Thanks for the response (though it seems the end got cut off).

I'm technically not a teacher yet. Our first child is due any day now, and we will probably be educating our children at home. For years now I have been trying to learn more about the history of education both in order to give myself a classical education and learn how to give my future children the best possible education. And thus I'm also interested in figuring out how to do this, from birth on through secondary education.

I have been reading some primary sources (math and science and others). I've read most of Euclid's Elements, read Nichomachus' Arithmetic, and just about to finish up Aristotle's Physics. I also have no problem benefiting from non-primary sources. For example, I'm reading "The History and Practice of Ancient Astronomy" by James Evans. He references and quotes primary sources constantly, and explains what the ancients knew and how they knew it. And it puts it on a practical level, showing how to do the calculations to solve problems, and gives problems to solve. Sounds like maybe an example of what you do in your classes. I'm really enjoying the book. And think it will be a big help when I eventually dive into Ptolemy.

One aspect of using primary sources (or in general using history) that really appeals to me (in addition to the things you point out) is that the order in which discoveries were made in history seems to be about the best order to teach them. Even apart from the story of it, It forms a natural progression, each step logically building on what came before, and one can understand the observations/reasoning that result in each step.

I don't know what's required for AP. My high school didn't have AP classes. I had never even heard of AP until I heard classmates talking about it in college. I don't know what BC is. But I do have a heart for math and science. I have an engineering degree and a masters degree in computer science, and am looking forward to learning more of the background of my knowledge by reading guys like Newton and Maxwell.

So if I'm understanding what you are saying, the thing that's needed to supplement the primary sources is applying the understanding to solving problems. So for Einstein, for an advanced example, one could supplement with something like "Problem Book in Relativity and Gravitation" by Lightman and Price.

What you say regarding Euclid is interesting because, from what I have heard, up until the past century or two books like Euclid were the sole textbook (and the students read it in Greek). I guess the teacher would then be the one to supply the interpretation, guidance, and problems to solve. Like choosing which problems the students should solve, or theorems they should try to prove, before reading Euclid's solution/proof.

I recently listened to your talk on the quadrivium here: http://www.societyforclassicallearning.org/index.php/resources/media/2-2010-conference-recordings

In the talk you urge caution when teaching via original sources (e.g., Euclid for geometry). Can you explain what you mean?

At another point you suggest that original sources lack "academic rigor". Can you explain what you mean by that? Maybe some examples?