by Petri Järvilehto
Imagine a cloud of dots. Each dot is a data point. These data points link to each other in various ways, and changes in one point may result into new connections between other points.
This is your “internal” or “personal” information. It’s what’s left when everything external is removed and you’re in an empty sensory deprivation chamber.
Now, we’ll add a book into that chamber. It shows up as a bunch of data points somewhere near the original cloud. As I’m reading the book, I’m making connections between the new datapoints and the existing datapoints.
Some of those new points are so interesting to me that they would gravitate inside the original cloud of dots, some of them would stay roughly where they are, but would be connected (since I could recall that the book had some info on the subject should I need to look it up) and some I would just forget, therefore not making any connection to those points.
WIth time, the connections that are not used will start fading away, and the datapoints that aren’t heavily connected or used will gravitate further from the information cloud that represents me.
The way I see thought process working in many cases is basically taking a bunch of data points and creating something new out of them. If I know that a+b=4 and I know that b=2 then I can create a hypothesis a=2 without any further information.
To further illustrate: imagine a wine expert. You can tell him that you’ve bought a wine from Chile from 1998 made out of tempranillo grape and he can instantly (without tasting or smelling) create a reliable hypothesis on what the wine will taste like, and what food you should serve it with.
This is the whole point of research. Having spent his life working with wines, the expert has a rich, connected cloud of information on and around the subject of wines. He can quickly deduct and even create new ideas based on that information cloud. Most of the stuff he doesn’t even have to look up, and if needed he’s guaranteed to have good connections to any question relating to his field (he’ll find it faster on the net than I would, or he’ll just pull a book out the shelf or check his personal notes).
While useful, I don’t think “just looking up data” is really the big deal. With that it’s possible to get a quick answer in a machinelike fashion (I’m serving asparagus risotto, what wine should I pair it with), but it’ll just create a single, isolated datapoint and connection that will not help me with any other similar issue and I won’t be able use that data in any new or creative fashion. To me this “fast data lookup” only becomes cool once it can be turned into something more than just automaton.
Whether the data is “internal” or “external” isn’t necessarily relevant (or even possible to define in many cases). I feel that it’s much more important would be whether 1) that data has been processed and something new can be created out of that, and 2) whether the data is shared or collaborative or not.
Anyways, that’s all a long way of saying that I like the definitions of “personal” and “shared” much better, but even with them it’s at best a line drawn in water.