On how to learn without getting a blinding headache.

[reddragdiva]

"Think!"
"Buh."

I have never been the greatest learner. I may be searingly intelligent, but systematic learning has always hurt my head. Unless I'm deeply interested already, I can't just read something and have it flow straight into my brain.

This stands me not too well at work. I enjoy my current job — Unix computer systems administrator, company with weird layers of heterogeneous systems, just bought a couple of other companies with their own weird layers of heterogeneous systems, I need to gain proficiency in most of it in reasonably efficient order — because there's a lot to learn, and fast. But I come home feeling like my head's gone five rounds with Mike Tyson. It's preferable to workplace boredom, but it's not preferable to just about anything else.

The trouble is that learning stuff quickly feels like pushing a large, heavy rock. Against considerable friction. Once I've learnt something, I'm fine, but the process of learning it hurts every time.

(The clearest example was learning mathematics at university. Once I'd taken in something, I was good to go. But the process of taking it in felt like I was carving every symbol into my forehead with a chisel. The ease of having knowledge versus the pain of gaining it.)

The Internet lubricates the path to knowledge by being the first place to look once you know the right question to ask. So how do you formulate the right question? And when you can't find the answer easily, how do you break it down into other questions?

There are various quick answers. One is to push ahead in a bloody-minded manner. This works to an extent, but leads to mental exhaustion out of proportion to the results. (It doesn't help that there isn't anything so simple as a clear curriculum.) Another is to find a suitably strong motivation. But when that comes down to "making money to fund my suburban lifestyle," it's just not that convincing a carrot, picture the stick as I might.

(I could try to find a career path I like better, but this one is the only field I've ever found that doesn't make me want to go postal.)

The question I'm asking is not how to push harder, it's how to reduce the friction. And accompaning smoke and screeching noises. Have you had this problem? If so, how do you deal with it?

Comments

[thorfinn]

IMO, the secret to being a "quick learner" is to correlate, correlate, correlate. All the quick learners I know are inveterate punsters... Because we're quick to link everything with everything, and don't mind the gaps. There's a story that goes around about a primary school teacher who was achieving amazing results with reading comprehension... When they peered into her classroom to find out why, they found an odd thing. She was sitting there, knitting, for most of the class. Every now and again, a kid would come up from their desk, she'd look up, say a short phrase, then send them back to their desk to continue reading. The phrase? "Say moses, and read on."

The trick to inhaling stuff fast is just to read it fast, skip over the stuff you don't grok yet. Then re-read it a bit later. And again. It's all about exposure... Your brain will grok the bits it can grok on the first pass, and settle the other stuff into the background for "next time". If you want to have an easier time learning stuff in a particular field, just get a lot of random books in that field, and read 'em. Don't worry if you don't understand them... it doesn't matter. Say moses, and read on. It's all about laying the foundations and jargon groundwork for your brain to work with later.

The sole problem with this approach is that you have to lay the groundwork as far in advance as you can. If you're trying to catch up in a particular area in a short time, then brain grinding hard work (to the point where normal language falls out of your head) is the only way to go. Happens to even us "quick learners" when we inhale a large chunk of stuff. It's just that we're used to it, and can go "oooer, brain grinding, better switch tasks momentarily."

[thorfinn]

Wups. The question, of course, was "Miss, what does this word mean?"

[ewen]

The trick to inhaling stuff fast is just to read it fast, skip over the stuff you don't grok yet. Then re-read it a bit later. And again.

Amen.

Accepting that you're not going to understand it all in the first pass (or second, or third, or...) is critical. As is giving your brain time to recover by doing something else -- there's a whole bunch of "background processing" that goes on to help form associations between things which just takes time.

Rather than struggling away at the same thing it's important to take another angle at it. Reading several books on something helps (each one generally covers some angle better, and the multiple descriptions helps with building associations). Or Google. Or whatever. The answer doesn't lie in the single document to be puzzled out; it lies in the commonality (or differences) between the various descriptions.

