00:48 (quit) corruptmemory: Ping timeout: 255 seconds
00:50 (join) neilv
00:50 (quit) neilv: Changing host
00:50 (join) neilv
01:14 (quit) jonrafkind: Ping timeout: 250 seconds
01:32 (join) evhan
01:34 (quit) evhan_: Ping timeout: 255 seconds
01:46 (join) mheld__
01:49 (quit) mheld: Ping timeout: 255 seconds
01:49 (nick) mheld__ -> mheld
02:17 (join) Senjai
02:17 Senjai: Hey guys
02:19 Senjai: Can anyone help me understand Binary Search Trees?
02:23 Lajla: Senjai, what do you want to know?
02:24 Senjai: Well, I understand how they work, kind of.
02:24 Senjai: Makes it easier to search for objects using less iterations
02:24 Senjai: But I want to do the exercise in HTDP on creating a function to create a BST from a list of name value pairs
02:25 Lajla: Ahhh
02:25 Senjai: I just, can't get anywhere, I came to the conclusion that the list should be sorted in an ascending matter before attempting to create the tree..
02:25 Lajla: How much help do you want? =P
02:25 Lajla: No,not at all.
02:25 Lajla: THat defeats the purpose a bit.
02:25 Senjai: I just need a .. push .. in the right direction
02:25 Lajla: Basically, you want your insertion algorithm to work like this:
02:26 Lajla: Basically, you have nodes and 'leaves'
02:26 Lajla: leaves are basically theend of your tree and contain no other branches.
02:26 Senjai: Leaves being the nodes branching off the left and right correct?
02:26 Lajla: No
02:26 Lajla: the inverse
02:26 Senjai: oh.. :/
02:26 Lajla: leaves don't branch
02:26 Lajla: branches branch
02:26 Senjai: Sorry..
02:26 Senjai: haha
02:26 Lajla: But like, what you want is an algorithm that does this:
02:27 Senjai: so a Leaf is (make-node key value false false) right/
02:27 Lajla: yeah!
02:27 Lajla: And there is another type, which is sometimes called a leave.
02:27 Lajla: Which is a null-leave
02:27 Lajla: So a leave which has no value
02:27 Lajla: Test that wit (null-leave? node)
02:28 Senjai: Reference to undefined identifier
02:28 Lajla: So what you want is (if (null-leave? node) (make-leave key value) ...)
02:28 Lajla: And if it's not a null leave
02:28 Senjai: OH
02:28 Senjai: create a helper function
02:29 Lajla: You want to compare the key of the leave to the key of what you insert
02:29 Lajla: If it's less
02:29 Lajla: you just recurse to the left branch
02:29 Lajla: if it's more, then you do to the right branch
02:30 Senjai: I understand. Racket is proving difficult for me, I have some experience in C++ (self taught) which makes the concept of racket.. hard to understand
02:30 Senjai: Now I'm in a comp sci class making us learn Racket, and I'm falling apart.. lol
02:30 Lajla: Senjai, ahhh
02:30 Lajla: I guess what is different here is that you don't want to update values per se.
02:31 Lajla: But this is a task that lends itself well for recursion.
02:31 Senjai: Okay let me write out a template
02:31 Lajla: Basically, you have a function which works like this (tree-with-value tree key value)
02:32 Lajla: What it does is it takes a tree as argument, a key and its value
02:32 Lajla: and produces the new tree with the value inserted
02:32 Lajla: the old tree is persisistent, it does not getmodified.
02:32 Lajla: And that can call itself recursively
02:34 Senjai: So my initial check-expect is (check-expect (create-bst (list 1 2) (list 2 2) (list 3 3)) (make-node 2 2 (make-node 1 2 false false) (make-node 3 3 false false)))
02:37 Senjai: I'm stuck, what should I do from here? http://pastebin.com/GhegQWXL
02:38 Senjai: My theory is still a bit rough I think, If someone wants to search the tree, they would have to start from the top of the tree right? Otherwise they can miss values, but can't there be more than one top of the tree?
02:38 Senjai: Lajla: recursion sucks :/
02:40 Lajla: Senjai, http://pastebin.com/dLqv7TW0
02:40 Lajla: You want something along the lines of that
02:41 Senjai: That's amazing, I'm unfamiliar with (let) though, and I think that's.. pointers in racket? I am in a Comp Sci class though, (don't worry this isn't a homework question, htdp just doesnt post solutions) I don't think I'm supposed to know that way yet.
02:43 Senjai: Lajla: look at this: http://pastebin.com/v5FXseyk Lecture notes for binary search trees, that's how he wants them done..
02:43 Senjai: Lajla: or at least the way he taught us
02:44 Lajla: That is not what I call a very clean, simple and elegant way.
02:44 Lajla: And * is part of the variable name
02:44 Lajla: identifiers are very permissive in scheme
02:45 (join) evhan_
02:45 Lajla: */7236 is okay as a variable name
02:45 Lajla: In fact * and -  and / and + are all technically variable names
02:45 Lajla: I basiclaly used key as the key of the node inserted node and key* as the key of the one I'm comparing
02:45 Senjai: Lajla: scary.. See, this issue in C++ would be easy, I'd just create a class for it, and use member functions to deal with it.
02:46 Senjai: What does let do?
02:46 Lajla: And (let ((name value) (name2 value2)) ...) is like {name = value, name2 = value2; ...} in C.
02:46 (join) mheld__
02:46 Lajla: Senjai, you can use classes in racket too.
02:46 Senjai: isn't that similar to define?
02:46 Senjai: Lajla: we're not allowed to do anything other than (define-struct) lol
02:47 Lajla: No, define defines a variable and binds some definition to it. =P
02:47 Lajla: That is weird, I assume you want to also define functions
02:47 Lajla: Basically, 'functions' and 'values' are the same.
02:47 Lajla: (define foo 3)
02:47 Lajla: foo now houses a number
02:47 Senjai: Yes
02:47 Lajla: (set! foo +)
02:47 Lajla: foo now houses the addition function
02:47 Senjai: Can't use set :P
02:48 Senjai: See if I could LIFE WOULD BE SO EASY
02:48 (quit) evhan: Ping timeout: 255 seconds
02:48 Lajla: Well, I did it without. =P
02:48 Lajla: And also, set! does not change any values
02:48 Senjai: I've talked to my prof.. but he wants us to learn without assigning values
02:48 Lajla: it changes a binding
02:48 Lajla: Good guy, good guy. =P
02:48 Lajla: But look at my code
02:48 (quit) mheld: Ping timeout: 264 seconds
02:48 (nick) mheld__ -> mheld
02:48 Lajla: (let ....) is just a way to create new bindings
02:49 Lajla: The code is not operational by the way, it's more a schematic of the basic algorithm.
