Berkeley to teach SICP in Python

http://www.cs.berkeley.edu/~bh/61a.html

thats an improvement, at least he's not using Logo

he ruined his book Simply Scheme by converting Scheme syntax to Logo

I'm glad they gave up on that piece of shit language, but the replacement isn't any better.  Python's closures aren't usable and it has lots and lots of warts :/.

old old news

I wonder if it was a coincidence that the /prog/snake became popular about the same time MIT started its switch to Python.

I wonder if it's a coincidence that all the top institutions are replacing Scheme with Python.

It's useless, it's over.

I'm a competent C programmer.  Is there anything I should read SICP for?

>>8
expand your understanding of abstraction

>>9
What are some "must-read" sections then?  I'm somewhat limited in the time I can spend to read this, along with other things I'd like to do.  I hate relearning basic topics, which is what a lot of books tend to consist of until the obviously interesting stuff starts.

>>10
I hate relearning basic topics, which is what a lot of books tend to consist of until the obviously interesting stuff starts
Same here, man.

>>10

The interesting part is the depth of the treatment of each topic, and the general attitude towards problem solving and mastery of the field.

But as a C programmer, you should start from chapter one, as it will introduce you to functional programming, from which it will veer off near the end of chapter two (which deals with data abstraction, so it's more worthy of skimming through it),  to start chapter three with the introduction of mutable state, its representation, consequences and organization (they introduce constraints-based programming and laziness there too).

I haven't yet reached chapter four, but they turn to languages in it, and in chapter five to machines and compilation.

>>8
Don't listen to >>9. Just start writing code. Use the hyperspec.

>>12
it just seems pointless to learn a dead language.  i know it's for learning purposes and to help understand the "elegance" of lisp, but pragmatically, why wouldn't you use your language of choice to learn the same things or conquer the same problems?

>>14

What, you think Scheme will clutter your head? It's one of the smallest non-joke languages there are, made for pedagogy and renowned for its minimalism. Besides, most languages will be more awkward to use while exploring the techniques in the book due to their more limited support for first-class functions, or their botched scoping rules. And let's not say anything about tail call optimization (because it's not really that relevant).

And not that it actually matters when talking about SICP, but Scheme is not really dead, at least in a way. Racket has a thriving community, and if you take further distance from Scheme's lineage, Clojure is full of hype still and runs on the JVM.

fuck Python

Coomon Lusp.

Racket has a thriving community
4/10

>>18
Yes, out of the ten people inside the Racket community, four of them are thriving.

Whats the difference between Racket and Scheme ?

>>15
Many Scheme implementations are also still actively developed. And those fucking toy implementations keep poping out of nowhere.

>>20
Racket is a dialect of Scheme, just like Scheme is a dialect of Lisp.
It's not ``a Scheme'' because it doesn't adhere to any standard (by default).

>>18
I've found the community helpful, friendly, and intelligent.

>>22
helpful, friendly, and intelligent
ahahahhaHAHAHAHAHAHA oh man, you just crack me up

>>23
are you okay

>>22,23
Remember to sage your posts for the benefit of all /prog/ users.

>>25
autist

replace 61A with something more modern
Since when Python became "modern"? This crap cant even do macros.

>>27
In this case, "modern" is considered ENTERPRISE