THE EVIDENCE EXISTS. CAN ANYONE EXAMINE IT? By 0pcter A record can be authentic, permanent, and practically useless. Its timestamp may be correct. Its digital signature may be valid. Its cryptographic fingerprint may match the original. Yet the information needed to understand what happened may no longer be available. This exposes a weakness in digital verification that is easy to overlook: proving that evidence once existed is not the same as preserving the ability to examine it. Consider an engineering firm responsible for inspecting a bridge. Each inspection produces photographs, measurements, sensor readings, maintenance observations, and a final report. The firm creates a cryptographic hash of the inspection package and records that hash in a permanent public ledger. Years later, investigators discover a structural failure and attempt to reconstruct what the engineers knew. The ledger confirms that a particular inspection package was committed at a particular point in the record's history, but the original photographs and measurements have disappeared. The investigators possess evidence of a record without possessing the record itself. They cannot determine whether the missing photographs revealed corrosion, whether measurements indicated deterioration, or whether the final report accurately represented the underlying observations. The hash cannot reconstruct the lost files. It can only test whether a subsequently recovered file matches the committed data. Without that file, the permanent commitment proves far less than its existence might suggest. This distinction matters because digital infrastructure increasingly separates information from the systems used to verify it. Organizations may store large files in private databases, cloud repositories, archives, or systems operated by third parties while recording only compact cryptographic commitments elsewhere. The arrangement can be efficient. A small fingerprint can represent a much larger collection of information without requiring every participant to store the complete collection. But that efficiency creates a dependency: the underlying information must remain accessible when verification becomes necessary. The dependency is not merely technical. Information can become unavailable through business failure, contract termination, deliberate withholding, inadequate retention policies, lost encryption keys, or incompatible software. A repository may still contain the original files while refusing access to an investigator. An organization may retain encrypted archives but lose the means to decrypt them. Even when the bytes survive, obsolete formats or missing interpretation rules can make their meaning difficult to recover. These failures have different consequences. If a file has been altered, a previously recorded hash may expose the alteration. If a file has been deleted, the same hash cannot restore it. If access is denied, the hash cannot compel disclosure. If the file survives in an unreadable format, its integrity may be established without making its contents intelligible. Each situation demands a different remedy, despite all four appearing under the broad language of record verification. The problem becomes more serious when records are preserved for people who were not present when they were created. A contemporary employee may know where an inspection file is stored, which software opens it, and how its measurements should be interpreted. An investigator twenty years later may possess none of that institutional knowledge. The original organization may no longer exist. Its software licenses may have expired, its employees may have retired, and its documentation may have disappeared. Long-term verification therefore requires more than preserving isolated digital objects. It requires preserving enough context to interpret them. A measurement without its unit, calibration information, or method of collection may be misleading. A database export without its schema may be difficult to reconstruct. A digitally signed approval without the associated policy version may establish who approved something without explaining what the approval authorized. Authentic evidence can still become ambiguous evidence. This suggests that the durability of a verification system depends on several independent properties. The information must remain intact, obtainable, interpretable, and connected to the circumstances that give it meaning. Integrity addresses whether the information has changed. Availability addresses whether it can be retrieved. Interpretability addresses whether its contents can be understood. Context establishes what the information actually describes. None of these properties automatically supplies the others. A publicly accessible document may have no reliable provenance. A cryptographically authenticated document may be inaccessible. A complete archive may lack the metadata needed to reconstruct its original meaning. A carefully preserved record may contain observations that were false from the beginning. Treating all these problems as one problem encourages solutions that address only the easiest part. There are conventional ways to reduce the risks. Organizations can maintain redundant archives, establish enforceable retention requirements, preserve open file formats, document schemas, and test recovery procedures. Independent custodians can hold copies of important evidence. Access controls can protect sensitive information while defined legal processes provide authorized disclosure. These practices are neither glamorous nor revolutionary, but they address practical failures that a permanent transaction ledger cannot solve. Bitcoin enters this discussion at a narrower and more defensible point. Its transaction history can provide independently verifiable commitments that help establish whether subsequently presented information matches information committed earlier. A record anchored to Bitcoin may remain independently checkable even if the organization that created it later disappears. That property can be valuable where disputes concern historical alteration, competing versions, or the timing of a recorded commitment. But Bitcoin does not guarantee that the committed information remains available. Storing a hash onchain does not preserve the original document. Storing the document itself may improve independent retrievability, but only if the relevant data can actually be obtained and the infrastructure serving it remains functional. Transaction inclusion, data access, long-term retention, and meaningful interpretation must be evaluated separately. Permanence of a commitment should never be mistaken for permanence of the entire evidentiary process. There is also an economic question. Maintaining accessible archives costs money. Storage must be paid for, systems maintained, formats migrated, and retrieval tested. If the organization responsible for those expenses receives no benefit from keeping old information available, its incentives may weaken long before the records lose their public importance. A technically durable commitment can therefore outlive the economic arrangement required to make it useful. That is where the architecture of a verification system must confront its institutional reality. Who is obligated to retain the evidence? Who can demand access? Who pays for preservation after the original business closes? Who verifies that archived files remain recoverable rather than merely listed in an inventory? A system without credible answers may provide impressive cryptographic guarantees while remaining vulnerable to ordinary organizational neglect. The strongest design would not assume that one technology solves every part of the problem. It would identify which evidence must be preserved, what must be publicly verifiable, what must remain confidential, and how authorized parties can recover the information when a dispute occurs. It would distinguish a record's integrity from its availability and test both over time. Where independent historical commitments add value, Bitcoin could serve a specific function within that design. Where conventional archival controls are sufficient, additional infrastructure may add little. The deeper issue is what society expects permanent records to accomplish. A permanent fingerprint can establish continuity between an earlier commitment and a later file. It cannot answer questions about evidence that no one can produce. Nor can it recover the knowledge necessary to interpret information whose surrounding context has disappeared. The future of digital accountability will depend not only on whether records can be altered, but also on whether they can still be examined when their meaning becomes important. Evidence that cannot be examined may survive as a record while failing as a means of accountability. OBSERVE • VERIFY • UNDERSTAND 0pcter