Professional Workflow Quiz: Backup and File Management

A hash or checksum is a mathematical fingerprint calculated from a file's exact contents. Recalculating the hash after a copy and comparing it to the original confirms that every bit arrived intact. Any difference, however small, changes the hash value.

If a recalculated checksum does not match the original, the file has changed since the hash was first created. This could indicate a failed write, a corrupted card, a transmission error, or bit rot in storage, and the copy should not be treated as reliable.

Verifying a copy by checksum after ingest confirms the copy is bit-perfect before the original card is cleared. Writing two copies before verification adds a second layer of protection against ingest failures or storage errors discovered only after the card has been reformatted.

RAID systems use multiple drives to tolerate drive failure without data loss, but they provide no protection against accidental deletion, ransomware, software corruption, or any catastrophe that affects the entire system simultaneously. They are a reliability mechanism, not a backup.

The 3-2-1 rule ensures redundancy at every failure mode: three copies means no single loss is fatal; two different media types prevents a format-specific failure from taking all copies; one offsite copy protects against any single-location disaster.

Any threat that reaches the physical location of your storage, whether fire, flood, theft, or building damage, destroys all copies kept at that location simultaneously. An offsite copy, whether cloud or physical, is unaffected by any single-location event.

A professional archive records which version was delivered to which client at what specification and when. Without this tracking, responding to revision requests, redelivering files, or resolving disputes about what was supplied becomes uncertain and error-prone.

Maintaining multiple unsynchronized catalog files creates risk: edits made in one instance may not appear in others, and conflicting versions produce ambiguity about which settings are authoritative. A single primary catalog with regular verified backups removes that ambiguity.

XMP sidecar files store all the editing instructions applied to a raw or original file as portable text metadata. They allow edits to be shared between applications, preserved independently of any proprietary catalog, and tracked separately from the original image data.

RAID protects against a single drive failing, not against all the other threats that destroy data: accidental deletion, ransomware, fire, or the failure of the RAID controller itself. Treating RAID as a substitute for off-site backups is a common and costly mistake.

Writing simultaneously to two cards creates an immediate second copy of every frame at the moment of capture. If one card fails or is lost, all the images are still on the second card without any gap in the sequence.

A consistent naming convention encodes date, job code, or sequence number directly in the filename. Automated ingest scripts, batch operations, and archive searches can all operate reliably on predictable filename patterns, eliminating manual sorting and reducing transcription errors.

Without systematic tracking of which version, size, and format was sent to each client at each stage, it is easy to resend an older version, deliver the wrong resolution, or lose track of what was approved for use in which context.

Hashing is meaningful only when exact bit-level preservation matters. Verifying a final archive against a known-good checksum confirms the file has not changed through storage, migration, or transmission, which is essential for archival integrity but unnecessary for casual viewing.

A professional workflow exists to eliminate the failure modes that result in lost files, incorrect deliveries, and unrecoverable mistakes. Consistency and redundancy built into the process are risk management, allowing the creative work to proceed without technical disasters undermining it.