Going away and doing something else any time the brain gets "behind" definitely helps. Even just grabbing a cup of coffee or chatting to someone about something else for a couple of minutes gives time to "catch up" a bit. If you're feeling "further behind" then lunch, or overnight, may be required. I even used to do this in tests -- read the "hard" questions and not try to do them until I'd done the easy ones. Then by the time I got back to them the "hard" questions weren't so hard (you do, of course, have to leave enough time for them!)

But accepting that it won't all make sense immediately, and that this isn't a problem, is the key to it. Plus a willingness to "wing it" on partial understanding once you know enough to know what you do and don't understand.

Ewen

[horngirl.livejournal.com]

That makes a surprising amount of sense in my own experience, I've just never heard it articulated as such :-)

[thorfinn]

It's one of those things that is articulated if you have academic parents. It's the big reason why kids of academic parents do better at school... They're often taught how to learn (not always - there's the fish have no word for water effect sometimes), as opposed to just being thrown into a "learning environment".

[horngirl.livejournal.com]

Nope, parents definitely not academic. They did teach me to read very early though (I was about a grade ahead of everyone else in primary school), and instilled in me a love of reading I will probably have for the rest of my life.

That, and never underestimate the way competition can develop between siblings. Being the youngest, I had to be the best (at least, that was how I saw it).

[juliann.livejournal.com]

My ADD makes sensory input difficult. I can formulate questions, but cramming material into my brain sometimes happens (hyperfocus) and most of the time doesn't. However, you can increase your chances by taking Omega 3s (or omega 3 6 and 9s in the form of hemp seeds or oil), ginko biloba can help with memorization too. Both of these are always-take rather than pop-right-when-you-need-it but are generally good for you. Also be sure you're getting decent vitamins, etc.

I have a book on optimal nutrition for the mind if you want to borrow it :)

[hestia.livejournal.com]

I know exactly what you are talking about. There are heaps of different ways to help yourself along. I can think of 3 straight away.

1) Make sure you are healthy, particularly that you have enough iron in your blood, and you're getting good quality REM sleep.

2) Association maps. Remember the infamous and evil acg and Jika Jika fuck maps with lines connecting everybody to all the people they've done? Adapt that idea anyway you like to your computer systems. I find these utterly invaluable in understanding any uni subject and writing the essays for them.

3) Hardest of all: stop resisting the process, relax and trust in your unconscious to do a good part of the heavy lifting. This is connected to the REM sleep where the unconscious gets to run rampant making sense of your day.

[topbit.livejournal.com]

2) Association maps
I heard these as 'Mind Maps', developed (or at least popularised) by Tony Buzan. http://freemind.sourceforge.net/wiki/index.php/Main_Page is one good (read: free - libre and beer) mind-mapping software I've seen.

[arkady.livejournal.com]

Whilst in training to be a station assistant for the Underground we spent a whole day going over mindmapping. I was told that apparently I mind-map naturally....

[dennyd.livejournal.com]

I wonder if that's something that programmers tend to do easily. I had my 3 month review at work recently, and one of the comments was that although they were very happy with my technical skills, what really impressed them was my problem solving abilities - the application of the right bits of what I know (or finding out something new) to solve the problem on hand. I've always though this was one of my strengths, and I suspect it's something that's common to a lot of (what I think of as) good coders...

[reddragdiva]

Yes, that happens to be exactly how programmers think.

This does not apply fully to operating systems administration - the individual pieces are often technically coherent and aesthetically pleasing, but there's too much legacy cruft and (with commercial systems) failed marketroid fantasies for a truly coherent system. This is why you can't really understand Unix, FreeBSD or even GNU/Linux without knowing a fair bit of computing history.

[ewx.livejournal.com]

Read over again (and if necessary over and over again), and get hands dirty (i.e. scribble/experiment/whatever fits the domain). Have a particular task in mind, skim the bits that don't seem relevant (but don't ignore them completely - you might find half way through that they are relevant after all). You can pick up the bits you skimmed when you've got a framework to place them in.

[arkady.livejournal.com]

I'm not using my brain at the moment. Would you like it instead?

In all seriousness, I can't fully explain just how it is I manage to "devour" stuff as a fast learner. I think a lot of it comes down to relience on my (pretty exceptional) memory; I read it, and even if it doesn't make sense I trust it will later.

