I built a six-figure education company doing these four things:

* Learn and practice effective non-fiction reading techniques: http://personalmba.com/10-days-to-faster-reading/. Once you've got it down, reading a book takes 10-30 minutes with very good comprehension.

* Synoptic reading: read 5-10 books about the same subject from different authors in a short period of time.

* Note-taking: after synoptic reading, take detailed notes about the key themes and concepts, as well as noting ideas you have about how to apply or extend the material.

* Writing: once you've distilled the essence of the subject, write about it to lock it into your brain and share it with others.

The useful bits of knowledge often come at the intersections between topics, so exposing yourself to many sources of information is key.

It even works for technical topics: when figuring out how to set up my VPS, this approach helped me learn which combination of programs to install and how to configure them correctly, even though there wasn't a pre-existing setup checklist to use. By compiling many different resources instead of relying on just one, I got much better results.

Thanks. I have found myself bouncing around books on a common topic because there is often no one book that explains things Just So. It's been that way with learning Haskell. While RWH has been a major help, prior (and now continued) dipping inro a few other sources helps flesh out the bigger picture and makes all of them more clear.

* Don't force anything, have a list of different subjects that you want to learn (Ex: guitar, emacs, emotional intelligence, judo). Study what you feel like in this list.

* While reading, tag you book in interesting parts, but don't break you flow. Have the tags at hand.

* Use something like anki or mnemosyne to dump all your tags.

* Enjoy what you are doing. This is connected to the first suggestion.

* Sleep well, exercise at least 5 times at week, eat healthy, and socialize.

Thanks.

"Study what you feel like in this list."

That's part of the problem. Some items crowd out others. And there are a lot of items. :)

Since I often don't follow up on setting aside explicit time to focus and study and practice, I end up focusing on some things to the exclusion of others.

That may be a way of telling myself the effort isn't worth it. But I think it's really me just doing what seems easiest. While it helps to enjoy what you're trying to learn, there are future benefits to things that may not be that enjoyable now (like practicing the violin).

That's why I'm trying out assorted "learn in the background" sorts of things. Like running a flash-card app or playing German language videos while coding; it's sort of on the peripheral, but I still pick up some things.

I've been slowly getting better with my bash skills by organizing it as a class on my learning website: http://curiousreef.com/class/bash-scripting/

It's useful because it helps me stay on track. I can take off a few days or weeks even and then go back and do a few bash problems out of that book. I know that I'm 38% of the way through it, and it's easy to pick up where I left off.

I don't think I've done the peripheral learning that you mention, but I think 30 minutes a week is still low-impact learning.

Very cool. Better bash skills is also target for gradual learning. And 30 minutes/week seems quite doable, something not too hard to commit to.

One way I learn things is to commit to giving a talk. I learned a whole lot of Haskell while preparing for my Hubris presentation at MountainWest RubyConf, certainly much more than if were to just move along in my free time.

      (I also wrote a desktop flash-card thing to do the same 
      thing, sliding into view every N minutes.)

I've looking for something like this for Japanese Kanji.

anki

the cutup method.

while reading a book, do a text processing project with it. extract and read a large subset of the sentences, phrases or words, that share some interesting properties--semantic, phonological, syntactic, etc.

it's a good way to get primed for a new subject. the dislocated language is strange and memorable, if not informative. you gain a kind of reflective facility with the language.