02:49 Senjai: yea
02:49 Senjai: I know
02:50 Senjai: That will insert, (i have to review insertion too) but what about reading from a list of lists?
02:50 Senjai: (list (list 1 2) (list 2 4)) etc
02:51 Lajla: Ahhh
02:51 Lajla: Well
02:51 Senjai: Im trying to find the question from htdp, I'll pastebin it, I just want to do it so Im sure I get BST's
02:52 (quit) evhan_: Quit: leaving
02:53 Lajla: Senjai, http://pastebin.com/M4ztDnzb
02:53 Lajla: Bottom part explains the general algorithm
02:53 Lajla: Or wait
02:53 Lajla: I can do better than that
02:55 Senjai: Lajla: Look at this http://htdp.org/2003-09-26/Book/curriculum-Z-H-1.html#node_toc_node_sec_14.2
02:55 rudybot: http://tinyurl.com/6z8vxvm
02:55 Senjai: Exercise 14.2.5.   Develop the function create-bst. It consumes a BST B, a number N, and a symbol S. It produces a BST that is just like B and that in place of one false subtree contains the node structure
02:56 Lajla: Senjai,
02:56 Lajla: http://pastebin.com/ntVyNmvY
02:56 Lajla: What does N stand for?
02:56 Senjai: N being the key
02:56 Senjai: S is the value
02:57 Senjai: then the left and right branchest
02:57 Senjai: branches*
02:57 Lajla: Ahh
02:57 Lajla: I just called that key and value
02:58 Lajla: But that is what my tree-with-value thing does right?
02:58 Lajla: It takes a tree, a key and a value
02:58 Senjai: I.. don't know?
02:58 Lajla: and produces the new tree
02:58 Senjai: But this doesnt
02:58 Senjai: err.
02:58 Senjai: it can
02:58 Senjai: but calling (create-bst false 40 "value for key #40) produces a leaf
02:59 Senjai: because a BST can be false, or (make-node key value left right)
02:59 (quit) mheld: Quit: mheld
02:59 Lajla: Senjai, yeah
02:59 Lajla: That's what mine does
02:59 Senjai: I don't udnerstand yours though lol
02:59 Senjai: If you look atth e exercise
02:59 Lajla: Except that I abstract how to represent null-trees.
02:59 Senjai: I have to create create-bst to create a single node.
02:59 Lajla: (define (tree-with-value tree key value)
02:59 Lajla: (if (null-tree? tree)
02:59 Lajla: (make-leave key value)) ; should make a leave node with no branhes
02:59 Lajla: Look at that part of the code
03:00 Senjai: THEN I have to create a function that recursively calls create-bst to create the total tree
03:00 Lajla: Which is what mine does
03:00 Lajla: Look, it calls itself
03:00 Senjai: that makes sense
03:00 Senjai: but i Dont understand this example
03:00 Senjai: create-bst (create-bst false 66 'a) 53 'b) = (make-node 66               'a              (make-node 53 'b false false)              false)
03:00 Senjai: oops
03:00 Senjai: thats ugly
03:00 Senjai: one sec
03:01 Senjai: http://pastebin.com/5AGkxWmQ
03:01 Senjai: the book says calling it like that, should = the following
03:01 Senjai: So when I design this function, I would assume that it would be Like (pastebinning)
03:03 Lajla: I'm not sure I get what you mean here.
03:06 Senjai: Lajla: so to my understanding the function looks like: http://pastebin.com/nU2e121n
03:06 Senjai: in order to create a single node
03:06 Senjai: according to the book @ http://tinyurl.com/6z8vxvm (look  at the exercise)
03:07 Senjai: But even my code doesn't work, I mean how can you assume everything coming off of the branch is  a leaf? or.. ught
03:07 Senjai: It gives me the correct answer though..
03:13 Lajla: Senjai: http://pastebin.com/4NLne2kY
03:15 Senjai: Lajla: reading
03:15 Lajla: Senjai, it's more the C++ esque code there that should perhaps explain the algorithm better
03:16 Senjai: Lajla: SO much better, I understand the pseudocode perfectly, but it appears I'm missing my logic on recursion is all
03:16 Senjai: I guess I can't wrap my hea don recusive functions
03:16 Senjai: I was one of those programs who always used loops, never recursions, and recursions are INTEGRAL in racket.
03:17 Lajla: Senjai, it takes time I suppose to get a feel for the recursive approach
03:17 Senjai: programmers* sorry its late
03:17 Lajla: Senjai, but the C++ code is in essense the same as the racket code
03:17 Senjai: Lajla: Yea, I just didnt get all the functions you used is all lol, and its the recursion that is getting me, give me a moment, I need to stare at this for like a minute
03:18 Lajla: Senjai, well, a lot of the functions I used in either don't really exist yet, it's up to you to make them. =P
03:18 Lajla: It's just the general algorithm
03:18 Lajla: But yeah, let it sink.
03:18 Senjai: Can i write this out in the way I know how and see if it's correct
03:18 Senjai: Hence pseudocode :P
03:19 Senjai: Let me write this up
03:19 Senjai: btw, thank you SO much for taking the time to write pseudocode for me.
03:19 Senjai: Can i check it with you after I write it/
03:22 Lajla: Sure
03:22 Lajla: and it's the morning
03:22 Lajla: I just woke up, doing stuff like this helps me clear my mind for the day.
03:22 Lajla: So it's not a problem.
03:28 (quit) sethalves: Ping timeout: 276 seconds
03:28 Senjai: Lajla: http://pastebin.com/LvBjqWWK
03:28 Senjai: Lajla: Thank you so much :)
03:31 Lajla: Senjai, yeah, that looks like what you need.
03:32 Lajla: I praefer to ue the let-block to keep stuff more organized though, but it doesn't really matter
03:32 Senjai: Lajla: I prefer that as well, though my prof would probably not allow it.
03:32 Lajla: He doesn't like the let block?
03:32 Senjai: Lajla: still trying to get my head around the recursion, No, he just hasnt taught us it yet, so he doesnt want us to use it
03:32 Lajla: Ahhh
03:33 Lajla: I was never a fan of that technique
03:33 Senjai: He is trying to get us to program with *minimal bindings/definitions*
03:33 Senjai: So we only use (define) and (define-struct) that is IT
03:33 Senjai: lol
03:34 Senjai: I encountered a problem
03:34 Lajla: What?
03:36 Senjai: Lajla: http://pastebin.com/0baNkw20
03:36 Senjai: I did some tests at the bottom and included the results
03:38 Lajla: Senjai, you inverted the left and the right
03:38 Lajla: You want to inserted it to the left when the < case applies, not the right
03:39 Senjai: O..M..G
03:39 Senjai: Im a loser
03:40 Senjai: That fixed that ine
03:40 Senjai: That fixed that one*, Ill try a more complicated one.
03:41 Senjai: I just kinda clued in a bit to my recursion issues
03:41 Lajla: Well, I didn't notice it when I looked the first time. =P
03:41 Senjai: Because the input is a BST, then you can assume that the BST is already properly sorted
03:41 Lajla: But it's one of the first thngs you look at when it's inserted at the wrong end I suppose.
03:41 Senjai: Beecause it has to be if it is a BST :P
03:41 Lajla: Senjai, yeah
03:41 Lajla: The algorithm does work on that assumption
03:42 Senjai: Ima try a more complicated oe :)
03:43 Senjai: err.. actually
03:44 Senjai: typing it out would be too complicated haha, Ill try making a function now to consume a list of name value pairs and produce a tree from that... This is the hard part
03:44 Senjai: Because I don't know if the list is sorted by key or not..
03:47 (quit) neilv: Ping timeout: 252 seconds
03:48 (join) neilv
03:48 (quit) neilv: Changing host
03:48 (join) neilv
03:48 Senjai: How should I start
03:52 Lajla: Senjai, oh, it doesn't have to be sorted actually
03:52 (quit) misterm: Read error: Operation timed out
03:52 Lajla: What you basically want is
03:53 Lajla: for the empty list return a null-tree
03:53 Senjai: Lajla: Then how can I do it? Im thinking of recursing ovver groups of three, and passing them onto the recursed.
03:53 Senjai: got that done
03:53 Senjai: (empty? lokvp) false
03:53 Senjai: OH
03:54 Lajla: Well, the recursion step would be (create-bst (lokvp->bst (rest lokvp)) (first-key lokvp) (first-value lokvp))
03:54 Lajla: Not sure if I'mdoing too much for you here.
03:55 Senjai: hmm
03:55 Senjai: It's so easy for you to do this lol
03:55 Lajla: Well, I am brilliant after all.
03:55 Lajla: NAhh, you'll get the hang of thinking recursively in a while.
03:56 Senjai: Ugh its that assumption that (lokvp->bst) returns a bst lol
03:56 Senjai: urgh
03:57 Senjai: But.. how do you position the nodes?
03:57 Senjai: E.g. make sure they go where they are supposed to
03:57 Lajla: It does so automatically. =P
03:57 Lajla: create-bst ensures it
03:57 Lajla: Basically, the logic is
03:58 Lajla: create-bst will return a correctly formed tree if its first argument is a correctly formed tree, right?
03:58 Lajla: A null tree is correctly formed.
03:58 Lajla: And lovkp->bst returns a correctly formed tree because it always returns a call to create-bst
03:59 Lajla: And therefore create-bst always consumes a correctly formed tree
03:59 Lajla: so the tree returned willalways be correctly formed
04:00 Senjai: yea makes sense
04:01 Senjai: so is it like this: http://pastebin.com/djSxTnXL
04:01 Senjai: is it really that simple?
04:07 Senjai: Lajla: If you look at the code, I copied my profs render tree function and rendered it out: http://imagebin.org/141411 << the image result
04:09 (join) mheld
04:09 Senjai: Lajla: What I don't understand about recursion is that (create-bst (lokvp->bst (rest lokvp)) ... ....)
04:10 Senjai: I know we're supposed to think: Okay, lokvp->bst always produces a bst, and create-bst inserts it properly, etc.
04:10 Senjai: but it actually doesn't
04:11 Senjai: I mean, create-bst should recieve the first node, but it just receives a list of nodes..
04:16 Senjai: Lajla: still there?
04:21 Lajla: Senjai, let me look
04:21 Lajla: was drinking coffee
04:21 Lajla: the one drug legalized since the dawn of man
04:22 Lajla: Senjai, it doesn't in what case?
04:23 Lajla: create-bst always receives a tree as first argument
04:23 Senjai: Lajla: lol funny, I'm drinking indulging in coffee as well, but it is night here, morning for you lol
04:23 Lajla: GMT+1 baby
04:23 Senjai: I know but where does lokvp  actually supply an individual tree?
04:23 Senjai: GMT-8 :)
04:23 Lajla: Senjai, supply?
04:23 Lajla: What do you mean?
04:23 Lajla: lokvp->bst
04:24 Lajla: evaluates to either the null-tree (false)
04:24 Senjai: create-bst requires a tree right?
04:24 Lajla: Or to a call to create-bst, rigt?
04:24 Senjai: yes
04:24 Lajla: Okay, so in the first case, it's a well-formed tree
04:24 Lajla: trivially
04:24 Lajla: In the second case, it also is, because create-bst returns a well-formed tree provided it gets one
04:25 Senjai: create-bst requires a tree to function, lokvp->bst provides that, but I don't see where it provides just an individual tree
04:25 Lajla: And it gets a well-formed tree because (lokvp->bst ...) returns one.
04:25 Lajla: It always evaluates to an individual tree
04:25 Lajla: either a null-tree, or a call to create-bst
04:25 Senjai: At least Im starting to locate my problems with recursion
04:25 Lajla: both of which return individual trees
04:26 Senjai: but the call to create-bst doesnt get a single bst, it gets (lokvp->bst (rest lokvp)) which will supply create-bst with a stream of trees right? But were does that functionalyity take place?
04:26 Lajla: No, that will return a single tree
04:27 Lajla: Not a list of trees
04:27 Lajla: Just look at how lokvp->bst is defined
04:27 Senjai: I understand
04:27 Lajla: it either returns a null-tree (a single tree) or whatever the call to create-bst returns, which is also a single tree
04:27 Senjai: But how? Where?
04:28 Lajla: http://pastebin.com/djSxTnXL there, at the top
04:28 Lajla: you defined it
04:28 Senjai: but there is something happening under the covers that when the call to create-bst takes place
04:28 Senjai: lokvp magically supplies it individual trees right?  because all lokvp->bst can do is produce random nodes
04:29 Senjai: create-bst has to sort them
04:29 Lajla: Yes, create-bst is the part that ensures that the trees are sorted
04:29 Senjai: What I'm asking is how the compiler, at runtime, does this recursion
04:29 Senjai: The stepper is too.. burdening to look through
04:30 Lajla: Ehh, it generates functions that call itself. =P
04:30 Senjai: renames them right?
04:30 Lajla: Senjai, no, not that I know
04:30 Lajla: function names no longer exist at run time.
04:30 Senjai: like with (local) it renames and lifts
04:30 Lajla: They are just identifiers used by the compiler to establish bindings at compile time
04:31 Senjai: hmm yea, renaming them would be inefficient actually, for large iterations it would suck up unneccesary ram'
04:31 Senjai is installing Adobe CS5 Master Collection ;)
04:32 Senjai: but do you see what my issue is? (create-bst (lokvp->bst (rest lokvp)) passes trees, one by one, to create-bst.. I understand that. But how, If I were to handwrite the execution, it would always be calling (rest lokvp) until it hits empty.
04:33 Lajla: Yeah
04:33 Senjai: I don't see where (create-bst) actually gets the individual trees is all
04:33 Lajla: Well, that is what lokvp->bst does
04:33 Lajla: It consumes a list of lists
04:33 Lajla: And produces a tree
04:33 Lajla: Basically, see it like this
04:34 Senjai: but if you look at the code, there is no (create-bst (makebst (first lokvp))), that makes sense to me
04:34 Lajla: We have a list of lists A := ((40 john) (30 claire) (50 Bill))
04:34 Senjai: because it is passed ONE value, before the recursion happens
04:34 Lajla: So, lokvp->bst on that creates a tree based on that right?
04:34 Lajla: Well, how does it do that
04:35 Lajla: It first simplifies to (create-bst (lokvp->bst '((30 claire) (50 bill))) 40 'john)
04:35 Lajla: Do you follow that step?
04:38 Lajla: Senjai?
04:39 Senjai: hold on
04:39 Senjai: Lajla:  Yes im following
04:40 Senjai: Lajla: I just created pseudocode for how I would do it: http://pastebin.com/4iAWdkdz in C++
04:40 Senjai: Lajla: I'm stuck in that way of thinking
04:40 Lajla: No need to sort it like that, it will go okay on itself.
04:40 Lajla: Okay, but
04:40 Lajla: That wil simplify to this:
04:41 Lajla: (create-bst (create-bst (lokvp->bst '((50 bill))) 30 'claire) 40 'john)
04:41 Lajla: And that will then simplify to this:
04:42 Senjai: Ok yes I see
04:42 Lajla: (create-bst (create-bst (create-bst (lokvp->bst '()) 50 'vill) 30 'claire) 40 'john)
04:42 Lajla: And finally, the base case:
04:42 Lajla: (create-bst (create-bst (create-bst false 50 'bill) 30 'claire) 40 'john)
04:42 Senjai: Oh
04:42 Senjai: Wow that makes a lot of sense actually
04:42 Lajla: As you can see, it expands in the end to multiple calls to create-bst, starting with a null-list and then just inserting all the elements.
04:43 Lajla: Each one feeding its result to the new call.
04:43 Lajla: Beautiful, isn't it?
04:43 Senjai: I forgot the key and value arguements allow create-bst to seperate the nodes
04:44 Senjai: Then after lokvp->bst recurses, create-bst works BOTTOM up.. to make sure everything is in the right place?
04:44 Senjai: I was thinking top down.. :/
04:45 Senjai: I need toast (too much coffee on an empty stomach
04:45 Lajla: coffee is good to help you adjust to recursion
04:45 Senjai: at 1:45 AM
04:45 Senjai: lol
04:45 Senjai: brb
04:46 Lajla: A fixed sleeping rhythm is for people who atually do some-thing productive with their life lol.
04:47 neilv: sleep rhythms are important, short-term and long-term. if you kick caffeine, in two weeks you'll be more consistently productive
04:48 Lajla is sitting comfortably on a few hours of sleep each 3-4 days for some time now.
04:48 Lajla: I can reccomend it, you get so much more stuff done by actually not spending half of your time onconscious
04:49 Lajla: Now all I have to do is is find stuff to get done.
04:49 Senjai: Lol, I havent slepped for a day or so
04:49 Senjai: You gain weight though, with caffeine, and not sleeping
04:49 Senjai: A lot of weight
04:50 neilv: ok, i'm off to code some racket
04:50 (quit) neilv: Quit: Leaving
04:50 Lajla: feeling important are we.
04:51 Lajla: Senjai, I have been sitting comfortably on about 55 kg for some time now.
04:51 Lajla: I wouldn't recommend that though
04:54 Senjai: lol
04:54 Senjai: Thats ddecent
04:54 Senjai: Why code racket Lajla ?
04:54 Lajla: Senjai, I don't really.
04:54 Lajla: I code in Scheme
04:54 Lajla: I use racket and icarus as implementations
04:54 Senjai: Lajla: difference?
04:55 Lajla: but racket also supports a scheme-ish language commonly called 'racket'
04:55 Lajla: Senjai, scheme is much, much, much, much, much, much smaller.
04:55 Senjai: So why use scheme instead of racket
04:55 Lajla: racket can be called a 'dialect of scheme', though it's not a superset of any official scheme standard
04:55 Lajla: Well, scheme is supported by more than one implementation.
04:56 Senjai: As in?
04:56 Lajla: And most of the things racket offers you can make yourself anyway.
04:56 Senjai: Scripting? I know emacs uses lisp
04:56 Lajla: As in icarus, stalin, bigloo, chez, chicken, guile, to name a few.
04:56 Senjai: is lisp the same as scheme?
04:56 Lajla: No.
04:56 Senjai: But what do they do?
04:56 Lajla: scheme is a  'dialect of lisp'
04:56 Senjai: Oh wow..
04:56 Senjai: This is getting confusing
04:56 Lajla: Senjai, well, racket started off as a scheme implementation but it begain adding extensions and at some points breaking compatibility with some standards
04:57 Lajla: Senjai, well, the different dialects of lisp are as divergent as Java and C really
04:57 Lajla: There are currently three popular dialects, common lisp, scheme, and clojure.
04:57 Senjai: sameish syntax but different libraries?
04:57 Lajla: Syntax is similar, semantics are fastly different.
04:57 Lajla: Most prominently, in common lisp, functions and values are not the same as they are in clojure and scheme
04:58 Senjai: Why use any of them though? I mean racket is amazing for a learner, to be able to test so quickly, but doesnt hold a time to C in power, speed, or efficiency
04:58 Senjai: dime*
04:58 Lajla: Senjai, well, racket is not an implementation geared for speed
04:58 Lajla: If you try out stalin ...
04:58 Lajla: It gets its name because it 'brutally' optimizes. =P
04:58 Lajla: Icarus is also pretty fast
04:59 Lajla: Scheme is also far more expressive than C and makes for easier to maintain code.
04:59 Lajla: And its a language that suits me I guess.
04:59 Senjai: But what do you use these dialects for? You know, every dialect has their niche community, java and opensource/web/portability, C#/.NET/Windows lovers etc
05:00 (join) lithpr
05:00 Lajla: Depends on praeference, though in the end Common Lisp and Clojure are more geared towards a business perspective than Scheme in the end.
05:00 Senjai: Business applications?
05:00 Lajla: Scheme has a tendency to not sacrifice elegance for speed.
05:00 Senjai is a Commerce major, good to know :))
05:00 Lajla: Or less so than Clojure
05:00 Lajla: But Common Lisp is ooold, and it shows
05:00 Lajla: It's not really as robust as a more modern language like Clojure.
05:01 Lajla: Each has its strength and weaknesses I suppose.
05:01 Lajla: Most importantly though, Scheme is the only dialect to allow fully hygienic macros.
05:01 Senjai: Can these dialects still produce executable files? or is it more similar to scripts run by a service on the system? Can they use the Win32 API, or a graphics library like OpenGL
05:01 Lajla: Which basically means that you can write new syntax which is as expressive as primitive syntax.
05:01 Senjai: Macros are scary things
05:01 Lajla: racket comes with a compiler.
05:01 Lajla: Not scheme macros.
05:02 Lajla: They are fully hygienic, they really never break.
05:02 Lajla: A lot of syntax you've been using in racket are actually macros.
05:02 Senjai: Wow, Id have to look into it then, but about the other question
05:02 Lajla: Well, racket comes with a compiler, like most implementations
05:02 Senjai: Racket can, but not without linking a crapload of DLL files, etc,
05:02 Lajla: Scheme and Clojure are easier to compile I would say than common lisp because of the lexical scope.
05:03 Lajla: Well, you can create standalone stuff and include them of course.
05:03 Senjai: meaning it does not "compile" its like racket packages it with a framework that reads the script
05:03 Lajla: No, it doesn't read the script.
05:03 Lajla: variable names and stuff like that is all gone
05:03 Senjai: Well I mean, the preprocessed version
05:03 Lajla: Clojure compiles to java bytecode and .net common language I believe.
05:03 Senjai: but its not machine code, or morphed into assembly right?
05:04 Lajla: Yes, it is.
05:04 Senjai: Really?
05:04 Lajla: Yeah
05:04 Senjai: Wow, i was wrong
05:04 Lajla: there is no reason scheme could not.
05:04 Senjai: Hey, I'll be right back, I have to reboot chrome to finish CS5 Installation (using webbased irc)
05:04 Lajla: Lisp has been at the vanguard of compiliation technology since 1950's or something.
05:04 Senjai: Give me a sec?
05:04 Lajla: Sure.
05:05 (quit) Senjai: Quit: Page closed
05:07 (join) Senjai
05:07 Senjai: Back :)
05:08 Senjai: Lajla: still here?
05:08 Lajla: Senjai, sure
05:09 Lajla: Senjai: http://en.wikipedia.org/wiki/Ikarus_Scheme#Design
05:09 Lajla: I think this is generally accepted as one that delivers the fastest code.
05:09 Lajla: Stalin delivers faster but also takes 30 billion years to compile.
05:10 Senjai: haha ptimization?
05:11 Senjai: optimization?
05:11 Lajla: Sure
05:11 Lajla: Stalin is one of the most optimizing compilers on the planet. =P
05:11 Lajla: It was basically created with the idea 'let's see how far optimization is possible'
05:11 Lajla: It's a scheme -> C compiler but it generates completely unreadable and indeciphrable C code.
05:11 Lajla: But it's fast, apparently.
05:12 Lajla: A lot faster than what is usually possible by writing the C program by hand
05:12 Senjai: Well it doesnt have to be neat if a human isnt intended to read it right?
05:12 Lajla: Yeah
05:12 Lajla: And Scheme code (to me at least) is much more readable than C code.
05:12 Lajla: It just takes a year to compile
05:12 Senjai: I feel uneasy by having anything more than What I Code In -> asm -> machine
05:12 Senjai: Like if something involves more steps than that I don't like it
05:13 Senjai: So if Stalin -> C -> asm -> machine I much rather would use C instead
05:13 Senjai: I like control over my code
05:13 Senjai: For every layer you add, you trust another programmer to do his job properly
05:13 Lajla: Well
05:13 Senjai: More steps, more places to break
05:14 Lajla: internally, almost every compiler uses a lot of intermediate steps.
05:14 Senjai: Well yea
05:14 Lajla: Which is actually what guarantees correctness more.
05:14 Lajla: Because each step can be done by a specialist in that field.
05:14 Senjai: But in terms of translating code to code. It's never going to be better, because you lose control over memory management, etc.
05:14 Lajla: I once read about how the GHC worked, that was a plethora of intermediate languages, most designed specifically for the GHC.
05:14 Senjai: like higher level to higher level is pointless. I think
05:14 Lajla: No, not really.
05:15 Senjai: GHC?
05:15 Lajla: It actually optimizes beter
05:15 Lajla: Glasgow Haskell Compiler
05:15 Lajla: A pretty impressive beast
05:15 Senjai: Wow I suck.. lol
05:15 Lajla: Never heard of it?
05:15 Senjai: Haskell is another type of code
05:15 Senjai: Ive heard of Haskell
05:15 Senjai: but I dont know anyone who usees it
05:16 Lajla: I sometimes 'use' it.
05:16 Lajla: I'm not a big fan of it though
05:16 Lajla: Like, it's whitespace significant
05:16 Lajla: It uses indentation rather than good old { ... } to determine block structure
05:16 Lajla: I'm always uncomfortable with that
05:16 (join) jake
05:17 DT``: afaik the forced indentation in Haskell is optional, you can use {}.
