Within /store/paperless, I have a directory structure like this :

• inbox (scanned PDF get saved here before I move them to their final location)
• 2008
  • divers
  • poste
  • swisscom
• priv
  • assurances
  • finances
  • loisirs
  • santé
  • voitures

When saving a file after the scan and OCR process, I name it $year.$month.$day-subject.pdf. Example : 2008.11.02-pinkrus_moleskine.pdf would be a bill from my purchase of a Moleskine agenda from Pinkrus dated 2nd of November 2008.

OS X Spotlight index all my files by name and content, so a few words into the search box of Spotlight brings me to my documents very quickly.

Once a document is scanned, it goes to paper-recycling or shredder depending on its content.

My next post will describe my backup procedure. Throwing away the paper version might be cool to regain office space, but a serious backup is a must.

At my company, we have a great joke called “STP”. It simply means “Spare Time Project”. It’s the cool stuff you would like to be paid for, but which you do in your spare time for free, just because you didn’t achieve to convince the customer that he definitely needs it.

Over time, we have had many great ideas for STP’s :

• build a Netnea-Linux distribution.
• build a Linux appliance to host these cool services like anti-spam, reverse-proxies or wind measurement system for the sail club.
• recognizing payment forms using the Macbook Pro embedded cam.
• and a few other I don’t remember or can’t mention here, because we might well do them and get rich, fast

In the last months, I have slowly moved over to a GTD-oriented way of working. It involves reading a few blogs like Lifehacker, 43folders, etc. They kept mentionning all the benefits of going the paperless way, so I bought the scanner mentioned everywhere and started working with it.

It’s the Fujitsu ScanSnap S510M. Notice the “M”, it define the Mac version. You surely don’t want the Windows version, as nobody runs Vista anyway, or so I have been told. For the Linux people : apparently some support exists with sane-fujitsu, but the real value you get with the S510M is the bundled software :

• Fujitsu SnapScan Manager (in fact, that’s the driver)
• Adobe Acrobat Professional v8
• a limited version of Abbyy FineReader v3

What is amazing when you start to scan pages is the speed of the beast : pages flow through it just like they would be ejected to get rid of them. Wire-speed it is, this thingie !

A PDF file is produced for every document (or document set) you scan, and then comes the software, asking what you want to do with the generated PDF :

• save to folder
• attach to an e-mail
• print (copy function)
• send to iPhoto
• OCR using FineReader

This last function works hard on your scanned document, and put words behind the image, so you search, index and classify your document. A nice piece of equipment indeed.

In the next posts, I will detail the process I use to classify, sort, index, backup and search the scanned documents.