where registry power should stop. A title office or title insurer does not become the commercial owner of land because it searches and assures the record. Likewise, a registry that maintains accurate number-resource records does not become the owner of the economic value created by operators, customers and counterparties. Its defensible function is narrower: preserve uniqueness, registration accuracy, publication continuity, security-adjacent records, transfer history and dispute metadata. A title-like assurance layer should make that function more reliable, not inflate it into a veto over price, customer model, geography or business strategy.
This matters because weak records often reward the wrong people. When the chain is unclear, value migrates toward insiders, repeat brokers and aggressive claimants who understand undocumented history. When the chain is legible, a buyer needs less private memory and fewer whispered interpretations. Certainty around the record is not the same as commodifying the registry. It is a way to reduce the premium paid to opacity.
The title-insurance analogy therefore supports a thin registry and stronger record discipline. It says: do not make AFRINIC the commercial owner; make the history legible. Do not make the registry a price regulator; disclose record risk. Do not let every dispute become a route disruption; classify the claim and preserve the last verified operational state unless a narrow reason requires change. The object is assurance around the chain, not sovereignty over the asset.
Why AFRINIC makes the question unavoidable
Every regional Internet registry faces the structural shift created by scarce IPv4. Addresses have become transferable, priced and embedded in balance sheets, financing discussions and business continuity. AFRINIC is the sharpest case because the stress is no longer theoretical. Its records have been tested by reported address theft, major litigation, court-supervised recovery, governance discontinuity and election controversy. Those events make visible an assurance problem that calmer registry operations can hide.
The 2019 reporting on African IPv4 blocks is the first warning. Public reporting described allegations that valuable address space connected to dormant or weakly monitored organizations had been moved into other hands, with companies linked to a former senior AFRINIC figure and with Ron Guilmette's investigations prominent in the record. The reported value of affected addresses was large enough to turn an administrative problem into an economic one. The institutional lesson is more important than any single allegation. Once addresses carry market value, a stale record is not merely a clerical inconvenience. It is an attack surface.
The Cloud Innovation dispute added a different test. AFRINIC challenged the use and status of a large holder's resources; court proceedings followed; public reporting described injunctions, attempted termination, frozen funds and a wider institutional crisis. Later Mauritian proceedings included orders concerning member position and rectification of corporate records. Observers disagree about the merits and motives of the parties. For market infrastructure the lesson is narrower: a registry action over number resources can become a high-stakes legal conflict whose outcome affects recognition, continuity and confidence.
Receivership and board discontinuity changed the question again. A buyer can diligence a seller, a prefix and a transfer instrument. But what if the recordkeeper's authority path is itself unsettled? Public reporting in 2025 described AFRINIC as having been unable to elect a board for years, with Mauritian court supervision and a receiver tasked with arranging elections. The Register reported that a planned election was suspended and later annulled after concerns about powers of attorney and voter documentation. The specific allegations belong to proper process. The market signal is immediate: counterparties need to know which registry act, by which authorized person, at which time, will be treated as durable.
That is the situation in which a title-like discipline becomes unavoidable. When a system is calm, people rarely ask for its sociology. When it is publicly tested, they ask for evidence. Who signed? Under what authority? Was the authority continuous? Was the record preserved? Were corrections made by overwriting history or by adding a traceable entry? Did a dispute attach to one resource, one holder, one election or the whole institution? Did technical continuity survive while legal questions were resolved?
AFRINIC therefore illustrates more than a broad governance risk. It illustrates the economic need for chain-of-registration assurance. The more valuable a prefix becomes, the more a buyer must know that the recognized history will survive old defects, adverse claims and institutional discontinuity. A market that cannot answer that question will still transact, but it will hide distrust in lower valuations, heavier covenants, larger holdbacks, slower transfers and more litigation.
What a title search means in a registry ledger
A title search over land begins with records, not with the seller's confidence. It moves backward through deeds, mortgages, releases, probate records, tax liens, judgments and boundary documents until the searcher can describe the chain and the exceptions. A number-resource search would use different materials but the same discipline. It would reconstruct the resource's recognized history and identify the risks that remain.