I do have the motivational issue though. I can't just learn for the sake of it - I get bored too easily (hence giving up on Italian after only two weeks - just couldn't be bothered to put the time in to carry on); there has to be an actual reason why I need to know something. I picked u what I know of Javascript that way - I needed to know how to do foo, so bought a book on the subject which I didn't really understand but read anyway, then found examples on the internet of what I wanted actually working - at which point my memory kicked in with what I read and said "Oh, so that's how it works!"

It's as I said about languages; the trick to learning a language is to start thinking in that language. Your problem here I think is working out what the "language" is in the first place before your brain can kick into thinking in that pattern, letting your memory come into play.

Oh, and my memory wasn't always this good. I was worse than you, in fact. I taught myself to have a good memory; and I started with journal keeping. Writing down everything that had happened from memory.

OK. Hmm. Not sure if that was a Death/Dream or a Delerium/Dream moment, but hurts now so I go chase mental butterflies for a bit and hope that helped.

[redcountess]

Break up what you're learning into small chunks - make screen dumps of references you can take home and read, yes, this does use up paper, but you can later use it for scrap (and you download other reading material anyway *g*).

Get a highlighter and mark out vital points, so you can go back to them later.

When your brain is full, don't keep trying to push yourself, change to another task.

Also, don't sit at your desk at lunch - go for a walk, and rest your eyes every so often. What others have told you about rest and nutrition is what I've been trying to say to you for months - you need to get to bed earlier and eat better.

[hawkeviper.livejournal.com]

Damn straight... and when you can't get your head around something, take a break... it'll be easy 10 minutes later. ;)

[inulro.livejournal.com]

You're asking me? "Making money to fund the suburban lifestyle" has come to a screeching halt, and I have to get a career before my brain implodes on itself. All these years, I've been using it & I didn't realize.

[tcpip.livejournal.com]

Hmmm... A lot of interesting and valid suggestions.

Thorf is perfectly correct in terms of cramming a basic understanding of issues.

Recountess is correct in terms of breaking the large components down into subcomponents.

Association maps, whilst good for the creative arts, are not good for technical tasks or engineering. "Association" refers to structural connections, whereas (nearly all) technical tasks refer to causal connections.

At the end of the day, you are referring to tasks. This means practise.

Cram first. Then subnet your task into a workable size and test the theory. Once you have a complete grasp of it, expand.

Yes it appears more time consuming and it is on the initial stage. But that's what a step learning curve is all about. It's almost impossible to make it in a single bound.

[thorfinn]

Yeah... Actually, I second tcpip says about mind mapping re technical tasks. It doesn't work particularly well, because the connectivity isn't really like that. You have to be willing to make testable hypotheses on your existing understanding, try them out, and see if they work. This means a scratch area, if necessary. If there's no scope for a scratch area, then you need to try and find domain experts, and pick their brain directly, with questions of "so, show me, exactly, how you do X", before going through it yourself. If you can document that process (or bring a tech writer with you to document it), then so much the better.

[reddragdiva]

Yes. Consider we're talking about Solaris here. It isn't something mappable — it's something packed. It's got lots of good bits, but lots of marketroid fantasies, shells of failed technologies and dead ends. The difference from Windows is any good bits at all.

[thorfinn]

Yeps. The shtick to Solaris is that someone at Sun decided that moving to SysV is a good idea... but there's still that old BSDish SunOS hiding away in dark corners... plus, as you say, some shiny marketroid fantasies whacked on top. Dunno if you have access to Sun's issues database, but that's worth getting into if you can. Also I think there's news:comp.sys.sun.* (or some similar hierarchy name) which might well be worth just inhaling chunks of for a while.

[hawkeviper.livejournal.com]

I know this is absolutely the last thing any sysadmin would say... however, I'm not a sysadmin - I am the sysadmins enemy, therefore the way I learn is backasswards... when I'm working on a box/network I know nothing about, I tend to learn on the fly... most things are just slightly different to things I already know, and therefore, using logic and google makes it a lot easier than enumerating the system beforehand and trying to learn how to use POSIX or BSD from a book.

