The sixth checkbox states:

In normal-order evaluation, (p) is not evaluated because predicate of if evaluates to false

It should state:

In normal-order evaluation, (p) is not evaluated because predicate of if evaluates to true

1-1-elements.html always crashes in Mobile Safari and Chrome for iOS. This doesn't happen with 1-2-procedures.html.

The codemirror interface is very slow when run in mobile browsers too.

The quiz is:

Exercise 1.12. The following pattern of numbers is called Pascal's triangle.

The numbers at the edge of the triangle are all 1, and each number inside the triangle is the sum of the two numbers above it.
      1
     1 1
    1 2 1
   1 3 3 1
  1 4 6 4 1
       . . .

Write a procedure that computes elements of Pascal's triangle by means of a recursive process. If a number lies outside of the triangle, return 0 (this makes sense if we view pascal as the combination > > function ). Start counting rows and columns from 0.

I did the Exercise, it gave me an unexpected evaluation,so I try to figure it out why.
End up copying the answer hided in the html, the site evaluates it to be right,
I check in DrRacket with the folowing code.
The answer hided in the site is wrong, #the followings.

; site's answer
(define (pascal row col)
   (cond ( (or (< row col)) 0)
         ((or (= 0 col) (= row col)) 1)
         (else (+ (pascal (- row 1) col)
                  (pascal (- row 1) (- col 1))))))
; the right one
(define (ps row col)
  (cond ((or (< row col)
             (< row 1)
             (< col 1)) 0)
        ((or (= col 1)
             (= col row)) 1)
        (else (+ (ps (- row 1) (- col 1))
                 (ps (- row 1) col)))
  )
)
(define (printPs-iter func row col limit)
  (if (and (= row 1) (= col 1)) (display func))
  (if (< row limit) (display (func row col)))
  (if (= row col) (newline))
  (cond ((> row limit) "end")
        ((> row col) (printPs-iter func row (+ col 1) limit))
        ((= row col) (printPs-iter func (+ row 1) 1 limit)))
  )
(define (printPs func limit)
  (printPs-iter func 1 1 limit)
)
(display "----------------") (newline)
(printPs pascal 5)
(display "----------------") (newline)

(display "----------------") (newline)
(printPs ps 5)
(display "----------------") (newline)

terminal print out:

----------------
#<procedure:pascal>1
21
331
4641

"end"
----------------
----------------
#<procedure:ps>1
11
121
1331

"end"
----------------
> 

A tab is the green thing on the right

The contents page should show all sections as well as exercises

For eg, if the desired answer is '((a b) a b) then we should accept ((a b) a b), (('a 'b) 'a 'b), '(a b) a b ) etc

These sections have footnotes in the original text. The numbers are present in the body, but not linked or superscripted.

Should show dependencies and pushes

Steal some features from light table

http://xuanji.appspot.com/isicp/1-2-procedures.html#Exponentiation

If you take a look, you can spot things like "14 multiplications.38 It is also" or " recursive algorithm.39".
These numbers correspond to the references at the end of each chapter that you can see in the original book. We should either remove them or link them to the references.
I can do it myself, but what should we do? My vote is to completely remove them.

Error: Not Found

The requested URL /isicp/index.html/ was not found on this server.

Tried the ^ <> at the top of the page, but ^ doesn't work. I think it should point to "/isicp/" instead.

evaluating
(integral cube 0 1 0.001)
on 1-3-hop.html on chrome gets
RangeError: Maximum call stack size exceeded, while
(integral cube 0 1 0.01)
works

SICP Lecture videos are now available on MIT's YouTube channel: https://www.youtube.com/watch?v=2Op3QLzMgSY

These videos should be displayed alongside the chapter text.

It's often unclear whether the scheme interpreter is running or not, and it may go into an infinite loop, freezing the browser for no apparent reason (to the user).

This can be fixed by having a "Code running"-indicator and cutting runtime short when it takes more than, say, 10 seconds.

They are now situated at http://mitpress.mit.edu/sites/default/files/sicp/full-text/...

I would like to help. How do you want to resolve this? Should I just fix the links or should I download all the images to the images folder to avoid future breakage?

Tell me which solution you prefer and I will make a pull request.

Not all of them, but it's pretty spotty. In some places it's been replaced by $ $ for inline equation rendering, but it doesn't seem consistent. I've been looking at the raw html of the original. There are also a lot of name tags for the index. If we're not doing an index, we probably don't need them, but they're probably harmless if we add some with copy/paste. It might have been easier to start from the html version, but we'll lose a lot of work if we try that now.

Is it possible to convert the whole site from English to Japanese or even Japanese-English bilingual versions? https://sicp.iijlab.net/fulltext/xcont.html

My code works but is not correctly evaluated. This occurs in other exercises as well.

(define (cons a b)
  (* (expt 2 a)
     (expt 3 b)))

(define (val n i)
  (if (zero? (modulo n i))
      (+ 1 (val (/ n i) i))
      0
  )
)

(define (car p) (val p 2))

(define (cdr p) (val p 3))

Grader should be allowed to run arbitrary code in order to grade a user-submitted function, instead of simply being able to compare strings.

Here: https://xuanji.appspot.com/isicp/1-1-elements.html

If you look at the paragraph:

"Notice that and and or are special forms, not procedures, because the subexpressions are not necessarily all evaluated. Not is an ordinary procedure."

the "and" and "or" should have some kind of distinct formatting but they are exactly the same as the surrounding text.

As seen in 1.3.1:

The subsections don't even include numbers in this version, so you can't find the reference they're talking about by a text search either. The original book has numbers on the headings and there are many references to them throughout the text, not just the footnotes. These other internal references should also be hypertext to the appropriate chapter and heading.

https://www.scheme.com/tspl3/intro.html#./intro:s40

The boolean values representing true and false are written as #t and #f. Scheme conditional expressions actually treat #f as false and all other objects as true, so 3, (), "false", and nil all count as true.

This is from 1-1-elements.html rendered in Chrome.
Edit: I just noticed a "load unsafe scripts" icon in Chrome. That seems to fix the rendering problem, and it enables the green sidebar which wasn't showing properly before. Is there a way to make the scripts "safe"? Or at least make a note of it so people know they have to enable the scripts manually?

Hey, I'm really digging this project. I noticed I have to run Ctrl-Enter to evaluate code; however, a common convention is to allow Mac users to use Cmd-Enter as well.

I noticed your'e using CodeMirror, and from a previous previous commit that solves #11, you could do something like this:

editor.setOption('extraKeys', {'Cmd-Enter': function() {
  editor.getOption("onBlur")();
}});

Thoughts?

For eg (define size 2)