For an AFRINIC prefix, the search begins with the original registration record. When did the resource enter the registry? Was it an allocation, assignment or later transfer? Which organization was named? Which contacts or maintainers controlled updates? What documents supported the action? If the record is old, does the archive show continuity between the entity named then and the entity now asserting control? If there were mergers, dissolutions, name changes or restructurings, are they evidenced by company records rather than by a convenient statement in a sale file?
The search then moves through registry events. Were there changes of holder, contacts, maintainers, route objects, reverse DNS or RPKI arrangements? Were those changes routine, or did they coincide with disputed corporate authority, dormant status, policy review, litigation, receivership or board discontinuity? Did the registry preserve a versioned history of who requested the change, what evidence was submitted, which staff or officer approved it and what reason was given? A search that cannot explain its own chain is not a search. It is a snapshot.
Operational history should be separated from registration history without being ignored. A prefix may have a clean holder record but a complicated route history. It may have been announced from networks unrelated to the registered holder, accumulated abuse reputation or appeared in inconsistent routing objects. These facts do not necessarily defeat the chain of registration, but they affect reliance. A title-like report should distinguish registration defects from operating impairments. A buyer wants both diagnoses, but it should not confuse them.
Dispute history is a further layer. The file should record known claims, court orders, registry holds, transfer denials, fraud allegations, abuse-contact escalations and notices from parties asserting authority. The purpose is not to decide every dispute inside the search. It is to stop hidden claims from surprising the next holder. In property practice, a disclosed exception can be priced, cured or accepted. The same principle applies to IPv4. A disclosed adverse claim is risk allocation; an undisclosed one is a market failure.
The search should end with a confidence conclusion, not a theatrical certificate of perfection. Serious title systems do not say that history is immaculate. They say what was searched, what was found, what was excepted, what can be cured, what remains uncertain and what standard will be used if a claim appears. That is the discipline AFRINIC-related space needs. A buyer should not be asked to believe a record is clean because someone says it is clean. It should be shown why the chain holds, where it is weak and what happens if the weakness becomes live.
The chain-of-authority file
The most valuable instrument in a title-like IPv4 transaction is not the number block alone. It is the chain-of-authority file. That file answers a practical question: why is the party signing today entitled to bind the resource position on which the buyer, lender or auditor will rely?
In simple cases the answer may be short. A currently active company is the registered holder. Its directors appear in current corporate records. The board has approved the transaction. The registry account uses current authorized contacts. No adverse claim is recorded. The route and security records are consistent with the holder's operating structure. The registry has no active hold. The file is boring, which is what finance prefers.
Older AFRINIC records may not be boring. A company may have changed name, merged into another group, been liquidated, revived or sold. A contact may be a former employee. A maintainer may be controlled by an address no longer associated with the holder. A historical update may have been made under an old policy or by a person whose authority is now unclear. A block may have been announced for years by a third party under a commercial arrangement that was never recorded as a transfer. None of these facts automatically proves a defect. Each creates a question the file must answer.
The file should connect corporate, registry, technical and legal evidence. Corporate documents show that the legal person exists, changed or authorized a transaction. Registry evidence shows that the recordkeeper recognized particular changes at particular times. Technical evidence shows that operational control has been consistent or explains why it was separated. Legal evidence shows whether courts, receivers, creditors or administrators have intervened. A board resolution from a company whose link to the registered holder is not proved is not enough. A registry snapshot without proof of the signer's authority is not enough. A routing screenshot without a registration chain is not enough.
This is where the analogy is better than a generic database-accuracy frame. Database accuracy asks whether today's field is correct. Title assurance asks why today's field can be trusted after walking through yesterday's changes. In an asset market, the second question matters more. The buyer is not buying a printout. It is buying confidence that the printout rests on a durable history.
The file also disciplines sellers. A seller with a clean file can command more confidence and faster execution. A seller with missing links must cure them, disclose them or accept that a buyer will price uncertainty. That is not punitive. It is how capital allocates risk. If a party wants asset-like pricing for a scarce prefix, it should be willing to provide asset-grade evidence.
