Lessig’s Plan To Fix the Government

I went to see Lawrence Lessig as Long Now last week and I was a little disappointed by his proposed solution to government corruption in the US.
  1. Amend the constitution
  2. Public campaign financing

He gave a rousing talk however and at the end offered this challenge.  Though it may seem hopeless for us to curb the influence of money on public policy, consider this thought experiment:  Suppose a doctor told you that your child had brain cancer and that there was nothing you could do.  Would you really do nothing?

The problem is that partisans are not really focused on the question of campaign financing and are too busy fighting one another.  And the rest of us are frozen with apathy or hopelessness.

For me, it’s simply a matter of style: http://rootstrikers.org/.

So what is meta-learning about?

Last week I was busting on (4-hour workweek, 4-hour body author) Tim Ferris for getting meta-learning (learning about learning) all wrong by focusing on memorization. Even Ferriss acknowledged that smart phones would be a globally disruptive force to transform learning.  To me the logical consequence is that memorization will become less and less important.  Even learning languages will be less important.  The military is already trying out realtime language translation devices.  I do hear that Google’s translation API is running into some snags because so many people used it that it started using it’s own results as input. (Which puts a whole new dimension on garbage in-garbage out.)  However, that’s probably a short term glitch.

So if memorization tricks aren’t the key to meta-learning, then what is?  Well internet search skills are an obvious candidate.  I’ve had several conversations this past week where we discussed how fundamental search is to work.  I can’t imagine any field where a worker could compete if denied the use of Search.   We can discuss some of the ways knowledge is becoming debased on the internet another day.  Today it is definitely possible to find useful information to make work easier and better in almost any field.

I haven’t spent much time researching search (yet!) But here is how I conceive of search right now.  I would break the problem into some general categories:

  1. Keyphrase craft – including and excluding the words most likely to return relevant results
    1. Using filters – knowing how to restrict results by date, category, or even a single site
  2. Results relevance analysis – determining how relevant each result is by scanning the results summary text and feeding back to recraft the keyphrase if needed.
  3. Evaluating  sources –
    1. how trustworthy is this site?
    2. if I trust the site, how useful or relevant is the info?

So we can argue about this characterization or improve it with the state of the art research (by using search of course).  But I don’t want to suggest that I think meta-learning is all about search.  I personally deeply rely on dialectics for learning.  It’s one thing to read a book or find some good search results, but you really start to understand what you read when you talk it over.  So it’s essential to:

  1. have relationships with people
  2. pick the right person to discuss a given topic with
  3. Clearly express your ideas
  4. Listen to and be receptive to opposing viewpoints (within reason of course – see #2)
Finally, we can’t ignore that Constructionist idea that Robin exposed me to that the best way to learn is to do.
“What I cannot create, I do not understand.”
― Richard P. Feynman
For me the useful handling of mistakes is a really essential and largely overlooked aspect of learning.  How can we learn without making mistakes?  There is much more to be said on this topic.
Anyway, the bottom line is: Meta-learning= learning about searching, speaking conversing, and doing.
Am I wrong?

Ferriss at Long Now

I was a little disappointed by Tim Ferriss’ talk at the Long Now this evening.  The topic was meta-learning which seemed like a subject with a lot of potential.  It’s hard to ignore the accelerating change all around us.  We must continuously learn new skills and knowledge to keep abreast of the latest developments in all fields.  So meta-learning should be incredibly useful.  This also ties into the “Lights in the Tunnel” meme-cloud about automation removing consumers from the economy.

The problem was that Ferriss presented a heuristic approach to decomposing and learning new subject matter which seems most applicable to rote memorization tasks or physical skills.  All of his examples revolved around language acquisition or sports such as weight lifting or swimming.  It was hard for me to see how his approach would help someone develop deeper understanding of complex subjects or even learn the new job skills demanded by our changing economy.

Kevin Kelly and Stewart Brand tried to draw Ferriss into a discussion about how important meta-learning would be in an era of  accelerating change, but I didn’t feel that Ferriss contributed much to that conversation.  He seemed more comfortable talking about his own personal experiences or specific techniques for learning than thinking about the big picture.

One thread of the conversation between Brand, Ferriss, and Kelly acknowledged the importance of self-tracking to meta-learning.  Kelly of course founded Quantified Self and Ferriss was proud to reveal that he was at the first meeting.  I was surprised to hear Kelly admit that he wasn’t a self-quantifier himself.  He just wanted to observe the early adopters of what he thinks will become normal behavior.   I personally find it hard to believe that most people will end up carefully collecting, curating, and deriving value from their personal data.  Huge amounts of data will certainly continue to be collected about us, but I am afraid that the majority of us will fail to understand how it is being used to influence our beliefs and actions.

I don’t find Ferriss to be a compelling teacher.  Gretchen and I lost more weight just by tracking meals via livestrong.com than we did trying to follow the 4-hour Body diet.  So my final analysis boils down to this: Tracking good, heuristics bad.