Hacker News new | comments | show | ask | jobs | submit login
CS61A Structure and Interpretation of Computer Programs (berkeley.edu)
83 points by miobrien on June 24, 2017 | hide | past | web | favorite | 17 comments



In case anyone's curious, this is a rather old iteration of the course (2011, right after it switched from scheme to Python).

The current iteration is at http://cs61a.org - taught by two undergrads for the summer.

The most recent two iterations taught by professors are http://fa16.cs61a.org and http://sp17.cs61a.org


I can't even conceive of CS61a not being scheme/lisp.


Yeah, I liked Scheme but that was the last I ever saw of it. They didn't even use it in CS 188. Now they beat Python into your head like a catechism. And while I really liked Scheme, I didn't like SICP which like the bible we actually never read and only worshipped.


> And while I really liked Scheme, I didn't like SICP which like the bible we actually never read and only worshipped.

Uh when did you take it? When I took the self-paced version with Harvey we read SICP. I mean, I'm sure some students skipped readings but they were still assigned nevertheless.


Harvey worshipped SICP so that makes sense.

I took 61A with Hilfinger (who usually taught 61B) and I didn't crack the book. His lectures and notes were very thorough. I can't remember if there were readings on the syllabus that I skipped but I didn't crack the book. Moreover, I was always big on doing the reading before class. Class was a review.

Later when I did read SICP, I didn't like it. Maybe that was like reading the book after seeing the movie. Dunno, but I didn't like it.

As an aside, there was a whole subculture of students who read nothing but just did the HW and worked past tests and sets. They did amazingly well. I had a study buddy who did that in EE20. I prepped him for the final and he told me he'd beat me; he narrowly did. He graduated with a 3.9+ GPA in EECS which is insane. He hated ideas then and now. But he tested insanely well. It was just pattern matching.


> Later when I did read SICP, I didn't like it. Maybe that was like reading the book after seeing the movie. Dunno, but I didn't like it.

Which introductory CS books did you enjoy? I still love my copy of Patterson and Hennessy but it is dry compared to SICP. To someone who had only ever programmed in C++ before 61A, SICP felt like an adventure.

> As an aside, there was a whole subculture of students who read nothing but just did the HW and worked past tests and sets. They did amazingly well.

Yep. Sadly this has only become more common with the rapid growth of the major. Most of my friends were brilliant but they didn't love EECS. If they had been 10 years older they probably would have majored in math or physics but they felt pressured to study something that would get them a job. They were able to do well just by studying old tests.

We would only get old tests from HKN but many students were somehow able to illegally obtain them. Cheating has only become more rampant and more sophisticated -- the department is going to have to address it at some point instead of sticking their head in the sand.


I strongly prefer Patterson and Hennessy Computer Organization and Design to their Computer Architecture which I really don't like; if I do look something up in CA, it's in an old edition. The Dragon Book 2nd ed is awesome. We didn't have a book for 61B. I didn't like the OS book whatever it was. Norvig + Russell is pretty darned good. The notes for CS 70 are awesome as are the CS 161 notes.

If Kubi is teaching just take the course.


I took it with Harvey.

one thing that struck me about SICP, is in the first section of the first paragraph they use an example of square roots by newtons approximation method. WTF!?!?

With all today's talk about being more welcoming to talented people with different backgrounds, talk about an off putting first example!!


Readings still get assigned, even in the python version - although my impression is that a pretty small portion of the class reads them. :)

They're at composingprograms.com - written by Denero to use Python, but based very heavily off of SICP.


Indeed; why don't they take the barbaric regression to its full conclusion in one fell swoop and go straight to line-numbered BASIC.


I did this course a while back just when I was getting started on my self-taught journey. I'd highly recommend this over Harvard's CS50 or MIT's intro course. This has great coverage of how a computer program works, without being too pedantic.


I feel that the biggest strength of CS50 is its wide scope of technologies used. Provides a real great introduction to the field of computer science


This Course does not come close to cs50 in anyway. cs50 makes computer programming look intresting to a lawyer


Yes, CS50 is quite lively but in terms of quality, 61A beats CS50 hands down.


I fear the current 'iterations' might not last long because of the video aspect. I have been following two other courses from Berkeley: data8.org and datastructur.es. However, because of a certain directive (and I don't fault university for it) the video lectures are no longer public. Sad.


Seems that CS61A video lectures have subtitles so it shouldn't be a problem: http://inst.eecs.berkeley.edu/~cs61a/fa16/


Really nice course. Labs come with solutions and would make an excellent "challenge pack" for anyone learning Python and CS right now.




Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact

Search: