2009 Apr 13 00:01:10  <chrisconley> hey all
2009 Apr 13 00:01:25  <inimino> hey
2009 Apr 13 00:01:25  <chrisconley> anybody else awake?
2009 Apr 13 00:02:12  <mariorz> heey
2009 Apr 13 00:02:25  -*- mariorz is still starting 2.20
2009 Apr 13 00:02:31  <aamar> hey
2009 Apr 13 00:02:46  <aamar> got through 2.52
2009 Apr 13 00:03:22  <chrisconley> i got to 2.42
2009 Apr 13 00:03:24  <chrisconley> still working on it
2009 Apr 13 00:03:30  <inimino> same as me
2009 Apr 13 00:03:45  <inimino> well, I haven't started working on it
2009 Apr 13 00:04:16  -*- inimino had a snack and a little nap instead
2009 Apr 13 00:06:30  <chrisconley> do you guys want to talk section 2.2 in general tonight?
2009 Apr 13 00:06:40  <chrisconley> then get all caught up for next meeting?
2009 Apr 13 00:06:54  <inimino> that sounds good
2009 Apr 13 00:07:04  <aamar> ok
2009 Apr 13 00:07:55  <inimino> 2.2 is a long section
2009 Apr 13 00:08:23  <chrisconley> aamar: how is 2.43->2.52?
2009 Apr 13 00:09:33  <inimino> I guess "hierarchical data" covers a lot
2009 Apr 13 00:09:47  <aamar> last few are fairly long
2009 Apr 13 00:09:53  <chrisconley> heh, yeah
2009 Apr 13 00:10:42  <aamar> 2.49(d) is tedious and long -- I suggest skipping it, actually.
2009 Apr 13 00:11:09  <chrisconley> hmmm, thanks for the tip
2009 Apr 13 00:11:30  <chrisconley> mind if i ask about 2.42?
2009 Apr 13 00:12:09  <chrisconley> i'm having trouble with safe?
2009 Apr 13 00:12:22  <aamar> ok
2009 Apr 13 00:13:09  <chrisconley> i might be heading down the wrong path, but I am passing "positions" as ((31) (21) (12)) for example
2009 Apr 13 00:13:15  <aamar> same here
2009 Apr 13 00:13:30  <chrisconley> (31) is my most recent position
2009 Apr 13 00:13:59  <chrisconley> i want to compare the row(3) to all of the other rows
2009 Apr 13 00:14:12  <aamar> (assume there's a space in there, i.e. (3 1))
2009 Apr 13 00:14:29  <chrisconley> something like (include? 3 (cdr postions))
2009 Apr 13 00:14:46  <chrisconley> actually, those parens were just a pause, so i actually meant 3
2009 Apr 13 00:15:27  <chrisconley> let me try to be a little clearer?
2009 Apr 13 00:15:47  <aamar> ok
2009 Apr 13 00:15:54  <chrisconley> so in the positions example above, i have queens in rows 3 2 and 1
2009 Apr 13 00:16:10  <chrisconley> those queens are in columns 1 1 2
2009 Apr 13 00:16:47  <inimino> columns 1 2 and 3, you mean?
2009 Apr 13 00:17:00  <chrisconley> yeah, bad example above
2009 Apr 13 00:17:28  <aamar> seems okay so far
2009 Apr 13 00:17:51  <aamar> so you want to check the first queen to see if it's safe against all the previous queens
2009 Apr 13 00:17:56  <chrisconley> yes
2009 Apr 13 00:18:18  <inimino> so you have to check if any row is the same
2009 Apr 13 00:18:23  <chrisconley> right
2009 Apr 13 00:18:25  <inimino> and the diagonals
2009 Apr 13 00:19:09  <chrisconley> the way i have it, i just have to check rows, the columns are already unique
2009 Apr 13 00:19:19  <chrisconley> ahhhhh
2009 Apr 13 00:19:28  <chrisconley> crap, i forgot about diagonals-heh
2009 Apr 13 00:19:35  <inimino> hehe
2009 Apr 13 00:19:53  <chrisconley> ok, forgetting that for a minut
2009 Apr 13 00:20:09  <chrisconley> is there a way to do something like (include? 3 (list 1 2 3))
2009 Apr 13 00:20:35  <chrisconley> returns true if 3 exists in the given list
2009 Apr 13 00:21:04  <inimino> you could write that as a helper function
2009 Apr 13 00:21:07  <aamar> yes, you could write an include? method
2009 Apr 13 00:22:28  <inimino> case (nil items) -> false, (car items == x) -> true, otherwise (include? x (cdr items))
2009 Apr 13 00:23:14  <aamar> (def (include? x others) (cond ((null? others) false) (= x (car others)) true) (else (include? x (cdr others)))))
