CM

I thought I’d share the basics of our new at-home scanning workflow:

• All lettermail is scanned via a ScanSnap ix1600¹ at the point-of-origin (basically as soon as it’s brought in)
• The ScanSnap uploads to a ‘Scans’ folder in Dropbox²
• The Scans folder syncs to an older Mac mini running a few utilities.
• Hazel processes the files:
  • Renames them with a date in the format YYYY-MM-DD_HH-MM-SS
  • Runs OCR on them using OCRmyPDF
• Keep it shot³ then reads the OCR and uses OpenAI’s API to:
  • Append 3 descriptive words to the filename
  • Append the addressee to the filename

Then anyone in the household can grab the scanned file when they’re ready.

I’ll eventually add some additional processing via Hazel to sort into folders, but that’s a future-Chris problem.

1. I’ve been using ScanSnap products for over a decade now and they’re fantastic. My first was the 1300 – which I recently moved to my office, which prompted the purchase of the ix1600. So yes, I was using a single ScanSnap for over a decade. ↩︎
2. This could really by any cloud service that’s supported. We still use Dropbox because it’s sync is still best-in-class and it’s handy for sharing random files with people. ↩︎
3. Technically, Keep it shot is invoked via a Keyboard Maestro automation. Clunky, but it works. ↩︎