A Durable File Naming Convention for Family Archives and Genealogy Records
Short answer. A durable family archive naming convention uses a consistent date format, a stable record identifier, a short subject description, and a clear indication of the file's role or version. Keep descriptive information in metadata and a separate catalog, so photographs, scanned documents, AI transcripts, and genealogy records remain identifiable even when software or folder structures change.
- Use a consistent date format and avoid making filenames depend on a particular genealogy application.
- Assign each historical item a stable identifier that remains unchanged when files are moved.
- Separate original scans, transcripts, edited images, and other derivatives through predictable suffixes.
- Treat AI-generated transcriptions as working documents that need source references and uncertainty notes.
- Maintain a portable catalog that connects filenames to people, places, dates, and archival relationships.
Build filenames around information that can survive software changes
A family archive may move between computers, external drives, cloud storage services, genealogy platforms, and cataloging applications. A naming convention should remain understandable after every move. The filename needs enough information to identify the item without depending on a particular database or application.
A practical structure is YYYY-MM-DD_RecordID_Subject_FileRole.ext. For example, 1948-06-12_FA0042_FamilyPicnic_MASTER.tif identifies a dated item, its stable archive number, a short subject description, and the role of the file. The date is written in year-month-day order, which sorts chronologically in ordinary file listings.
Historical dates are not always precise. Use a year alone when only the year is known, and avoid inventing a month or day to make the filename look complete. If the date is estimated, record that uncertainty in the catalog. A filename should communicate the available evidence, not create an appearance of certainty.
Assign a permanent identifier to each archival item
The record identifier is the most important part of a durable naming system. Assign it when an item enters the archive and keep it for every associated file. A photograph may have a master scan, a smaller access copy, a restored derivative, and a descriptive record. All of them should retain the same identifier.
For example, FA0042 might identify a single photograph showing relatives at an outdoor gathering. Its related files could include FA0042_MASTER.tif, FA0042_ACCESS.jpg, and FA0042_DESCRIPTION.txt. The identifier remains stable even if the estimated date or identification of a person changes.
Avoid putting a person's full name into every filename. Family relationships can be uncertain, people may have several names, and privacy considerations may make extensive personal details inappropriate in filenames. Keep richer descriptions in the catalog, where they can be corrected without renaming every associated file.
Choose consistent labels for different file roles
A naming convention becomes useful when every file role has a defined meaning. MASTER should identify the preserved source file. ACCESS can identify a smaller copy intended for convenient viewing. EDITED can identify an altered derivative, while TRANSCRIPT can identify text derived from a document or recording. Use a short controlled vocabulary and document its definitions.
For example, a scanned letter might have the following related files: FA0081_MASTER.tif for the original scan, FA0081_ACCESS.jpg for a viewing copy, and FA0081_TRANSCRIPT.txt for its transcription. If an AI handwriting recognition tool produces an initial transcription, label that document as a draft or AI-assisted version until a human review has been completed.
Do not use vague labels such as final, final2, new, or latest. These descriptions lose meaning when several people work on the same archive. A date or explicit revision number provides a clearer history. Where a transcription is corrected, preserve the previous version if its changes are relevant to understanding the editorial process.
- MASTER: preserved original digital capture.
- ACCESS: convenient viewing or listening copy.
- EDITED: modified derivative.
- TRANSCRIPT: text derived from a source document or recording.
- LOG: processing and editorial history.
Make room for genealogy and AI-assisted transcription
Genealogy AI tools can assist with searching, indexing, handwriting recognition, and extracting information from historical records. Their usefulness depends on the quality of the source, the legibility of the handwriting, and the reliability of the resulting interpretation. A naming convention should make it easy to connect an AI-assisted document to the exact record from which it was derived.
For example, a digitized parish register entry might be assigned the identifier GR0127. The original image could be GR0127_MASTER.tif, while an initial AI transcription could be GR0127_AI-DRAFT.txt and a corrected human transcription could be GR0127_REVIEWED.txt. The catalog should identify the register, page, repository reference if available, and relevant image location.
Transcribe old records with special attention to surnames, place names, dates, occupations, and abbreviations. These details often carry the greatest genealogical value and can be especially difficult for recognition systems. Preserve uncertain readings with a consistent notation such as [unclear] or [possible: Miller]. Do not silently replace an uncertain word with a confident-looking guess.
Keep the source image available beside the transcription. A future researcher may have different information about a family name or historical location, and the original record allows that interpretation to be reconsidered.
Keep the catalog independent of the folder structure
Folders are useful for browsing, but they should not carry the entire descriptive burden. A portable spreadsheet or tabular catalog can connect each identifier to its filenames, date, people, location, source, format, and processing history. CSV is a useful exchange format because it can be opened by many spreadsheet and database applications.
Include separate fields for the date shown on the item, the estimated creation date, and the date of digitization. These are different events. Add a date confidence field with values such as exact, approximate, and unknown. If an image depicts several relatives, list the identifications separately and indicate whether each is confirmed, probable, or unidentified.
Avoid making the catalog dependent on a single software application's internal record number. The archive identifier should be present in the filenames and in the exported catalog. This makes it easier to reconstruct relationships if a program is discontinued or a database must be migrated.
Test the convention against difficult archival material
A naming system needs to accommodate undated photographs, several versions of the same document, photographs of the same event, and items with uncertain identities. Before applying it to thousands of files, define how these situations will be represented. For an undated photograph, a filename might begin with UNKNOWN-DATE_FA0152 rather than an invented date. For a photograph dated approximately 1930, use an approximate-date marker in the catalog and a documented convention in the filename.
Use only characters that are broadly supported across operating systems. Letters, numbers, underscores, and hyphens are generally practical choices. Avoid special symbols, unnecessary punctuation, and excessively long filenames. Keep the file extension intact because it identifies the format and helps software interpret the contents.
Finally, document the naming rules in a short README.txt stored at the archive's top level. Include the date format, identifier structure, role suffixes, uncertainty conventions, and instructions for creating new records. Keep the README and catalog in the backup arrangement alongside the collection itself.
Adopt the convention with a controlled migration
Renaming an established archive all at once can break links, duplicate files, and disconnect images from existing genealogy notes. Start with a small collection and create a mapping between old and new filenames. Retain the original filenames in a catalog field called previous_filename so that earlier references can still be followed.
After renaming, confirm that the catalog points to the correct files and that related master, access, and transcript files share the intended identifier. Preserve the migration record as part of the archive's documentation.
The next step is to create a README.txt, define five file-role labels, and assign stable identifiers to the first ten items in the collection before extending the convention to the rest of the archive.
Questions readers ask
Should a family archive include names in filenames?
Short subject descriptions can help, but personal names are better maintained in the catalog. This allows uncertain identifications and changing family knowledge to be updated without renaming associated files.
How should duplicate photographs be named?
Give each distinct physical item its own identifier, even when two prints show the same image. If one file is simply another copy of the same digital capture, document its duplicate status in the catalog instead of creating a misleading new historical record.
What should happen when an estimated date changes?
Update the catalog's date and retain the previous interpretation in the change history. If the date forms part of the filename, record the old and new filenames in the migration log.
Are spaces acceptable in archive filenames?
Spaces work in many modern systems, but underscores and hyphens provide predictable separation across applications and command-line environments. Consistency matters more than the particular separator.
How often should the naming rules be revised?
Review the convention when a new material type enters the collection or when a genuine compatibility issue appears. Version the README so that earlier cataloging decisions remain understandable.