For AFRINIC, standardized files would reduce pressure on the institution itself. Staff asked to process transfers during public stress should not improvise legal judgment under political heat. They should be applying an objective file standard: entity continuity, authorized representative, current control, no active hold or a clearly classified hold, operational continuity plan, dispute disclosure and audit log. The more objective the file, the less room there is for discretionary overreach or accusations of favoritism.
The defects old records can hide
Not every defect is the same. Mature assurance classifies defects rather than treating every problem as either fatal or irrelevant. This is one of the strongest parts of the title-insurance analogy. Property practice distinguishes forgery, missing release, boundary issue, lien, easement, identity mistake, probate gap and recording error because each has a different remedy and risk. IPv4 needs the same habit.
An identity defect exists when the current claimant is not clearly the same legal person as the registered holder, or when the connection is undocumented. This appears when old records name entities that have dissolved, merged, rebranded or migrated across jurisdictions. It also appears when a resource was historically associated with a trading name rather than a formal legal person. The cure is evidence: company registry extracts, merger documents, liquidation orders, court records, assignments, corroborated affidavits and registry history.
An authority defect exists when the holder may exist but the signer may not have power to act. Old contacts, stale emails, former staff, rogue consultants, compromised credentials or disputed directors can all create authority risk. The sensitivity of delegated authority was visible in the reported election disputes over powers of attorney. A transfer file should not ask merely whether a document exists. It should ask whether the grantor had authority, whether the scope covers the act, whether the document is current and whether the registry can test it.
A record-integrity defect concerns the archive itself. Historical updates may have been made without adequate evidence, or the archive may not show how the change occurred. The address-theft reporting made this concrete. If a registry record can be altered around dormant resources, the market must know which records have robust audit trails and which depend on trust in a thin archive. A missing audit trail is not always fatal, but it should be disclosed and compensated with stronger external evidence.
A dispute defect arises when another party claims the resource, challenges a transfer, alleges fraud, asserts creditor rights or relies on a court order. The answer should rarely be immediate destruction of operational use. It should be classification: active claim, dormant claim, court-bound claim, duplicate-claim risk, fraud hold, curable documentation gap or resolved dispute. Buyers can price different categories. They cannot price fog.
Operational impairments belong in a separate bucket. Route leaks, spam reputation, abuse history, broken reverse DNS, missing RPKI continuity or inconsistent IRR objects may not break the registration chain, but they affect usability. A prefix with a clean chain and messy reputation can be repaired operationally. A prefix with a broken chain and clean routes remains dangerous. The market needs both diagnoses.
The AFRINIC-specific category is institutional-continuity defect. If a registry has experienced receivership, board discontinuity or contested authority, counterparties must know whether a given action was taken by an actor whose authority will survive later review. The defect is not in the prefix alone. It is in confidence around the recordkeeper's act. A title-like system would record the authority basis for registry actions during such periods instead of pretending that all timestamps are equal.
Dormant records and address-theft risk
Dormancy once looked mainly like waste. In a scarcity market it also looks like opportunity, and sometimes like prey. That is the institutional lesson of reported African address-theft episodes. Dormant companies, old contacts, unused prefixes and weakly maintained records can become valuable precisely because nobody is watching them closely. Once IPv4 prices rise, a dead record is no longer dead. It may be a vault with a broken lock.
Public reporting around the 2019 AFRINIC episode described valuable IP blocks connected to entities that were no longer active or had been acquired long before. It also described alleged record alteration and companies linked to a former senior registry figure. The details have been debated for years because they illustrate a wider market problem. Scarcity turns old administrative gaps into economic incentives. If the record system does not preserve authority evidence, a sophisticated actor may be able to convert obscurity into control.
This changes how buyers should treat old AFRINIC space. A long period of inactivity is not automatically suspicious. Many organizations received resources in an earlier era and used only part of them. Some changed business plans. Some became universities, public agencies, carriers or legacy enterprises with irregular documentation. But dormancy creates diligence questions. Was the holder alive when later updates occurred? Who had authority to revive the record? Did any intermediary profit from the revival? Were old contact emails reactivated or replaced? Was the prefix routed during dormancy by someone else? Were there abuse reports suggesting unauthorized use?