2009 Apr 13 00:23:22  <aamar> right, that sort of thing...
2009 Apr 13 00:23:57  <aamar> It's hard to use an include like that for diagonals, I think.
2009 Apr 13 00:23:58  <chrisconley> ahhhh, thanks! i had a major brain block and wasn't thinking recursively
2009 Apr 13 00:24:07  <chrisconley> have i not learned anything!? :)
2009 Apr 13 00:24:16  <aamar> how soon we forget!
2009 Apr 13 00:24:55  <inimino> for diagonals you need to check whether any (col+n, row±n) is occupied
2009 Apr 13 00:25:13  <chrisconley> yeah cool
2009 Apr 13 00:25:57  <chrisconley> ok i should be good to go, thanks!
2009 Apr 13 00:26:30  <aamar> jump back to 2.33?
2009 Apr 13 00:26:39  <inimino> sure
2009 Apr 13 00:27:44  <inimino> oh yeah, I did this one wrong the first time and had to go back
2009 Apr 13 00:28:11  <aamar> http://mibbit.com/pb/EOw9nl
2009 Apr 13 00:28:25  <inimino> I was thinking of accumulate as a fold-left
2009 Apr 13 00:28:32  <inimino> http://inimino.org/~inimino/projects/2009/SICP/chap_2/2.33
2009 Apr 13 00:29:55  <inimino> hm, my length lambda only takes one argument
2009 Apr 13 00:30:02  <chrisconley> http://github.com/chrisconley/sicp-exercises/blob/66ffd41c76d8840c5987d60d7e778a616fbd5f74/chapter2/ex2-33.ss
2009 Apr 13 00:30:07  <inimino> yours takes two
2009 Apr 13 00:30:50  <inimino> oh yeah, it's going to be called with the element
2009 Apr 13 00:31:18  <chrisconley> i have the same length as aamar it seems
2009 Apr 13 00:32:14  <inimino> yeah mine was incorrect
2009 Apr 13 00:32:46  <inimino> your map is different
2009 Apr 13 00:33:13  <aamar> chrisconley: your map uses (p x y) while we have (cons (p x) y)
2009 Apr 13 00:33:21  <chrisconley> yeah
2009 Apr 13 00:33:54  <inimino> your map is more of a fold
2009 Apr 13 00:34:02  <chrisconley> yeah whoops
2009 Apr 13 00:35:41  <aamar> okay, otherwise everything matches
2009 Apr 13 00:35:45  <inimino> yeah
2009 Apr 13 00:35:50  <aamar> should we go on to 2.34?
2009 Apr 13 00:35:53  <chrisconley> yeah
2009 Apr 13 00:35:57  <inimino> http://inimino.org/~inimino/projects/2009/SICP/chap_2/2.34
2009 Apr 13 00:36:22  <chrisconley> http://github.com/chrisconley/sicp-exercises/blob/66ffd41c76d8840c5987d60d7e778a616fbd5f74/chapter2/ex2-34.ss
