With all of the excitement I see online about SICP and Lisp/Scheme, do you (or anyone else) know why they are transitioning away from the book and Scheme?
You may have missed the other excitement online about schools changing their computer science programs to make them more attractive to students. MIT's not really ratcheting down the difficulty, but starting off everyone in Scheme was really off-putting to a number of students.
The new intro classes are in Python, which you could argue has the immediate benefit of being used in the world outside of MIT. They're also now more focused on learning while working towards project-style goals (like building a robot) instead of towards academic goals (like getting Scheme to do OOP).
Personally, I loved 6.001, but I'm the first to admit it was not for everyone, even people who could become great coders.
I took 6.001 in Spring '87. I loved it, aside from the slow computers in the lab. I pulled an all-nighter on a Saturday for a problem set not due until Thursday so as to avoid the crowds.
After the class was over I didn't use Scheme for another 10 years, but did everything in C. Scheme was just too slow in 1987. Both the language implementations and the hardware have come a long way. I use Scheme all the time now, and there are numerous popular languages, notably PHP and Ruby, that are slower than Scheme today.
It surprises me that people didn't know about these lectures already (of course I realize there's always a first time). They are considered classic and anyone interested in lisp (and programming in general) should watch them.
Imagine being unprivileged enough to be born in a third-world country and not even have a chance to pay your application fee for higher education. If you're bright enough and have Internet access you can still learn from the best and maybe prepare yourself and earn a scholarship in the future.
Well, it's very common at my university (Faculty of Information Technology, Brno university of technology: http://www.fit.vutbr.cz). We're in the 21st century after all. Lectures are streamed in real-time and also recorded and published later.
At Georgia Tech, they filmed a lot of lectures and you can go back to the library and watch the tapes there. I found it to be one of the best ways to study for an exam. It's funny how the second time you watch a lecture, it all seems so easy to understand.
I'm watching the Gilbert Strang linear algebra course to pick up enough for other classes I am taking. Saves me using up an entire semester for that (I have a tuition benefit which I am using to take one class for semester; so I can take Information Retrieval now instead of taking Linear Algebra and waiting until next Spring.).
I was surprised to see the equivalent of a powerpoint in '86. It's weird to see the same ideas that we have now, but implemented in an older technology. Kinda like when you watch the mother of all demos.
I can see why old-timers think nothing's changed in the past couple decades sometimes.
All the Intro to CS classes start the same: the teacher writes COMPUTER SCIENCE in the board and then says 'this is not a science and it is not about computers' and crosses them out.
I took 6.001 in the spring of 1997 and we did the simple "Decaf Java" interpreter (in Scheme). Decaf Java is an s-exp representation of a Java program's parse tree. It is foggy and distant now but I believe we did do a problem set in actual Java and then later in the semester after doing the Scheme in Scheme metacircular evaluator we did the Decaf Java interpreter. I don't have any other 6.001 experience to compare it to but I loved the version I took.
http://swiss.csail.mit.edu/classes/6.001/abelson-sussman-lec...
They're pretty awesome.