Digital archive services in Dubai are often quoted as scanning with a bigger price tag. The distinction is real, though, and it is about time horizon. A scanning project answers the question of how to stop walking to the store room. An archive answers the question of whether a file created today can still be opened, trusted and understood by somebody in thirty years who has never met anyone involved in creating it.
A folder of PDFs is not an archive
Three things separate the two. An archive keeps a master file that is never edited, distinct from the compressed copies people use day to day. It carries evidence that each file is unchanged since capture. And it holds enough description alongside the file that its meaning does not depend on the memory of the person who scanned it. Miss any of those and you have storage, which is a different and much cheaper thing.
Choosing formats
There is no single correct archival format, only formats with different trade-offs against the material. The reliable rule is to prefer open and documented specifications over proprietary ones, and to keep a lossless master even when the copy circulated to users is compressed.
| Format | Typical archival use | Trade-off |
|---|---|---|
| TIFF, uncompressed or lossless | Master image files for scanned documents, drawings and photographs | Large files and storage cost. Not convenient for everyday access, so an access copy is needed alongside |
| PDF/A | Access and delivery copies, and multi-page documents where text should stay searchable | A constrained subset of PDF. Fonts and colour handling must be embedded correctly or conformance fails |
| JPEG 2000, lossless mode | Large image collections where storage cost makes uncompressed masters impractical | Less universally supported than TIFF. Check your platform can render it before committing a collection |
| JPEG | Thumbnails and lightweight access copies only | Lossy. Each re-save degrades further, so it should never be the master |
| Native office formats | Born-digital records where the original file is itself the record | Depends on software that changes. Usually paired with a PDF/A rendition for readability |
| Proprietary DMS container | Nothing archival | Ties the collection to one vendor. If the product is retired the archive has to be extracted first, if it can be |
Fixity: proving nothing changed
Digital files degrade quietly. A bit flips on a failing disk, a faulty copy operation truncates a file, a sync client resolves a conflict badly. None of it announces itself, and a corrupted image often still opens, just missing part of the page. Fixity is the countermeasure: a checksum generated at capture and stored with the archive, then recalculated periodically and compared. A mismatch tells you which file to restore, and from which copy. Without checksums you will restore a corrupted file over a good backup and not find out for years.
Fixity is also the evidential half of the archive. Being able to show that a file matches the checksum recorded on the day it was captured is a stronger statement about integrity than any assurance about access controls.
Preservation metadata
Descriptive metadata tells you what a document is about. Preservation metadata tells you about the file itself, and it is what a future custodian needs. Record what the source material was, when and how it was captured, at what resolution and colour depth, what format the file is in, what checksum it carries, and every action applied to it since. When the archive is migrated to a new format in a decade, that history is what lets somebody confirm the migration was faithful rather than take it on trust.
Plan the next migration during this one
Every format eventually becomes awkward, then obscure, then unreadable. The way this is managed is not by picking a perfect format but by expecting to migrate and preparing for it: masters kept in formats that convert cleanly, a documented inventory of what formats you hold, and a scheduled review point rather than an open-ended assumption that today's choice will hold. An archive with a review date and a documented format inventory can be migrated by people who were not there when it was built. One without either has to be reverse-engineered first.
Independence from any one vendor
The most common way archives are lost is not disaster but decommissioning. A platform is retired, the export is partial, and the metadata that lived in the database rather than in the files goes with it. Build for exit from day one: open formats, metadata written alongside the files as well as into the system, file naming that carries meaning outside any one product, and a tested extraction route. If you cannot describe how the archive leaves its current platform intact, you do not yet have an archive, you have a dependency.
Frequently asked questions
What makes a digital archive different from a folder of scans?
Three things. A master file that is never edited, kept separate from the access copies people use daily. Evidence through checksums that each file is unchanged since capture. And preservation metadata describing what the file is and how it was made. Without those you have storage, which is useful but will not survive decades unattended.
Which file format should archival scans be kept in?
Usually a lossless master such as uncompressed TIFF, with a PDF/A access copy for everyday use and searchable text. Choose per material rather than applying one format to everything, and prefer open documented specifications over proprietary containers. The master should never be a lossy format, because every re-save degrades it further.
What is fixity checking?
Generating a checksum for each file at capture, storing it with the archive, then recalculating and comparing it periodically. It detects silent corruption from failing disks, bad copies or sync conflicts, which often leaves a file that still opens but is damaged. It also provides evidence that a file is unchanged since the day it was captured.
How often should an archive be migrated to new formats?
There is no fixed interval, which is why a scheduled review matters more than a fixed rule. Keep an inventory of the formats you hold, set a review date rather than leaving it open, and migrate when a format's support is visibly narrowing. Masters in formats that convert cleanly make each migration cheaper.
What is preservation metadata?
Information about the file rather than its subject: the source material, capture date, resolution and colour depth, format, checksum, and every action applied since. A future custodian uses it to confirm that a format migration was faithful. Descriptive metadata helps people find a document; preservation metadata is what keeps it trustworthy.
Does cloud storage remove the need for a preservation plan?
No. Cloud storage handles hardware durability well, but replication copies corrupted files just as faithfully as good ones, deletions propagate, and no provider warns you that a format is drifting out of support. Fixity checking, format migration planning and preservation metadata remain your responsibility wherever the files are stored.
Related reading
- Paper to digital conversionThe capture project that produces the material an archive is built from.
- Classification and retentionWhich records are permanent, and therefore belong in an archive at all.
- Formats and output optionsWhat we deliver, in which formats, and how the index data arrives.
- Security and handlingAccess control, encryption and audit trails around archived material.
- Athena Global TechnologiesParent-company background, including its records governance practice.