2009 Apr 13 00:36:22  <aamar> http://mibbit.com/pb/eBcA3f -- 2.34
2009 Apr 13 00:37:15  <inimino> all the same
2009 Apr 13 00:37:34  <inimino> wait
2009 Apr 13 00:37:38  <inimino> mine is backwards
2009 Apr 13 00:37:44  <inimino> whoops :-)
2009 Apr 13 00:38:27  <aamar> right, results in an extra x factor.
2009 Apr 13 00:38:35  <aamar> Okay, otherwise the same...
2009 Apr 13 00:38:38  <inimino> yeah
2009 Apr 13 00:39:48  <aamar> 2.35
2009 Apr 13 00:39:53  <inimino> http://inimino.org/~inimino/projects/2009/SICP/chap_2/2.35
2009 Apr 13 00:39:56  <aamar> http://mibbit.com/pb/4Y194N -- 2.35
2009 Apr 13 00:40:50  <chrisconley> http://github.com/chrisconley/sicp-exercises/blob/66ffd41c76d8840c5987d60d7e778a616fbd5f74/chapter2/ex2-35.ss
2009 Apr 13 00:40:59  <aamar> nice, keep forgetting that "+" is a useful synonym for (lambda (x y) (+ x y))
2009 Apr 13 00:41:13  <chrisconley> i cheated a bit on this one
2009 Apr 13 00:41:50  <inimino> aamar yours and mine look just the same
2009 Apr 13 00:42:24  <chrisconley> i replaced (map (??) (??)) with (enumerate-tree t)
2009 Apr 13 00:42:35  <inimino> ah
2009 Apr 13 00:43:07  <inimino> alright, 2.36?
2009 Apr 13 00:43:37  <chrisconley> http://github.com/chrisconley/sicp-exercises/blob/66ffd41c76d8840c5987d60d7e778a616fbd5f74/chapter2/ex2-36.ss
2009 Apr 13 00:43:39  <aamar> ok
2009 Apr 13 00:43:59  <aamar> http://mibbit.com/pb/WlfcIK -- 2.36
2009 Apr 13 00:44:53  <inimino> all the same
2009 Apr 13 00:45:27  <chrisconley> 2.37?
2009 Apr 13 00:45:34  <inimino> http://inimino.org/~inimino/projects/2009/SICP/chap_2/2.37
2009 Apr 13 00:45:35  <aamar> ok
2009 Apr 13 00:45:49  <chrisconley> http://github.com/chrisconley/sicp-exercises/blob/66ffd41c76d8840c5987d60d7e778a616fbd5f74/chapter2/ex2-37.ss
2009 Apr 13 00:46:10  <aamar> http://mibbit.com/pb/zItO4C
2009 Apr 13 00:48:01  <chrisconley> heh, my transpose has an unnecessary lambda
2009 Apr 13 00:49:15  <aamar> chrisconley -- in your dot product (define (dot-product v w)   (accumulate + 0 (map * v w))) -- I'm not sure of the use of map there...
2009 Apr 13 00:49:16  <chrisconley> and looks like i should've replaced my second map in matrix*matrix with matrix*vector like inimino
2009 Apr 13 00:49:31  <aamar> the map I have takes only one sequence
2009 Apr 13 00:50:29  <inimino> oh I didn't look at that dot-product
2009 Apr 13 00:50:43  <inimino> since it was given in the problem
2009 Apr 13 00:50:56  <chrisconley> dot product is V0W0 + V1W1 +...+ ViWi
2009 Apr 13 00:51:08  <inimino> but that's how it was given
2009 Apr 13 00:51:25  <aamar> you're right -- I see that now.
2009 Apr 13 00:51:28  <aamar> Hm.
2009 Apr 13 00:51:40  <chrisconley> yeah i just went with the book's definition; my matrix operation knowledge is a bit rusty
2009 Apr 13 00:51:57  <inimino> I think there was a footnote about it
2009 Apr 13 00:52:24  <inimino> http://mitpress.mit.edu/sicp/full-text/book/book-Z-H-15.html#footnote_Temp_190
2009 Apr 13 00:52:25  <chrisconley> footnote 17 talks about the extended version of map
2009 Apr 13 00:52:51  <inimino> http://mitpress.mit.edu/sicp/full-text/book/book-Z-H-15.html#footnote_Temp_166         (actually this)