The title-insurance analogy gives a disciplined answer. It does not say old resources should be confiscated or frozen. It says they need a search. The search should reconstruct corporate survival, authority chain, registry update history, route use and adverse-claim notices. If the chain is clean, the block can move with confidence. If the chain has gaps, they should be disclosed and either cured or excepted. If there is evidence of fraud, the registry should preserve the last verified state and send the issue through an independent process rather than acting as prosecutor, judge and executioner.
This is why simple "use it or lose it" rhetoric is dangerous. Dormancy may indicate inefficiency, but it is not proof of invalidity. A title-like system should not punish a resource merely because it was underused. It should ask whether the person asserting control today derives authority from a legitimate chain. Markets can decide whether underused resources should be sold, leased, financed or redeployed. The registry's job is to prevent false claims and keep the record honest.
For lenders and auditors, the address-theft lesson is straightforward. Old AFRINIC records should not be treated as self-authenticating. A prefix may be valuable precisely because it is old, and fragile for the same reason. The difference lies in evidence. That is what title-like assurance supplies.
Adverse claims and proportional handling
In a low-value system, a complaint can look like administrative noise. In a scarce-asset system, an adverse claim is a market fact. It may be weak, strategic or abusive. It may also be the first sign of a real defect. The point is not that every claim should stop a transaction. The point is that claims must be recorded, classified and handled according to a known standard.
An adverse claim in IPv4 can take many forms. A former shareholder says the seller lacked authority. A dissolved company's successor asserts that a block was transferred without approval. A creditor claims security rights. A court order freezes corporate assets. A registry concludes that a prior update was fraudulent. A third party says it has long operated the prefix under contract. A receiver questions actions taken during a governance vacuum. A public entity asks whether its historic resources were improperly alienated. Each claim affects confidence differently.
The worst response is discretionary opacity. If a registry silently marks a resource as suspicious, buyers cannot price it and holders cannot cure it. If it ignores all claims until litigation erupts, the eventual correction may be more destructive than necessary. If it treats a claim as a reason to impair unrelated resources, it turns a file dispute into a continuity crisis. Title-like discipline exists to avoid these extremes.
The first rule should be notice. The holder should know the nature of the claim, the evidence relied upon and the resource scope affected. The second should be classification: documentary, corporate, fraud-related, operational, abuse-related, creditor-related, court-bound or policy-based. The third should be proportionality. A claim over one prefix should not impair unrelated prefixes without specific evidence. A claim over a contact object should not invalidate an entire network. A complaint about commercial use should not become a security-service penalty unless a technical integrity emergency exists.
The fourth rule should be preservation of the last verified operational state. In property markets, a claim may cloud title while a dispute is resolved, but it does not automatically burn down the house. In IPv4, routes, reverse DNS, RPKI and registry publication should continue unless an independent decision, binding order or narrow security emergency requires change. That is not favoritism toward the current holder. It is continuity discipline.
The fifth rule should be independent review before serious adverse action. If AFRINIC proposes to revoke, reclaim, transfer, freeze, impair security services or deny recognition based on a disputed defect, the holder should have access to a review path outside the same staff chain that made the initial decision. Courts will remain necessary for some disputes, but not every registry-state question should require years of litigation. A specialized review mechanism could order interim preservation, require disclosure, compel correction or defer to court where the issue is genuinely legal rather than registry-functional.
The final rule is claims memory. Once a claim is resolved, the outcome should remain searchable. Was it withdrawn, rejected, cured, settled, adjudicated or converted into a corrective record? This helps future transactions and discourages repetitive claims. It also protects the registry from the suspicion that inconvenient history disappears. The title-insurance analogy is practical here: a good title system searches, excepts, underwrites, defends and cures. IPv4 may not need the same product form, but it needs the same institutional temperament.
Private warranties cannot repair public records
Commercial lawyers often reach for warranties and indemnities. The seller warrants that it has authority. The buyer can claim if the warranty proves false. The seller indemnifies the buyer against defects. A portion of the price may be retained for a period. These tools are useful, but they are too private to solve AFRINIC-related title confidence.