"Empty your mind, be formless. Shapeless, like water. If you put water into a cup, it becomes the cup. You put water into a bottle and it becomes the bottle. You put it in a teapot it becomes the teapot. Now, water can flow or it can crash. Be water my friend." - Bruce Lee - I'm damn certain that holds true for techs too.

[feanelwa.livejournal.com]

I don't know how it works, but I know how I make it work better. These are the things I do:

1) I have a constant part of my brain looking at everything I read and thinking "Is this in the best possible logical structure? Would it be better if I put that section before that one, or stuck those together?" I always, always make things into small chunks before I even try learning them. You cannot eat a potato without dividing it up.

2) I keep my digestive system on a very, very light load when I want my brain to work well - lots of fibre, lots of fish and as little meat that walked as possible. I don't know why it makes my brain work better, but it does. Maybe there are fewer large fluctuations in blood distribution or something. I also take iron tablets :)

3) The most important rule - you do not control your brain, you only live in it. It rearranges itself to hold what it needs to know, and its opinion of what it needs to know is often different to yours. If it evaluates the knowledge you are trying to give it and finds it wanting, it will not expend the effort to make a space, no matter how urgent it is to you. Learn how your brain values different kinds and topics of knowledge and learn accordingly how to disguise things to make them look valuable enough to remember.

[thorfinn]

Heavy proteins sit in your stomach and intestines for a lot longer, which means your intestines retain a lot of blood. This is not conducive to thinking, because you want a lot of blood up there in the brain, not in the stomach. OTOH, you really don't want to be hungry, because heavy thinking chews through bloodstream glucose almost as fast as physical exercise does... Low GI food is good.

[feanelwa.livejournal.com]

I have to say, stopping eating white bread has made things slightly clearer up here. I still eat sweets throughout the day though, mostly because in this weather I use up so much energy keeping warm that I need every joule I can get back again.

[hobbitblue.livejournal.com]

This is something we had to figure out recently - I went the traditional route, school, college, university and learnt good study skills along the way. My partner forestwalker left school at 16 and after doing a few evening classes here and there suddenly signed up to do a science degree with the Open University 25 years later, and the first year we spent working on study skills and how best to approach things. Its been interesting seeing how many "learning" type things I take for granted, and which are completely alien in approach for him.

So, my suggestion is to work out how you personally learn stuff best, and then learn new stuff in a similar way. For example, I assimilate large chunks of text rapidly, can skim read and take in the relevant points and see patterns and linkages with my existing knowledge; I can do this with any subject and can "translate" stuff outside my field enough this way. Forest on the other hand works best visually, he likes mindmaps and diagrams and finds writing things out in notes helps, and having lists of things to refer to. In the past I've used mnemonics and rhymes and even setting things to music to memorise basic stuff, my brain seems to like that, he just can't see the point.

Just pushing ahead through deadly dull doesn't get you anywhere. Break things into small chunks and if you can't concentrate, go and do something else and then come back to it. Another tip is to read actively - don't just have a book or manual in front of you, be scribbling notes or underlining stuff, or making summaries or checklists. Try music in the background, classical music especially can help with learning stuff, and most importantly if you get stuck on something don't sit worrying away at it for hours, move on to the next thing, or sideways, and then come back later. Sitting staring at the same problem all day just makes it seem bigger and more impossible (still haven't persuaded Forest not to do this as yet).

Oh and if its practical programming stuff, try the examples, physically typing them in to see the results, even the basic stuff seems to stick better if your brain can see cause and effect rather than just theory.

[belegdel.livejournal.com]

I suffer from a different problem but the solution might be valid in your case. I tend to fogure things out quickly but retain very, very little. My poor memory is legend (though, oddly, I can quote "Aliens" and "Excalibur" almost in their entirety)

So, reading expansively on one topic is largely a waste of time to me. Instead, I tend to go after very specific solutions to the problem at hand and follow up on surrounding context only if I can't grok the solution without it.
It tends to leave me with patchy knowledge but gets the job done. In some ways it implements Thorf's suggestion by leaving the other stuff for assimilation at a later date and redcountess' idea in that it chunks the learning. However, it also accelerates the process by focussing it.
The down side is that it does risk the occasional fuck-up through missed details and requires a bit of "try it and see". Perhaps working at a Uni has spoiled me a bit in terms of having the time to "play".