2009 Apr 13 00:53:13  <aamar> thanks, completely missed those notes.
2009 Apr 13 00:53:49  <inimino> so map is basically map-n (by analogy with accumulate-n)
2009 Apr 13 00:55:08  <inimino> all of the rest looks similar
2009 Apr 13 00:55:44  <inimino> 2.38?
2009 Apr 13 00:55:56  <aamar> yep, sure
2009 Apr 13 00:56:02  <chrisconley> http://github.com/chrisconley/sicp-exercises/blob/66ffd41c76d8840c5987d60d7e778a616fbd5f74/chapter2/ex2-38.ss
2009 Apr 13 00:56:20  <inimino> http://inimino.org/~inimino/projects/2009/SICP/chap_2/2.38
2009 Apr 13 00:56:20  <aamar> http://mibbit.com/pb/H9CVrp -- 2.38
2009 Apr 13 00:57:00  <inimino> when I got to this problem I had to go back and redo all the ones involving accumulate ;-)
2009 Apr 13 00:57:44  <chrisconley> hehe, that must've been fun...
2009 Apr 13 00:58:16  <inimino> yeah
2009 Apr 13 00:58:43  <inimino> hm, you have commutativity where I have associativity
2009 Apr 13 00:58:56  <chrisconley> yeah i think you're right; i think i got them mixed up
2009 Apr 13 00:59:02  <inimino> and aamar didn't answer that part ;)
2009 Apr 13 00:59:23  <aamar> forgot to paste it in
2009 Apr 13 00:59:28  <aamar> ; associative property -- (op (op a b) c) = (op a (op b c)) -- guarantees equality
2009 Apr 13 00:59:41  <inimino> it's also necessary for the initial value to have some property
2009 Apr 13 01:00:32  <aamar> do you in fact need both associativity and commutativity?
2009 Apr 13 01:00:38  <inimino> I guess to be the identity under that op, since it will actually come at different ends of the list
2009 Apr 13 01:00:47  <aamar> trying to think of something that has one but not the other...
2009 Apr 13 01:00:59  <inimino> I think you only need associativity
2009 Apr 13 01:01:03  <inimino> http://en.wikipedia.org/wiki/Moyal_product is one
2009 Apr 13 01:01:30  <inimino> but you also need for the zero op x = x op zero
2009 Apr 13 01:01:54  <inimino> where zero is whatever initial value you provide
2009 Apr 13 01:02:09  <inimino> (I think)
2009 Apr 13 01:02:32  <aamar> prob. right
2009 Apr 13 01:03:44  <aamar> Hm. I think it would take a while for me to use the Moyal product to test this claim.
2009 Apr 13 01:03:59  <inimino> heh, yeah, me too
2009 Apr 13 01:04:14  <inimino> probably easier to construct such an operation directly by using a table
2009 Apr 13 01:07:22  <aamar> Well, it's easy to make one that's commutative but not associative: (abs (- a b))
2009 Apr 13 01:08:20  <inimino> true
2009 Apr 13 01:08:58  <aamar> (fold-left (lambda (a b) (abs (- a b))) 0 (list 10 4 1)) ; => 5
2009 Apr 13 01:09:07  <aamar> (fold-right (lambda (a b) (abs (- a b))) 0 (list 10 4 1)) ; =? 7
2009 Apr 13 01:09:16  <aamar> ; => 7 I mean
2009 Apr 13 01:09:21  <aamar> okay, that's good enough for me
2009 Apr 13 01:10:19  <aamar> 2.39 ?
2009 Apr 13 01:10:35  <aamar>   Complete the following definitions of reverse (exercise 2.18) in terms of fold-right and fold-left from exercise 2.38:  (define (reverse sequence)   (fold-right (lambda (x y) <??>) nil sequence)) (define (reverse sequence)   (fold-left (lambda (x y) <??>) nil sequence))
