How to keep track of what you've learnt | Hacker News Reader

How to keep track of what you've learnt

freestylemind.com

35 pointsoscardelben30 comments

This is a different method, but for the last few years I've been using Personal Brain (http://thebrain.com) for note taking on pretty much everything I read. There's a free version available that is unrestricted enough for most uses.

Think of it as a mind map where every node has the potential to be the central node and where connections can be made from any node to any other. Mine, for example, has several thousand nodes on topics from everything ranging from science to literature to my business ideas to pretty much anything else.

Having everything in one place is great because I find that I'll often stumble on something I added months or years before and I can associate it with what I'm doing now. It often exposes patterns, reminds me of stuff I'd forgotten and it's fun to explore.

I haven't found any other software like it. I'm not particularly anxious to replace it but I think some competition would be great. A web app or something non-Java would be interesting.

I've written about it more here, and in several other blog articles: http://marcusvorwaller.com/blog/archives/2008/07/18/personal... . I also did a quick video of how I use it awhile back http://www.youtube.com/watch?v=eTNlRPAq3gE .

Hopefully that doesn't sound too much like an infomercial, I'm not associated with PersonalBrain in any way, but I do really, really like using the software and I'm surprised it isn't used more widely.

It'd be cool to make a whole desktop UI around that. Very cool.

If someone made a really intuitive web app for this I would totally use it. Usability would be key though. Painless to enter data, easy tagging system, and easy wandering system to keep on top of everything.

Making an intuitive app would be half the problem. The other half is finding enough customers to pay for it - there is no obvious way to segment and/or target users, other that a single life-hacker post that you're going to get.

Make it free for users and work out how to monetize the aggregate data (sell it to Google?) somehow

Here is my app. Needs some more work on the really intuitive part: http://flashcarddb.com

looks like the '90s

How about some more constructive criticism?

make buttons bigger and use more shiny colors. Better yet: get a designer.

I tried this. It failed due to lack of revenue. I was trying an ad-supported model.

My problem would be trusting all of this information to a third party. At the very least, I would want a means of exporting the information in some open format to my desktop.

Use mnemomesyne for that: http://www.mnemosyne-proj.org/

It is a great tool for learning and works with an efficient algorithm. Make it a habit for learning every morning or every night.

I always find it interesting that say if someone asked me to describer every article I read on HN in a day I wouldn't be able to but if conversations with friends come up where each of them were relevant I'm usually able to describe there point of view and information and detail.

Also it's similar with studying, I will go through the material and write all the key points down. I won't actually read them again, unless I forget something relevant later in study. I find in the exam when I'm given a relevant question I have near perfect recall on what I've studied no matter how large the amount of knowledge that I've gone through in a couple of days.

I write down things I want to remember later in Emacs org-mode files. It would be easy to set this up with org's SCHEDULED marker.

Just make a macro to emit a bunch of SCHEDULED markers for suitably spaced future dates, run it on new items, and review the items when they come up in the agenda in the future.

And looks like someone's making an add-on to org-mode that implements the spaced repetition with an actual Supermemo-style algorithm instead of an inflexible fixed scheduling: http://www.mail-archive.com/emacs-orgmode@gnu.org/msg18554.h...

I don't think this is necessary. If something is boring, you'll forget it. If something is interesting, you'll remember it. You'll think about it a few times, and perhaps discuss it with other people. You'll link it mentally to other parts of your worldview. All these things will increase your ability to recall it when necessary, without effort.

To improve memory generally, stay fit and live a sober and useful life.

NB: I'm leaving this comment because I'll have forgotten about the article by tomorrow :-)

I guess, if you want to only know basic stuff. But engineers, programmers, physicists (etc..) can benefit from maintaining a large knowledge base of stuff that they wouldn't find interesting or have a conversation about. If you only try to remember stuff that interests you, you probably won't be as successful as someone who can remember even the uninteresting stuff.

Besides, knowledge is a VERY powerful thing.

The ability to forget certain things is important. Knowledge advances by rejecting false theories.

Scientific and engineering knowledge fits together because reality is consistent. So it isn't necessary to memorise everything.

Rather than trying to 'drill in' certain pieces of essential info, it's better to improve the state of your mind, so that your memory works better generally. We waste a great deal of vitality by overeating, watching TV, etc. It seems to dull the mind and the memory significantly IME.

Forcing yourself to memorise stuff that doesn't interest you will tend to shut the mind down. It's like being at school.

One thing he didn't mention that helps is learning and reviewing similar things together. If you learn things that are too similar at different times they can interfere with each other; proactive and retroactive inhibition. If you link them and learn them together, they will not only not interfere with each other, but can help reinforce each other.

Interesting choice of image. The gears cannot turn.

http://www.freestylemind.com/wp-content/uploads/2009/11/brai...

Maybe I'm being obtuse, but why not? The radius and tooth count appear to be the same on all four, and the directions of rotation match if each turns opposite the two it touches.

Because the gear movement vectors add up to a vector equal to the applied force, but opposite in direction :-D Remember, the magnitude and direction of a vector are the sum of the magnitudes and directions of the vector components. If you applied a given force to any of the gears, in any direction, you could tabulate the resulting forces and directions and add them up. What you will have then, for the image, is a four quantities of equal magnitude but where the odd numbered quantities have opposite sign than the even number quantities. Classic parallelogram.

Say, you applied a force to the green gear to move it clockwise, let's call it +N (for positive force of magnitude N; we chose +ve to mean clockwise.) All four gears are of equal size, so they have identical torques; they will all move by the same speed. Taking friction as negligible, we can expect any gear to produce a force of the same magnitude as that applied to it, but with opposite direction.

So back to the green gear, when we apply a clock wise force of magnitude N, it gives us back a counter clockwise for of N, which acts of the yellow gear (or the red one, if we choose to.) Yellow one is acted upon by +N, and it in turn gives us -N back (a counter clockwise one.) Feed that to the blue one and you get +N, which acts on red to give us -N.

Now, a force of -N is exerted by red upon green; BUT! we are applying a force of +N to green! What's going on?

The two forces cancel out and you have nothing.

If you apply clockwise force of magnitude N to green gear, you would be applying +N. Which would apply -N to yellow (not +N as you stated), +N to blue, -N to red, and +N to green again, thus rotating the gears. You obviously have a better understanding of this than I do, but you must be explaining some part of this equation incorrectly.

Ceci n'est pas un engrenage? :)

I believe you to be mistaken. The green and blue can turn clockwise, the yellow and red anti-clockwise.

They can turn, but it requires an external driving force.

I don't see a single mention of long term memory in this article. Cognitive psychology is not a grab bag to use for marketing.

The idea is pretty much spaced repetition learning (http://en.wikipedia.org/wiki/Spaced_repetition), even though article doesn't call it that or make any references to the existing work.

If you want to master a new technique on your sport, the best way is to repeat it as many times as possible until it becomes second nature for you.

Wrong. The best way is to create as many as possible associations of different kind - abstract, visual, verbal. The net of connected associations (or a mental map) is the key, not a while loop.

Write a list, picture your associations. Read aloud and capture it to a webcam. Then listen and watch yourself. That's much more useful than repeated reading.

Well, you're both right, as you are talking about two different types of memory.

Associations are best with "normal" memory, but with sports (or piano/instrument playing, or touch-typing etc) we're talking about muscle memory, and there is no way around repetition-repetition-repetition

Sure! That repetition is called practice, which is the best teacher.

Reply on news.ycombinator.com