05:17 Senjai: Eww!
05:17 Senjai: I would screw that up all the time!~
05:17 Lajla: DT``, yeah, but then I get even more confused.
05:18 Lajla: What I often use in C is that I don't use { ... } if I only have one expression
05:18 Lajla: No matter the people who complain about the style.
05:18 Lajla: And it'smore the let ... in syntax's indentation rules that I'm not really comfortable with yet.
05:19 (quit) jake: Client Quit
05:19 Senjai: Im so stoked I have Adobe CS5 lol
05:20 (quit) lithpr: Remote host closed the connection
05:21 Lajla: Senjai, you mean photoshop?
05:22 (join) lithpr
05:23 Senjai: No
05:23 Senjai: I mean the whole collection
05:23 Lajla: Ahhh
05:23 Lajla: I have some of that
05:23 Lajla: pirated, ofc.
05:23 Senjai: Photoshop Dreamweaver, browserlap, after effects, premeire, flash
05:23 Lajla: And it's even legal where I live.
05:23 Senjai: the whole thing :)
05:23 Lajla: Eewl, dreamweaver.
05:23 Senjai: My Uni lets i=us have it for free
05:23 Senjai: dreamweaver is AMAZING
05:23 Senjai: The new flash, you can make flash apps without code.
05:24 Senjai: it's done really well for none-coders
05:24 Senjai: and you can export to iphone formats etc..
05:24 Senjai: Back to racket!
05:24 Senjai: must learn.. study..
05:25 cpach: Lajla: pirating is legal where you live...?
05:27 Senjai: cpach, probably in sweden?
05:28 cpach: i'm in sweden. pirating is not legal here.
05:29 Senjai: Shit your in sweden, lucky..
05:30 Senjai: Technically, piracy is legal as it is defined. Modifying and distributing software is not. Sharing is. The problem comes when you cant distinguish the difference
05:30 Senjai: I agree with free software for personal use, and charge commercial users.
05:31 Senjai: If your not making money with a product, then dont charge for it. Though it would be hard to set that up
05:31 Senjai: Impossible even
05:31 rapacity_: it's hard to find legal software in my country, as most of the stores sell pirated software/films
05:31 Senjai: rapacity_: where are you from
05:32 cpach: i thought that bsa et al had cracked down on things like that, but apparently not
05:32 rapacity_: secret :p but it's ranked in the top 40 worldwide for piracy
05:32 Senjai: Lajla: still here?
05:32 (quit) mheld: Quit: mheld
05:33 Senjai: rapacity_: you cant even say the country name? Crap you have a hostmask and using ssl
05:33 Senjai: lol
05:33 cpach: Senjai: when the winter season is taken into account, i'm not sure i would consider myself lucky to be here ;-)
05:33 Senjai: cpach: take a look at the girls dude.. haha
05:33 Senjai is Canadian
05:35 cpach: Senjai: i knew that would come up! i never really understood that thing. maybe i should spend less time staring at a terminal window :)
05:35 Senjai: lol cpach
05:36 Lajla: cpach, yeah.
05:36 Senjai: are you a racket user? What distro you use? (assuming linux, you said terminal, could be mac)
05:36 Lajla: Basically, the logic is quite simple
05:36 Lajla: If I give you a bike or a cake or whatever
05:36 Lajla: Is it then your responsibility to check if Ilegallyobtained that
05:36 Lajla: or if I havethe intellectual property secured to give you that?
05:36 Lajla: No, it's not, it's mine.
05:36 Senjai: Lajla: Logic win :)
05:37 Lajla: In the Netherlands, the same principle applies to stuff 'given' to you online.
05:37 cpach: oh.
05:37 cpach: that's curious :)
05:37 Senjai has never seen this channel so active
05:37 Lajla: If you see a random link on the net, it's not your responsibility to go check if the site who offers that link has all legal stuff done properly.
05:37 Lajla: cpach, well,admit, it's actually quite sensible.
05:37 cpach: convenient at least :)
05:37 Lajla: cpach: http://akiscode.com/projects/SOS/86random-f1.1.gif
05:37 Lajla: If you click that link
05:38 Lajla: you download it
05:38 Lajla: Who knows
05:38 Lajla: it might be copyrighted?
05:39 Senjai: lol
05:41 cpach: i guess intent is often taken into account though. eg if you want to download something via the pirate bay, you most likely are actively searching for it.
05:41 cpach: Senjai: i'm trying to learn racket. i'm on ubuntu.
05:42 Senjai: cpach, haha I have ubuntu :)
05:42 Senjai: cpach on windows atm though
05:42 Senjai: cpach Im learning it too, you using htdp? I have to learn it for my course
05:42 cpach: i haven't used htdp that much so far.
05:43 Senjai: lol
05:43 DT``: cpach, sicp?
05:43 Senjai: Should have seen Lajla and I's intense BST discussion
05:44 cpach: Senjai: i saw that :) was those examples based on htdp?
05:44 Lajla: cpach, Dutch law doesn't feature 'intent' in that way very often, I guess because you can't prove that.
05:44 cpach: Lajla: cool :)
05:44 Lajla: We also have very little 'victim-less crimes'
05:44 Senjai: cpach yes
05:44 Senjai: the book
05:44 Lajla: The only thing is that recently sex with animals has again been criminalized, I can't say I agree.
05:45 cpach: DT``: yeah, i've worked through a small part of sicp. i really like that book.
05:45 Senjai: scip?
05:45 Senjai: sicp*
05:47 cpach: Senjai: http://mitpress.mit.edu/sicp/full-text/book/book.html
05:48 Senjai: Nice!
05:48 (join) lucian
05:49 cpach: definitively worth having a look at, imho.
05:50 cpach: i also really enjoyed working through _the little schemer_. it was a great way to get familiar with recursion.
05:50 (quit) lithpr: Remote host closed the connection
05:51 Senjai: I hatae recursion
05:51 Senjai: hate*
05:52 DT``: even tail-recursion?
05:53 bremner_: Senjai: scheme/racket may not be the language for you if you hate recursion.
05:53 Lajla: recursion is awesome.
05:53 Senjai: bremner_: I need to get better at it anyway, I'm just so accustomed to looping. I have to learn racket, its taught in our comp sci ocurse
05:54 Lajla: (define (recursion) (recursion))
05:54 Senjai: Lajla: once you figure it out
05:54 Lajla: Senjai, have they explained tail-recursion to you?
05:54 Senjai: Lajla: lol!