The first problem is solvency. A warranty is only as good as the party standing behind it. A seller of an old prefix may distribute proceeds, dissolve, move jurisdictions or become difficult to sue. A dormant entity revived for a transaction may have no meaningful asset base. A small operator may be honest but unable to absorb a large claim. A buyer financing a valuable block cannot treat a paper promise from a thin seller as equivalent to a clean chain.
The second problem is evidence. Private warranties do not improve the public or quasi-public record. If the chain is broken, a warranty may compensate after failure, but it does not tell future operators, lenders or auditors why the registry history is reliable. It also does not help the registry decide whether to record a transfer. The same defect can reappear at the next transaction because it was settled privately rather than corrected in the ledger.
The third problem is externality. An IPv4 defect does not always harm only buyer and seller. It can affect customers using the prefix, upstreams carrying routes, platforms relying on reputation data, lenders using the block as collateral and future purchasers inheriting the history. A private indemnity cannot fully internalize those network effects. The asset's usefulness depends on recognition beyond the contract parties.
Timing is equally important. Indemnities pay, if at all, after the fight. Title-like assurance tries to prevent the fight, disclose it before closing or provide a claims path that keeps the resource usable while the fight is resolved. For a network operator, continuity may matter more than damages. A court award years later does not restore customers lost when registry recognition was disrupted.
This does not mean warranties should disappear. They should become part of a broader assurance stack. The seller should still warrant authority and disclose known defects. The buyer should still receive remedies for misrepresentation. The lender may still require covenants. But these promises should sit on top of a searched and classified record, not substitute for one. In property terms, the warranty deed did not eliminate title search. It made title search more meaningful.
What lenders and auditors are really buying
The lender does not route packets. The auditor does not update reverse DNS. Yet both increasingly matter because scarcity has moved IPv4 from engineering inventory into capital planning. A company may buy addresses for expansion, carry them as an intangible asset, pledge them in financing, move them within a group, impair them after a dispute or disclose them in an acquisition. At that point, the question is not only whether engineers can use the prefix. It is whether finance can rely on it.
Lenders think in recoverability. If a borrower defaults, can the lender enforce against the asset position? Can it sell or transfer the block? Will the registry recognize the secured party, receiver or buyer? Does the borrower truly control the resource, or merely use it under a fragile contract? Are there adverse claims that would defeat enforcement? Has the registry ever challenged the holder's use, identity or authority? Could a future registry discontinuity prevent realization?
Auditors think in evidence. If management says the block is worth a material amount, what supports the assertion? Is the holder record current? Is there a chain showing how the company obtained the resource? Are there court orders, disputes or policy holds? Is the asset separable from a broader service contract? Are there impairment indicators from abuse reputation or registry uncertainty? Did management overstate value by ignoring transferability risk?
Both lender and auditor need title-like confidence. They need a file that can be reviewed without informal registry relationships. They need enough disclosure to distinguish clean assets from assets with exceptions. They need assurance that the record will survive a transfer, enforcement action, merger, receivership or board discontinuity. They also need to know what the assurance does not cover. A title-like report should not promise that the future price will rise, that every network will accept the route or that no policy dispute will occur. It should promise that the chain was searched to a defined standard and that known exceptions were disclosed.
This is where the analogy has its strongest economic force. Insurance is not only about indemnity. It is about making an asset financeable by converting unknown history into classified risk. A bank can lend against property because it receives a record search, exceptions and a policy. It knows old defects are either cured, excepted or backed by a claims process. Without that, it demands a deeper discount, more collateral or no loan.
The same logic applies to IPv4. If AFRINIC space is perceived as historically or institutionally riskier, capital will either avoid it or demand compensation. Some compensation will appear as lower valuation. Some will appear as longer diligence, more legal cost, larger holdbacks and stricter covenants. The objective of title-like assurance is not to inflate value artificially. It is to let value reflect known facts rather than fear of unknown files.
The registry is a ledger, not owner or guarantor
A registry should not become the insurer. That boundary is crucial. AFRINIC's proper role is to maintain the authoritative record for the functions it is supposed to serve: uniqueness, registration accuracy, publication, contactability, transfer recording, reverse-DNS and security-adjacent continuity, and dispute metadata. If it tries to guarantee asset value, approve commercial merit or stand behind every historical defect as an insurer of last resort, it will become both over-powerful and undercapitalized.
