Nutrinet users create many bogus barcodes (probably related to scanner quality)

[teolemon]

What

• Nutrinet users create many bogus barcodes (probably related to scanner quality)
• issue reported by segundo on the bad Nutrinet barcodes probably related to the barcode scanner used (non native)
• see https://openfoodfacts.slack.com/archives/C0116U37JCE/p1640626240022400

How ?

• there is a technical solution on Nutrinet's side
  • Get a better web barcode: Scandit for Web, but it's proprietary and costs money (they can do probono in some cases) - (I have seen the tech on Verif Covid)
  • Create a native app
• Another solution on our side: quick verification based on brand spaces (do robotoff brand insights match the barcode space?)

Part of

• 🧽 Data quality (tracker) #5538

[stephanegigandet]

A quick way to eliminate many issues is to look for the number of digits: 15 digits or 10 digits is probably an error. We can also check if the barcode is valid:

https://fr.openfoodfacts.org/editeur/nutrinet-sante/codes

[github-actions]

This issue is stale because it has been open 90 days with no activity.