05:54 Senjai: Lajla: I dont remember
05:54 DT``: Lajla, ((λ (λ) (λ λ)) (λ (λ) (λ λ)))
05:54 Lajla: Senjai, well, you end up using recursion n mathematical definitions either.
05:54 Lajla: Senjai, that's another cool part of scheme
05:54 Lajla: You have no reserved words
05:55 Lajla: He just used an identifier that is commonly a keyword as a bound variable in a function
05:55 (join) lithpr
05:55 bremner_: just because you can do something...
05:55 DT``: that helps so much when writing obfuscated code.
05:56 bremner_: Is there an obfuscated scheme contest, like the old obfuscated C contest?
05:56 Senjai: Lajla: do you have an exampl?
05:56 DT``: I don't know, but it's possible.
05:57 DT``: Senjai, of tail-recursion?
05:58 Senjai: DT``: yea
05:58 Lajla: Senjai, well (define (recursion) (recursion))
05:58 Senjai: Okay but what is the point?
05:58 Senjai: How is that different from regular recursion?
05:59 lithpr: the neat thing about tail-recursion in scheme is that it runs in constant memory space
06:00 lithpr: so you don't use memory stack after memory stack
06:00 DT``: Senjai, the difference is:
06:00 lithpr: it's the "pure scheme" form of iteration
06:00 Lajla: Senjai, a simple example: (define (list-ref l k) (if (zero? k) (car l) (list-ref (cdr l) (- k 1)))
06:00 Lajla: This guy runs in constant space.
06:00 DT``: it's the last function called, you don't ``delay'' any computation after the recursion.
06:00 DT``: it's just a loop, written in ``functional'' style.
06:01 DT``: bremner_, http://paste.lisp.org/display/120250
06:01 Senjai: what does (car) and (cdr) do?
06:02 lithpr: Senjai: you should check out the "racket guide".  It explains this stuff very well (imo)
06:02 DT``: they are list accessors.
06:02 DT``: maybe you know them as ``first'' and ``rest''.
06:02 Senjai: DT``: yes I do
06:02 DT``: or head and tail.
06:02 Senjai: car is first? cdr is rest?
06:02 DT``: yeah.
06:03 Lajla: Senjai, yeah
06:03 Lajla: but car and cdr is so much cooler
06:03 Senjai: What do they stand for?
06:03 DT``: my car and my other car.
06:03 Lajla: contents of acessor part of register / contents of decrement part of register
06:03 Lajla: If I'm correct.
06:03 DT``: you're correct.
06:04 Lajla: Basically, the first lisp implementation was implemented on a machine that had these assembly instructions and it used it to implement lists
06:04 DT``: *address part
06:04 Senjai: My computer is lagging audio
06:04 DT``: Senjai, http://en.wikipedia.org/wiki/Car_and_cdr
06:04 Senjai: Lajla: you know everything lol
06:04 Lajla: and it was like 'Eurr, I need a name for them, let's just call them after the assembly'
06:04 Lajla: name them after*
06:04 Lajla: Senjai, never say that when chandler is around again.
06:05 Senjai: huhwhat?
06:06 Lajla: Senjai, well, let's just say that anyone with basic knowledge appears to know a lot when you're just starting to learn it yourself.
06:06 (quit) lucian: Read error: Operation timed out
06:07 Senjai: Yea
06:07 Senjai: At least i have a basic understanding from other languages
06:08 Lajla: So do I, I guess.
06:08 Lajla: Senjai, but basically
06:08 Lajla: to explain it
06:08 Lajla: say you have your call stac
06:08 Lajla: and you have a function foo
06:08 (join) lucian
06:08 Lajla: and it ends on return bar (x,y);
06:09 Lajla: As a random example.
06:09 Lajla: Why would you add another stack frame?
06:09 Lajla: Obviously foo will just return what bar returns.
06:09 Senjai: Yea
06:09 Lajla: So, nothing goes wrong if the stack frame of bar just raplces that of foo.
06:09 Senjai: Makes sense
06:09 Lajla: This is basically in a nutshell what happens with tail-call-optimization
06:10 Senjai: DrRacket has a memory leak.. :/
06:10 Lajla: And an added benefit is that when it ends on return foo(x,y);
06:10 Lajla: Is that it compiles into a loop, baiscally.
06:10 Senjai: Because there are no conditional statements that loop can execute without delays right?
06:10 lithpr: (define (got-milk? lst)
06:10 lithpr: (cond
06:10 lithpr: ((null? lst) #f)
06:10 lithpr: ((eq? 'milk (car lst)) #t)
06:11 lithpr: (else (got-milk? (cdr lst)))))
06:11 lithpr: 
06:11 Lajla: Senjai, well, if you have (define (recursion) (recursion)), which in C++ like code would be int recursion() {return recursion();}
06:11 lithpr: just something concrete to refer to :)
06:11 Lajla: In C++ you will get a stack overflow, depending on your compiler
06:11 Lajla: In Scheme, the thing will just enter an infinite loop
06:12 DT``: That's just how you write loops in Scheme.
06:12 Lajla: You usually end up doing that
06:12 Lajla: but there are some loop macros
06:12 Lajla: but they are just implemented on top of tail-recursion basically
06:13 Senjai: Lajla, recursion is also a better way to program, so ive heard
06:14 lithpr: iteration is often preferable to recursion because is uses less ram- that's why there is tail recursion
06:15 lithpr: so you can "think recursively" but use iteration when possible
06:16 Lajla: Senjai: http://codepad.org/N61gNczc
06:16 DT``: Senjai, if you want to read something on this, you can read the first chapter of SICP.
06:16 Lajla: THat's also a funny trick
06:16 Lajla: mutual tail recursion
06:16 Lajla: Senjai, some would say that it keeps stuff clearer
06:16 Lajla: or at least, not mutating stuff
06:16 Lajla: Maintaining 'referential transparency'
06:17 Senjai: Yea ill read the sicp chapter
06:17 Senjai: its interesting for sure
06:19 Senjai: Mutual Recursion.. eww
06:19 Lajla: Senjai, a mistake a lot of people make though is that they just end up using tail recursion to simulate loops when they shouldn't
06:19 Lajla: Senjai, why?
06:19 Lajla: Your tree function used mutual recursion
06:20 Senjai: Lajla, Mutual recursion occurs when one recursive function calls another right?
