This is an excellent essay which all EdTech people should read. We're all down with more Tech in the Ed, but more interesting is getting the Ed into Tech, i.e., not stupid arithmetic math and puzzles in computer games, but real math reasoning.<p>> <i>Over the long run, humanity will no doubt build more powerful new patterns of explanation into our media platforms, permanently changing and expanding what we mean by explanation. We're only just beginning to explore these possibilities, but it will be exciting to see what happens in the decades ahead as we reinvent explanation.</i><p>this<p>> <i>A second problem with educational games lies in the word "educational". The most important fact about compulsory schooling is that students do not -- indeed, cannot -- choose to attend. Instead, they are required to attend, for what society deems "their own good". This is true even in the most enlightened schools. A student in such a coercive environment does not have full responsibility for their own learning. And, in my opinion, it is not possible to do serious intellectual work without full responsibility for your own learning. Put another way, I believe that compulsory schools, by their nature, are places where serious intellectual work cannot occur.</i><p>True. For many students, schools are prisons.
Given this context, the quality of the teaching material is more important.
If I were in prison, I'd wish for good books...<p>> <i>what would happen if we put the resources and talent of a major video game or movie studio toward creating great explanations, rather than pure entertainment products?</i><p>It's called the Khan Academy ;)<p>______<p>My take on this is that we should aim for lowest common denominator: a system based around HTML (like standalone .html files or .epub) with JavaScript enabled (for MathJax and scripting canvas/svg interactives). Content written in this system will be renderable on most pixel-based devices (for interactive exploratory learning) and with a toLaTeX() method should also be printable (for analytical learning).<p>____<p>Here's my attempt at the kidney studies' Simpson paradox in text.
For large kidney stones Treatment A helped 55 people out of 80,
while Treatment B helped 192 out of 263, in two separate studies.
For small kidney stones treatment A helped 234 people out of 270,
while Treatment B helped 81 out of 87, in the same two studies.<p>We shouldn't really compare the statistics from the two trials since they were performed on different patients. Nevertheless, since both studies had 350 patients in total, we can say that Treatment A is better since it helped a total of 289 patients on that trial which is > than the 273 saved by Treatment B.<p><pre><code> Treatment A Treatment B
Large kidney stones 69% (55 / 80) 73% (192 / 263)
Small kidney stones 87% (234 / 270) 93% (81 / 87)
All patients 83% (289 / 350) 78% (273 / 350)</code></pre>