2009 Apr 13 01:10:35  <inimino> ok
2009 Apr 13 01:10:50  <aamar> http://mibbit.com/pb/TsWXLx -- 2.39
2009 Apr 13 01:10:50  <inimino> http://inimino.org/~inimino/projects/2009/SICP/chap_2/2.39
2009 Apr 13 01:11:37  <aamar> same.
2009 Apr 13 01:11:51  <chrisconley> http://github.com/chrisconley/sicp-exercises/blob/66ffd41c76d8840c5987d60d7e778a616fbd5f74/chapter2/ex2-39.ss
2009 Apr 13 01:12:07  <chrisconley> i didn't get the reverse working with fold-right
2009 Apr 13 01:12:42  <chrisconley> seems simple now :)
2009 Apr 13 01:13:29  <aamar> actually took a while to get that one
2009 Apr 13 01:13:44  <aamar> And some trial and error too.
2009 Apr 13 01:13:44  <inimino> yeah it was tricky
2009 Apr 13 01:14:08  <inimino> I never test anything ;-)
2009 Apr 13 01:14:29  <inimino> I think something like (append x (list y)) has appeared before
2009 Apr 13 01:14:52  <inimino> but actually,,,, that was probably when I was using the backwards version of accumulate, heh
2009 Apr 13 01:16:12  <inimino> 2.40?
2009 Apr 13 01:16:16  <aamar> ok
2009 Apr 13 01:16:27  <inimino> http://inimino.org/~inimino/projects/2009/SICP/chap_2/2.40
2009 Apr 13 01:16:29  <chrisconley> http://github.com/chrisconley/sicp-exercises/blob/66ffd41c76d8840c5987d60d7e778a616fbd5f74/chapter2/ex2-40.ss
2009 Apr 13 01:16:37  <aamar> http://mibbit.com/pb/k3sGym -- 2.40
2009 Apr 13 01:17:27  <aamar> I was actually talking to someone who'd done most of SICP when she was in college, she actually did all the assigned exercises by hand on paper.
2009 Apr 13 01:17:46  <chrisconley> all look the same
2009 Apr 13 01:17:49  <aamar> It was a while back I guess, when it wasn't likely you'd have a scheme interpreter runnable on Windows 3.1 or whatever.
2009 Apr 13 01:18:12  <chrisconley> geez; i did everything on paper for most of chapter 1
2009 Apr 13 01:18:36  <inimino> they all look the same
2009 Apr 13 01:18:49  <inimino> yeah, I've only been running the ones that require you to
2009 Apr 13 01:19:17  <inimino> like the ones that say "test your procedure by finding the largest prime after 100000000" or whatever
2009 Apr 13 01:19:18  <chrisconley> i haven't done it lately because i've needed more hints
2009 Apr 13 01:19:42  <aamar> ha, I'm glad I got the scheme interpreter working -- I just hit C-x C-e and it tests if I'm right
2009 Apr 13 01:19:51  <aamar> I guess rudybot_ here is just as good
2009 Apr 13 01:19:55  <inimino> which shows in the number of dumb mistakes that remain in my answers, like calling functions with the wrong number of arguments ;)
2009 Apr 13 01:20:30  <chrisconley> inimino: i always wondered why you had plain text files :)
2009 Apr 13 01:20:42  <inimino> that's why :-)
2009 Apr 13 01:21:12  <inimino> I can cut and paste into MIT/GNU scheme when necessary
2009 Apr 13 01:22:12  <inimino> for ones like "what is the result of (fold-left 1 / (list 1 2 3))" I seem to learn better by working it out mentally
2009 Apr 13 01:22:33  <chrisconley> i should try to start the problems on paper more often; sometimes you can cheat by testing and not fully understand
2009 Apr 13 01:22:48  <chrisconley> yeah
2009 Apr 13 01:23:32  <chrisconley> i guess i should replace "you can cheat" with "you end up cheating"
2009 Apr 13 01:23:32  <inimino> all of this accumulate and map and fold stuff is actually quite dense
2009 Apr 13 01:24:19  <inimino> even though I was exposed to some of it previously, I still get confused a lot
2009 Apr 13 01:24:20  <chrisconley> yeah there a lot of layers to keep in your head
2009 Apr 13 01:24:42  <aamar> definitely attempted to cheat in 2.43, though it turned out not to help
2009 Apr 13 01:25:39  <inimino> just think of the poor undergrad with no programming experience trying to get through this :)