The ledger role is narrower and more defensible. A ledger records recognized claims and changes. It preserves version history. It identifies who requested a change, what evidence supported it, which authority approved it and what dispute or hold applied at the time. It distinguishes current status from historical fact. It corrects mistakes without erasing the mistake. It publishes enough information for the market to know which record is operative and which claim is contested.
This role does not require AFRINIC to declare that IPv4 is ordinary property. Nor does it require the registry to deny that holders have economically significant interests. It requires the registry to act like a serious recordkeeper in a scarce-asset environment. The more valuable the resource, the more disciplined the ledger must be. A low-value administrative database can tolerate informal changes. A capital-significant ledger cannot.
The boundary also protects the registry. If AFRINIC presents itself as the commercial source of all rights, it invites claims proportionate to the value it purports to control. If it disclaims all responsibility while retaining broad discretion, it invites mistrust and litigation. The stable middle is ledger accountability: strong duties to preserve record integrity, publish status, process objective changes, maintain audit logs and isolate disputes; no duty to guarantee business models, prices or profit.
The same boundary protects holders. A registry that is not the guarantor of commercial value has no reason to judge whether a buyer paid too much, whether a seller should have waited, whether a prefix should remain in a preferred geography or whether a business model is attractive. Those questions belong to markets, courts, regulators with proper authority and private contracts. The registry's question is narrower: does the record update preserve uniqueness, accuracy and continuity while disclosing material disputes?
AFRINIC's recent stress makes ledger discipline urgent. If corporate governance is contested, the ledger must become more auditable, not more discretionary. If board authority is discontinuous, the record must show the authority basis for actions taken during the discontinuity. If a receiver or court is involved, the record should preserve the connection between supervised authority and registry action. That is how a ledger remains trustworthy without pretending to be an owner.
Independent assurance and additive correction
If AFRINIC should not be the insurer, assurance must be layered. Some of it must come from registry records themselves. Some can come from independent search providers, auditors, lawyers, technical diligence firms, insurers or market utilities. The institutional design should avoid both extremes: a registry monopoly over assurance and a purely private market of comfort letters with no connection to the ledger.
The registry contribution is the title plant. In property markets, a title plant is an organized archive of records that allows efficient search across time. The number-resource equivalent would be a versioned and independently auditable archive of registration events, transfer evidence, holder identity records, contact changes, dispute flags, reverse-DNS and RPKI-relevant changes, and authority documents where publication is appropriate or controlled access is justified. The plant need not expose confidential documents to the world. It must preserve them so a qualified search can verify the chain.
Independent searchers can then produce title-like reports. They would not ask AFRINIC to certify commercial value. They would ask for record history, compare it with corporate and technical evidence, classify defects and state exceptions. A searcher might conclude that a chain is clean to the current holder except for unresolved corporate-name continuity in an old year; that no active adverse claim is recorded; that route history shows third-party operation over a defined period; that abuse reputation requires operational cleanup; and that the transfer instrument requires board approval plus a registry-recognized authorized contact. The value lies in specificity.
Underwriting standards would define when a report can support stronger assurance. A clean file with current corporate evidence, continuous registry history, no adverse claims and aligned operational controls might receive high confidence. A file with missing archives, dormant-entity revival or unresolved route use might receive assurance only with exceptions. A file with an active fraud allegation or court hold would not be described as clean. This is not ideology. It is risk classification.
The assurance layer could be voluntary at first. Buyers, lenders and auditors would demand it for larger transactions or older blocks. Sellers with clean files would adopt it because it reduces delay and improves credibility. Insurers or financial counterparties may insist on it before taking exposure. Over time, the market would create a norm: serious AFRINIC space should come with a chain-of-registration file and a defect schedule.
Correction is where the layer either earns or loses credibility. Every record system makes mistakes. The question is whether correction improves confidence or creates a new uncertainty. Destructive correction is especially dangerous in a scarce IPv4 market. If a registry deletes, overwrites or silently reverses history, it may solve an immediate embarrassment while making the chain less reliable for everyone. Title-like assurance requires additive correction: preserve the old record, identify the defect, state the authority for correction and show the new operative position.
