I think his advice best applies to a group of students who care about learning and want to learn. It doesn't apply to those who don't possess these qualities.
More than a few people should pass because that's what the Board of Regents, the college president, the Governor, the Legislature, and society at large want and they pay my salary. This isn't a flippant statement. In my state one of the Board of Regents wants my pay to be dependent upon how many people pass my class. Every year the president of the college sends a report to the Department of Mathematics and asks what we are doing to improve the passing rate.
The state wants three things,
1. we must accept anyone who applies 2. we should pass almost everyone who takes our classes 3. everyone who passes should 'know' the material
You can have any two of these things but not all three. I have a lot of pressure from the administration and so my perception of what skills it takes to pass has been a decreasing function for the past 10 years.
Almost all of my colleagues give point based on things that a well trained chimp can do. They give points for attendance. They give points for turning in homework. (The homework doesn't have to be correct to get the points.) My colleagues try their hardest to give students enough points so that they can reach the magical 70% thresh hold. Some have lowered this magical thresh hold to 60%.
What it means to pass a class is a level of achievement that most people in the class can reasonably achieve. It has nothing to do with knowledge, skills, or anything that really matters in terms of learning. Will 70% of the students pass with this grading scheme, with these assignments, with this level of work and effort? If the answer is yes then you've got a good idea of what it means to pass a class.
This has bad, long term consequences for society writ large. I'm aware of this and I try the best I can to reach as many students as possible.
I definitely don't teach my students as if I was in the audience. I'm an outlier in terms of mathematical talent and interest. It would be insane for me to teach as if I was in the class.
Can't let it stop with college or a job, now can we?
I love teaching a room full of interested, motivated people. It's too bad I can't become a teacher to do that. :(
There's also this delightful post from btilly http://bentilly.blogspot.com/2009/09/teaching-linear-algebra... that I found wonderfully informative. Make sure you read the q/a in the comments.
The way I learn mathematics is suited only for a small percentage of the population. As an example, I have little use for diagrams. I don't think in terms of geometry but in terms of abstract definitions. I don't need a diagram to conceptualize and understand that pi is the ratio of the circumference of a circle to its diameter. Even with diagrams most of my students don't understand what this means.
Meyer's enthusiasm is contagious and I appreciate that about him. It's exceptional, in fact. It's also not duplicable by most teachers. He's found a style that works for him. It doesn't work for everyone though.
One of his methods involves watching videos of water pouring out of a container and experimentally verifying whether or not his calculations make sense. For me, this is a great big yawn. I could not bear watching such a thing in a class as a student.
The real problem in education, as I see it, is that schools do a terrible job of matching student personality with teacher personality. I believe that solving this problem would significantly increase retention and learning.
It's hard to fit a teaching method to a personality that isn't predisposed to such a method. It just doesn't work. The human interaction dynamic of such a class is a mess.