Experiential Learning

Experiential learning gathers the site’s resistance to lecture-first education. Why E-Learning Fails the 70-20-10 Test opens with a coding-class moment that crystallizes the argument: a student gets stuck on a Code Combat level, the instructor starts to explain variable scope, then catches himself — “He didn’t need a lesson. He needed to fight the ogre again, differently, and lose in a new way until the pattern clicked. He needed the 70%.”

The 70-20-10 model — Lombardo and Eichinger’s mid-1990s synthesis of work conducted with Morgan McCall in the 1980s — assigns 70% of meaningful learning to challenging on-the-job experiences, 20% to relationships and feedback, and just 10% to formal training. The essay treats the exact ratio as “precisely wrong and directionally right,” then turns it into a diagnostic. Looking at the e-learning industry through that lens, the indictment is that “an entire ecosystem built almost exclusively around the 10%. Video lectures. Interactive slides. Multiple-choice quizzes. A three-hundred-billion-dollar industry flowing into the thinnest slice of the model while the other ninety percent went undesigned.” The rest of the essay is a working report of three courses — UI design, game-based coding, film studies — that tried to flip that ratio, with three different results.

Seymour Papert’s Legacy supplies the deeper educational imagination behind that critique. The essay opens with a memory of trying to learn BASIC from a CCTV broadcast in 1980s China, and then reads Papert’s verdict on BASIC as a kind of late-arriving explanation: “BASIC is to computation what QWERTY is to typing.” Papert’s point in Mindstorms is that BASIC’s small vocabulary, often praised as beginner-friendly, makes expression so contorted that “only the most motivated and brilliant children would learn to say more than ‘hi’.” The argument the essay extracts is that environments matter more than language minimalism — that learners need powerful ideas and rich problems, not artificially impoverished sandboxes.

Unboxing AI carries the same logic into the present. Once AI tools start doing some of the cognitive work that schools used to assign, the question shifts from “how do we teach the tool?” to “how do we keep students doing enough that they actually learn?” — a 70-20-10 question dressed in 2026 clothes.

Taken together, these pieces make the point that experiential learning is not a slogan but a constraint on design. The best learning environment is not the one that narrates thought most clearly, but the one that gives learners meaningful problems, feedback, and room to act. The opposite — over-explanation, lecture polish, frictionless paths — looks like respect for the learner and is, in practice, the e-learning industry’s most consistent failure mode.