When you are trying to replace a technology like paper, it helps to know how it is used. You can roughly split paper usage into two categories. What you put on the paper, and what you do with the paper. For the first question we knew we had to build a surface for recording notes that would work for ink or text (and even audio, pictures, etc), and would be as flexible as paper – or as close as we could get. But what did that really mean? We needed to know what real people did when they wrote notes. Some people guessed that there would be a lot of hierarchical notes (those of us who took notes that way). Others guessed there’d be mainly messy notes, still others thought there would be a lot of diagrams, and so on. We could argue in our hallway, or we could find out what people really do.
We started out with an effort to collect handwritten (and typed) notes. We asked about 1000 registered users of Microsoft products (who had agreed to be contacted in advance) to send us a manila envelope full of their notes (whatever that meant to them), in return for $20. We got notes from about 500 people, which we posted on the walls of our hallway (picture to come later). We categorized them into groups. People who use outlines (hardly any, confirming my bias), people who make lists, people who mix drawings and text a lot, people who write paragraphs, people who use no discernable structure, etc. We also had classes of notes like post-its, course notes, and interview notes. And notes from other countries, lawyers, notes from meetings on handouts, etc.
We put these notes on the wall because I wanted them to be easily accessible. If we ever had a discussion about whether “users do this” or “users do that”, it should be as easy as walking out into the hallway to check. And we did. Many arguments were settled pretty quickly. The “wall o’ notes” was one of the more successful things we did in the OneNote project.