Systems · Fioka
Digital chaos, filed into drawers.
Point it at the drive nobody has opened in three years. It reads what is actually there, files it into drawers, and shows you which files are copies of each other — without deleting, moving or renaming a single one.
What it is
Fioka is for files: PDFs, photos, video, scanned paper, and the twelve years of folders behind them. It builds an inventory of everything it finds, fingerprints each file by its contents, and sorts it into drawers — documents by type, photos by date and place, code and system clutter out of the way.
The rule the whole tool is built on is that it does not touch anything. Fioka reads, counts and proposes; you decide. Deleting is a separate step you run on purpose, it moves files to quarantine rather than destroying them, and there is a script that puts them back. Nothing disappears because a tool decided it should.
We built it because we needed it. On our own working drive it read 53,448 files — 187 GB — in thirteen minutes, and found 10,425 groups of identical files wasting 72 GB between them. For 837 files it could not identify a safe original, so it proposed nothing and flagged them for a human. That is the behaviour we care about.
This is not business intelligence. Fioka does not care about your margins; it cares about whether you can find the signed annex. Numbers live in business dashboards, and the two stay separate on purpose.
How it works
-
01
Point it at the mess
A folder, a drive, or an Android phone. It does not need to be tidy first, and it is not tidied behind your back.
-
02
Everything gets fingerprinted
Each file is hashed by its contents, so identical copies are found even when somebody renamed them.
-
03
Read the report
What you have, what is duplicated, how much space the copies waste, and which files it is not sure about.
-
04
You decide what goes
Deletion is a deliberate second step, into quarantine, reversible with one script.
What you get
- Read-only by design — A scan deletes nothing, moves nothing, renames nothing.
- Duplicates by content — Matched on a sha256 of the file itself, not on the filename.
- Drawers, not a folder tree — Documents, photos, screenshots, video, archives — sorted on the way in.
- Photos by date and place — EXIF date and GPS, plotted on a map of where they were taken.
- Says when it is unsure — Files with no safe original are flagged for you instead of guessed at.
- Auditable afterwards — Every decision written to CSV and a SQLite database you can re-read later.
Who it is for
- Drives and shoeboxes of scans copied between three computers and two backups, none of them authoritative.
- Phones and photo libraries where the same picture exists in five places.
- Anyone about to buy storage for files they already own four copies of.
Questions
- Will it delete my files?
- No. A scan never deletes, moves or renames anything. Deletion is a separate step you run on purpose; it moves files to quarantine rather than destroying them, and a restore script brings them back.
- How does it find duplicates?
- By hashing the contents of every file. Two files count as duplicates when their contents are identical, which catches copies that were renamed — and never flags two different files that happen to share a name.
- Does it read the text inside scanned documents?
- Not yet, and it does not pretend to. A PDF that already carries a text layer is read and sorted by what it says. A photographed or scanned page without one is filed as an unread scan and labelled as exactly that, rather than being quietly sorted into the wrong drawer. Reading those is the next piece of work, not something on offer today.
- How is this different from a shared drive?
- A shared drive stores whatever you put in it, including the four copies. Fioka inventories it, tells you which files are the same file, and gives the archive a structure.
- Is this the same as business dashboards?
- No. Fioka handles files and documents. Business dashboards handle numbers. They are separate products.