So today, on the drive home from a doctor's appointment, I was daydreaming about my ideal note taking app. You know, like a normal, well-adjusted person would do. I don't even approach the level of competency in computers necessary to create something like this so if anybody else wants to take the idea and run with it, all I ask in return is a free copy of the hardware and software through the beta and final release process.
First of all: This app would run on a tablet. Visually, it would just be a blank white screen, and you'd write on it with a stylus. The device would probably have to have a digitizer, so that it could respond to pen pressure and so you could rest your hand comfortably on the device without adding extra random lines.
But it would also have a recording component: when you start taking notes, unless you shut this feature off, it would be recording audio of the room. Video too, if it can be set up to do so functionally.
So, you start the recording, and start taking notes. That process goes pretty much how you'd expect.
But after the fact, when you review the notes, you can select any given piece of writing -- any line or phrase -- and the device will have saved timestamps corresponding to the recording, and can play back the audio (and video) of the subject at exactly the time you took those notes.
There would be as many available interactions between the recording and the notes as possible. You should be able to save new layers of notes over the original layer, corresponding to separate events of review of the material. You should be able to ask the app to highlight areas with a particularly high density of notes by timeframe, or spots where there are longer than average time gaps between notes.
So if there's a chunk of a lecture where you kind of nodded off, you could go back and click on exactly the part of the recording you missed.
It should be easy to insert a new section of blank space after those spots, sliding the later content down the page, so you can re-take the corresponding notes, in continuity.
A good voice-to-text engine should be part of the package, so if you want to transcribe the content, you can run that on the recording then listen through for corrections instead of having to transcribe word-for-word.
You should be able to look at the recording itself as a timeline, with Soundcloud-style comments distributed along it.
Anybody up for making that for me? (And, you know, students and journalists and note-takers everywhere?)