3 ms·

>Anyway, it is possible to do this again and with two small kids I'm very interested to know how. Here's a context-relevant place to start: http://slatestarcod

• mncharity 9y ago
  I wonder if the review is missing some aspects of "concentrate on one subject"? On one subject, a mentor can be a master, and master instructor. And such instruction is crucial for developing deep expert understanding. But in the future, with better content, and improved learning infrastructure, one might imagine these becoming available for more than one subject. Both the concentration on one subject, and that subject being chess, contribute to the effort having nice properties. Like non-superficiality. An integrated and deeply organized body of knowledge and skills. Developing transferable knowledge (within the subject at least), rich feedback, and reflective building of mastery. But it's the properties that matter. Chess, or another one subject, might be taught in a way that fails to have the nice properties. And as learning infrastructure improves, the nice properties may become available without the "one subject" or "and it's chess" restrictions. There's a long history of exceptional masters teaching their field/craft to their children from a young age, who in turn become exceptional masters. The challenge is to scale that. The straightforward "a gaggle of masters in everyone's pocket" is AI-complete implausible. But the art of the next decade is crafting human-computation hybrid systems. Things like eye tracking, and big data, provide opportunities that no master mentor has ever had. We just need to get around to building on them.