Actually, you've hit upon the root problem. Truly effective education doesn't scale. Why? Because truly effective education is personal and personalized. A teacher is very effective when teaching small classes, because they can give students individual attention. That teacher is not going to be as effective when teaching 100, or 1000, or a million students. There's no way to provide individual attention there.
It can be hard for us to understand, because readers of HN are likely to be autodidacts to some extent who don't need that kind of attention. We need to understand that this isn't the norm. The research has been very consistent on this topic for decades now.
That doesn't mean that mass education a la khan and some of the other solutions that are out there are bad. The frustration educators are feeling about these rise from several different areas:
1) We know that these are not the most effective means of teaching. Most students do experience long-term learning in these environments.
2) Mass education is very dehumanizing. This isn't just websites that experience this, but teachers in actual classrooms who have to teach larger and larger class sizes every year. It's a dehumanizing experience.
Basically, to me it seems like we don't care about 2 any more, and the sacrifice of effectiveness is one we're willing to make in order to get quantity. I don't like that, and many teachers out there completely reject that idea.
So if you can find a way to scale personal (in the social sense), and personalized education, you're going to be positioned very well. Khan academy isn't it, and I'm not even sold that doing it on large scales is even possible.
Maybe scaling personalized education is not that complicated.Basically you just create wide arrays of explanations to something , and automatically try to offer each student the explanations he will benefit most from. Yes it demands work on content creation , and it demand a way(automated, crowdsourced, by teacher) to offer each student recommendations on content. But the gains could be immense: offering great education globally.
After you solved this problem, offering social support becomes somewhat easier.Instead of the teacher focusing on lecturing to full classes, creating lesson plans, he can focus on working with each student individually and offering social support. That's the whole idea of reverse classroom:children listen to lectures at home, and do homework at school with help of the teacher.
But that's just one way of a scalable solution, but there are many startups working on finding the solution and building interesting things. I wouldn't bet against all of them.
>> Most students do experience long-term learning in these environments.
That's interesting, because that doesn't match the many positive reviews from students. That also doesn't match the fact that khan offers exercises to some students and see results improve.
(And I would argue huge college lecture halls, and courses such as the Standford AI class also illustrate that small class size isn't what it's made out to be.)
-frequent teacher feedback
-the use of data to guide instruction
-increased instructional time
How many of those are not even doable by Khan?
By the way, when I say 'class size matters' I'm not saying that 'large class sizes aren't effective.' I'm saying 'smaller class sizes are more effective than large class sizes.' We all know learning can happen in pretty much any environment. The question is how to maximize that value. The fact that you can learn in the stanford AI class is very far away from answering the question 'is that the most effective way to learn.'