This principle is familiar in serious ledgers. A bank does not improve auditability by erasing a wrong entry as if it never occurred; it posts a correcting entry. A court record does not become more trustworthy when an order disappears; it shows the order, appeal, variation or discharge. Land records preserve mistaken deeds, corrective deeds, releases and judgments because future searchers need the sequence. IPv4 registries should do the same for record-significant events.
Additive correction matters for several defect types. If a stale contact was replaced, the record should show when and on what evidence. If a transfer was recorded and later disputed, the record should show the dispute rather than pretending the transfer was either unproblematic or nonexistent. If a fraud investigation concludes that a block was moved improperly, the corrective entry should identify the basis for restoring or changing recognition. If a court or receiver orders a corporate-record correction, the registry chain should connect that authority to the registry action.
The benefit is not historical neatness alone. Additive correction reduces litigation incentives. A holder that knows a dispute will be recorded without immediate destruction has less reason to seek emergency relief. A buyer that sees a resolved defect can price the remaining risk instead of assuming concealment. Registry staff can process future requests without reinventing the past. An auditor can see whether management's asset position rests on a corrected chain or on silence.
Independence is vital throughout. If the registry alone provides assurance, it may defend past errors or expand its discretion. If private actors alone provide assurance, they may lack authoritative history and create inconsistent standards. A good system would let independent parties search against a registry plant, with clear access rules, audit trails and dispute procedures. The registry remains the ledger. The assurance layer interprets the ledger for market reliance.
Mobility without discretionary ownership
The final test is mobility. A title-like assurance layer should make AFRINIC-associated IPv4 more mobile, not because it ignores risk, but because it makes risk legible. Clean chains should move faster. Defective chains should reveal what must be cured. Disputed chains should be recorded without contaminating unrelated resources. Buyers should know what they are buying. Sellers should know what evidence they must bring. Lenders and auditors should know what reliance is reasonable.
This does not require AFRINIC to become a commercial owner. It requires the opposite. The registry should not decide whether a price is high enough, whether a buyer is morally attractive, whether a seller should have retained the resource, whether an operator's customers are in the preferred geography or whether scarcity should be rewarded. It should decide whether the record update is supported by objective authority, preserves uniqueness, maintains publication and security continuity, and properly records any adverse claim.
A title-like assurance system would also reduce the temptation to use registry discretion as a substitute for institutional repair. If AFRINIC worries about fraud, it should strengthen identity evidence, audit trails and claims handling. If it worries about old records, it should support chain-of-authority searches and corrective entries. If it worries about disputes, it should isolate them and provide review. If it worries about market abuse, it should disclose record facts rather than policing lawful commercial terms outside its competence.
The benefits would be distributed. Buyers would gain confidence that a prefix will not be reopened because of an old hidden defect. Sellers with clean files would receive fairer treatment. Lenders could finance with clearer collateral assumptions. Auditors could review management assertions without relying on folklore. Operators could continue serving customers while disputes are classified. The registry would gain legitimacy by becoming more predictable. The market would spend less on fear and more on productive redeployment of scarce addresses.
There will still be discounts for uncertainty, but they should be consequences rather than the organizing principle. Some files will be messy. Some claims will be real. Some old records will prove impossible to cure fully. Title-like assurance does not abolish risk; it prices, routes and narrows it. Mature markets do not pretend history is clean. They build institutions for living with history.
AFRINIC's recent past has made one lesson hard to avoid. A scarce IPv4 block is not just a routable object and not just a contract entry. It is a record-dependent economic position. When the recordkeeper's continuity and authority are publicly tested, the market needs more than one-off representations and vague discretion. It needs a way to search the chain, classify defects, disclose exceptions, preserve operational state and handle claims.
That is what the title-insurance analogy contributes. It is a metaphor for trust architecture, not a demand to turn addresses into land. It asks AFRINIC and the market around it to protect the ledger without worshipping the gatekeeper, to assure the chain without making the registry an owner, and to let scarce assets move without forcing every buyer to rediscover the same old uncertainty in a new diligence room.