2009 Apr 13 01:26:26  <inimino> as an introduction to programming it really is pretty hardcore
2009 Apr 13 01:26:28  <chrisconley> seriously! i never would've survived at the pace they go at
2009 Apr 13 01:27:14  <aamar> some of the math might be fresher in mind though.
2009 Apr 13 01:27:22  <aamar> it's been a while since I did any calculus.
2009 Apr 13 01:27:35  <inimino> yeah, same here
2009 Apr 13 01:27:58  <chrisconley> yeah true
2009 Apr 13 01:28:35  <inimino> though I probably would have choked on matrices as a freshman
2009 Apr 13 01:29:07  <inimino> but then I didn't go to MIT either ;)
2009 Apr 13 01:29:36  <inimino> hm, where are we, 2.41?
2009 Apr 13 01:30:07  <inimino> http://inimino.org/~inimino/projects/2009/SICP/chap_2/2.41
2009 Apr 13 01:30:25  <inimino> that's the last one I finished
2009 Apr 13 01:30:57  <chrisconley> http://github.com/chrisconley/sicp-exercises/blob/66ffd41c76d8840c5987d60d7e778a616fbd5f74/chapter2/ex2-41.ss
2009 Apr 13 01:30:58  <aamar> http://mibbit.com/pb/7S0EgC -- 2.41
2009 Apr 13 01:32:51  <inimino> (cadr (cdr x)) -> (caddr x)
2009 Apr 13 01:33:41  <chrisconley> all look similar
2009 Apr 13 01:34:07  <aamar> yep, good point
2009 Apr 13 01:34:22  <inimino> yeah they look the same
2009 Apr 13 01:34:59  <aamar> chrisconley: yours does an additional map, but conceptually I think it's the same
2009 Apr 13 01:35:24  <chrisconley> yeah the extra map is just to create a list with the triple and the sum
2009 Apr 13 01:35:41  <chrisconley> ((6 5 4 . 15)
2009 Apr 13 01:35:41  <chrisconley>  (7 5 3 . 15)......)
2009 Apr 13 01:35:49  <aamar> ok, got it
2009 Apr 13 01:36:01  <chrisconley> similar to make-pair-sum in the book
2009 Apr 13 01:36:07  <inimino> yeah
2009 Apr 13 01:36:34  <inimino> we all came out with the very same unique-triples
2009 Apr 13 01:36:55  <inimino> so we must be doing something right :)
2009 Apr 13 01:36:57  <aamar> I prob. should have used a fold for triple-sum like both of you did.
2009 Apr 13 01:37:30  <inimino> I thought about doing it your way
2009 Apr 13 01:37:33  <chrisconley> heh, yeah, i would've thought our solutions would be pretty diverse
2009 Apr 13 01:37:52  <inimino> but it just seemed easier
2009 Apr 13 01:38:39  <chrisconley> the wife has dinner ready, so i should run
2009 Apr 13 01:39:03  <inimino> it was funny how they just introduced cadr without comment in an example
2009 Apr 13 01:39:18  <inimino> chrisconley ok
2009 Apr 13 01:39:27  <aamar> ok cool
2009 Apr 13 01:39:47  <chrisconley> see you  in two weeks? up to 2.52? or more?
2009 Apr 13 01:40:08  <aamar> how about 2.58 ?
2009 Apr 13 01:40:24  <inimino> that sounds good
2009 Apr 13 01:40:44  <inimino> I guess the next ten should not take us a whole two weeks (though I haven't looked at them)
2009 Apr 13 01:41:03  <aamar> 2.44 through 2.48 are also trivial
2009 Apr 13 01:41:10  <chrisconley> yeah that sounds good
2009 Apr 13 01:41:12  <chrisconley> cool
2009 Apr 13 01:41:18  <aamar> ok cool
2009 Apr 13 01:41:22  <aamar> let's go for it
2009 Apr 13 01:41:23  <chrisconley> ok see ya then!
2009 Apr 13 01:41:25  <inimino> sounds good
2009 Apr 13 01:41:35  <aamar> ok catch ya in a couple weeks
2009 Apr 13 01:41:37  <chrisconley> (and thanks again for help on 42 earlier!)
2009 Apr 13 01:41:45  <chrisconley> see ya