APP VS LIBRARY
A finished app vs a self-hosted redaction library
If you have engineers, time, and a pipeline to put it in, a self-hosted open-source redaction toolkit is a genuinely good answer, and it is free. DataAnonymiser is for the case where the person handling the sensitive document is not going to write Python.
Side by side
The differences that matter.
Compared at the level of the category, not a specific vendor's current feature list.
| Dimension | Self-hosted library | DataAnonymiser |
|---|---|---|
| Licence cost | Free and open source | Paid subscription |
| Who can use it | Developers | Anyone who can use a desktop app |
| Setup | Runtime, dependencies, models, configuration, hosting | Install, activate, and it sets itself up |
| Customisation | Total — you own the recognisers and the pipeline | Custom terms and settings, not arbitrary code |
| Restore round trip | Build it yourself | Built in, session-scoped, local |
| Documents, images, folder scanning | Build the extraction pipeline yourself | Built in |
| Maintenance | Yours: upgrades, model updates, breakage | Signed updates delivered to the app |
Free is not the same as cheap
A redaction library is free to download and then costs engineering time forever: dependency upgrades, model management, text extraction for every file format you care about, a UI for the people who actually handle the documents, and a review workflow so nobody ships an unchecked output.
If that work is already on your roadmap, or you need redaction inside a pipeline you control, self-hosting is the right call and we would rather you did it than bought something ill-fitting.
What a product adds
DataAnonymiser is the whole path, not just the detection step: text, code, PDFs, office documents and images all handled; consistent labelled placeholders across a document; a session mapping so you can restore the AI's answer; a residual-risk report a person can read; a folder-level scan that is quick to repeat; and signed, verified delivery of the app itself.
The trade is flexibility. You cannot write your own detection logic in our app the way you can in a library — you can add your own confidential terms, and that is the extent of it.
Delivery integrity
Self-hosting means you are responsible for verifying what you pulled. Our installers are published with checksums you can verify, the macOS build is signed with an Apple Developer ID and notarized, and everything the app downloads afterwards is verified before it is installed.
Questions
Common follow-ups.
Is DataAnonymiser open source?
No. The application is proprietary and the source is not public. Our local-only claim is backed by things you can check yourself instead: the app keeps working with the network disconnected, and you can watch it with a network monitor or firewall.
Can I add my own detection rules?
You can add custom confidential terms — project names, client references, internal codes — which get their own placeholders. You cannot supply your own detection code; that is what a library is for.
What happens if my subscription lapses?
Activation and downloads require an active subscription. The app is designed to keep working offline for a grace period after that; see the terms for the specifics.
Related
Other comparisons.
ON-DEVICE VS CLOUD
On-device redaction vs cloud redaction APIs
Why sending data to a cloud service to have it de-identified creates the exposure you were trying to avoid.
DESKTOP VS BROWSER-ONLY
Desktop app vs browser-extension-only redaction
Why the browser is the right place to catch a leak and the wrong place to do the redaction.
REDACTION VS SETTINGS
Redaction vs relying on the AI vendor's privacy settings
Opting out of training is a promise about what happens to your data after it lands. Redaction changes what lands.
DataAnonymiser is a best-effort redaction tool, not legal advice. It does not guarantee detection or removal of all personal data, nor compliance with GDPR, CCPA or any other law. Always validate outputs against your own obligations.