Anyway, that's my suggestion. I've managed to drag myself from a Perl CGI programmer to a passable Solaris admin in about 2 years this way, so it can't be completely ineffective (unless of course that's way slower than normal :)

[beckyzoole.livejournal.com]

Sleep. Eat right. Drink lots of coffee. Get in and do things hands-on.

Also say things out loud -- I tend to remember expressions much better if I say them out loud.

[morganjaffit.livejournal.com]

Lots of good suggestions, but they're mostly people telling you how *they* learn. Which is only part of the battle. People tend to have different manners in which they learn best - and by the sounds of what you're saying, you don't learn by reading so well.

Most of the suggestions seem to be pretty reading focused - I'm not sure of the best way around that, but it would probably be worth looking down other alleys, methinks.

[vatine]

These are not so much "learn everything", these are "coping skills" for the interim.

"Find a spot you feel you almost understand. POlish it until it makes sense. Move to the next spot."
"Learn how to use man(1) to your advantage" [s/man/pydoc/ and you have me giving python advice to someone with 3 years of python experience after 3 months]
"Learn how to use man(1) to your advantage" [this is not optional, having a vague idea what would do what, approximately, is a good thing and learning how to go "this almost does that, let's check references" is Not Optional, it is Mandatory]

Another thing to do would be to dump the contents of /usr/man/* to a printer and just bloody read 100 pages a day (skim-read!). That's roughly how I initially learned unix. A binder of the full contents of a version 6, lying on my back in bed, reading in multiple times, garsping just a tad more each time.

[ heterogenity ]
The pain problem there would be, I guess, odd interactions between Wedge A and Slot B (what with A not being a tab and all). And the cognitive dissonance of almost-buit-not-quite similar unix boxes. I, for one, don't actually notice that. In my first job, I herded D-NIX, HP/UX7, HP/UX8, AIX2, AIX3, SunOS4 and (when no one else was around) VMS. Oh, ah, yes, the lab machine ran Interactive 386/ix. We learned to write "portable code", we did. And find and rely only on a suitably portable subset of shell commands.

[sushidog.livejournal.com]

Hm, not sure I'm the best person to comment on this, since I'm not a coder; my learning tends to be factual/associative, and more focussed on understanding how things fit together, than remembering how to do something, if that makes sense?

However, since you asked, how do I learn? Well, I guess partly, I talk to myself, and explain it to myself; that is, I read it one way, and re-phrase it another. I draw lines from one bit to another (not literally; I've never been into mind maps, although the OU is very keen on them; I'm just not a very visual person) and link what I'm learning to what I already know. Essentially, I try to understand it, rather than memorising it.

I guess I use these same tactics when teaching; my style is associative and anecdotal; I'll illustrate points with stories from personal experience or whatever.

As an example, there's a psychological construct called Locus of Control; I remember that very well, largely because during the lecture on it, I doodled a large grasshopper in thigh-high boots and long gloves, clutching a whip; the Locust of Control. I've never forgotten it, and it has meant that the concept behind the pun (I notice someone else mentioned punning up above!) has stayed with me too.

[reddragdiva]

I certainly hope you saved a scan of the original doodle for your future lectures to undergraduates.

[heliumbreath.livejournal.com]

Looks like lots of good advice already. To that I add: Recover. Make sure you get in some decent Recovery everyday.

And I think stuff has to come together chaotically; if you stay on one thing until you're squirming you're past the optimum point, so jump to another spot, see how it all fits together, muck with things a lot, keep decent notes and scribbles. Don't be afraid to use paper.

Bottom line is that I'm cribbing hard from this thread, since I'm recently Just A Unix Admin again, and can shed the netadmin and sysprog chunks out of my head and try frantically to replace them with whole swathes of Unix that I've never come closer to than hearing people rant about them.