Recently, an article was published on the effectiveness of taking notes by hand on paper vs. via typing/computer.
I found it interesting in part because it reflects something that has been part of my learning and creative habits, that I always assumed were formed mostly by the unavailability of cheap portable computers, and the ready availability of pen, pencil, and paper. Even though typewritten notes were more thorough, there was effectively no impact on the ability to remember facts when questioned a short time later, but there was a noticeable difference - in favor of those taking notes by hand - in how well ideas were retained.
When the experiment was run again, with the results being measured by a test taken a week later, the differences were even more pronounced.
Why is this? I don't know. Part of me has long felt that the time taken to write things out - since writing is muscle memory - forces you to focus more on what you are writing, and that the need to condense the information simply to keep up as you're writing it forces you to re-work and better understand the information. You also have the aspect that repetition and/or greater sensory involvement (tactile and / or spatial when it comes to diagrams and notes) helps improve menory and understanding.
The upshot is that I realize I've always done something like this. When I want to concentrate and actively understand something, I don't type out the notes. I doodle, or write them by hand. Or simply don't take notes so I can utterly focus on a conversation (parent teacher conferences, for example). When studying for advancement exams as a mechanic in the Navy, my practice was to read through once (getting an overall feel for the main poitns presented), to read through with a highlighter annotating the most crucial information, and then to go through the hilighted sections and make my own annotated handwritten notes.
I scored quite well.
Additionally - and this is a habit I see in a number of digital and 3D artists who grew up with ready access to computer-only tools - drawings, models, and sketches almost always start out on paper or other physical media before being scanned in to use as a starting point on the computer. Many many artists only convert to digital after the work is finished. Yes, I expect to see some changes to this with some of the excellent tablet-driven sketching programs, but then these programs work hard to provide the feedback and feel of a piece of paper and pencil/pen/paint.
Finally - whether it's mapping out roles in a program, or the functions and hardware in a network, that almost always is first done on paper as well, regardless of what drawing tool (Viso, the google drawing app, Omnigraffle, etc.) is used.