06:20 Senjai: Lajla, Mutual recursion occurs when one recursive function calls another (that also has a recursion)
06:21 Lajla: Senjai, not exactly
06:21 Senjai: create-bst did not call lokvp->bst like your codepad example
06:21 Senjai: btw, codepad seems awesome for pastes
06:21 Lajla: mutal recursion occurs when a foo calls a function bar that calls foo.
06:21 Lajla: Or actually, there can be indefinite steps between it.
06:21 Lajla: A parser I've written for some other dialect of lisp is mutually recursive.
06:22 Senjai: You wrote a parser?
06:22 Senjai: No big deal.. crazy person
06:22 Senjai: lol
06:22 DT``: Parsing Lisp is easy.
06:22 DT``: (not counting reader macros)
06:23 Senjai: For you guys it is
06:24 Lajla: Senjai, naahh, lisp is written to parse easily.
06:24 Lajla: the grammar is pretty simple as you should have noticed by now.
06:24 Lajla: No special rules for keywords and all that.
06:25 Lajla: It was so simple, no need for al that yacc or LARL or whatever stuff
06:25 Lajla: It was basically recursing through
06:25 Senjai: lol
06:25 DT``: it's like writing a brainfuck or forth parser.
06:25 Senjai: HAHA
06:30 Lajla: Well, the parsers are fairly simple.
06:31 (join) MayDaniel
06:39 (join) lucian_
06:40 Senjai: Lajla
06:40 Senjai: Or anyone: http://codepad.org/0cPB2hRh can someone help me with the error/
06:41 (quit) lucian: Ping timeout: 264 seconds
06:41 Senjai: line 36: car: expects argument of type <pair>; given empty
06:41 bremner_: Senjai: I guess you are missing a #lang line at the top
06:42 Lajla: bremner_, no, he means an error racket gave him
06:42 Senjai: bremner_: it's not that, I have it set to intermediate student
06:42 Lajla: Senjai, well, car doesn't work on emtpy lists
06:42 Lajla: car works on nonempty lists
06:42 Lajla: It's an alias of first
06:42 Lajla: you cannot get the first of an empy list
06:42 Lajla: (car (rest lob)
06:42 Lajla: That part is where it goes wrong
06:42 Senjai: Lajla: I know :) its not that, its that ball-r is attempting to get the radius of empty, thats the error
06:42 Lajla: If lob is a list of length one
06:42 Senjai: Lajla: I just dont know how to fix it
06:43 Senjai: Lajla: your right im wrong
06:43 Lajla: Senjai, well, what you want is this case is to add an extra case
06:43 Senjai: Lajla: oops, but still, How do I fix this
06:43 Lajla: if lob is of length 1
06:43 Senjai: I tried
06:43 Senjai: oh
06:43 Lajla: Basically, you have 2 base cases
06:43 Senjai: okay
06:43 Lajla: of length 0
06:43 Lajla: in this case return the empty list
06:43 Lajla: and of length 1
06:43 Lajla: Well, how do we sort a list of length 1?
06:44 Lajla: You want an extra case in your cond
06:44 Senjai: just return the one item
06:44 Senjai: Lajla: I tried, but wont it fail to cons onto the recursed list?
06:45 DT``: if (< (length lob) 2) lob
06:45 Senjai: (and (ball? (car lob)) (empty? (cdr lob))) (car lob)
06:46 DT``: Senjai, you can just change the (empty?) case in (< (length lob) 2), and return lob.
06:46 Lajla: DT``, that could be done, but it's not very efficient
06:46 DT``: Lajla, then, (or (null? lob) (null? (cdr lob))
06:47 DT``: *empty?
06:47 Lajla: Senjai, I think (cond [(empty? lob) empty] [(empty? (rest lob)) lob) [else ...])
06:47 Lajla: works okay
06:47 Lajla: DT``, yeah
06:47 Senjai: DT``: Lajla Now I have anothe rerror on (cons (car lob) (sortlob (cdr lob)))
06:47 Lajla: Because it will actually go recurse through the entire list obtaining the length
06:47 Senjai: cons: second argument must be of type <list>, given (make-ball 40 "blue") and (make-ball 50 "yellow")
06:47 Lajla: Senjai, which?
06:48 Senjai: Lajla: line 37
06:48 Senjai: throws that argument
06:48 Lajla: (cons (sortlob (cdr lob)) (car lob)))]))
06:48 Lajla: that one
06:48 Lajla: is the error I think
06:48 Lajla: You got your arguments to cons reversed there
06:48 Senjai: yes
06:48 Senjai: I have to
06:48 Lajla: Or wait
06:48 Lajla: You want to add it to the end of the list?
06:48 Senjai: yes
06:49 Lajla: (define (extend-list l . xs) (append l xs))
06:49 Senjai: (cons (car lob) (sortlob (cdr lob)))  gives the error
06:49 Lajla: Not sure if your language allows that
06:49 Lajla: Yeah, it doesn't add an element to the end in that way.
06:49 Lajla: My  extend-list procedure takes a list and some arguments and adds the argument to the end of the list.
06:49 Lajla: But I'm not sure if the language you have to use allows for that one.
06:49 Senjai: hmm
06:50 Senjai: (append) maybe?
06:50 DT``: with just (append), you'd need to wrap the second argument in (list).
06:50 Senjai: Lajla: I think its fine that way, it is saying that sortlob is giving two balls bnot a list
06:51 Lajla: Senjai, what append does is join lists together
06:51 Lajla: So if you want to extend a list, you can use (append a-list (list 1))
06:51 (join) lucian
06:51 Lajla: Adds 1 at the end of a-list
06:52 (quit) lucian_: Ping timeout: 252 seconds
06:52 Senjai: (cons (car lob) (list (sortlob (cdr lob)))) adding list fixed it
06:53 Senjai: maybe racket could not determine the type (sortlob) was returning?
06:53 Senjai: But if you wrap list for that you get a nasty error later on ball-r: expects argument of type <struct:ball>; given (list (make-ball 30 "green") (list (make-ball 40 "blue") (make-ball 50 "yellow")))
06:54 (join) tfb
06:55 Senjai: The answer key says to use an insert helper function
06:55 (join) lucian_
06:55 Senjai: Doing this requires operating on arbitrary sized data (the result of the natural recursion), so we have to  use a helper function to do it.
06:56 Senjai: Doing this requires operating on arbitrary sized data (the result of the natural recursion), so we have to  use a helper function to do it.
06:56 Senjai: Ooops: so now we have the first ball (first lob) and the result of the natural recursion. we know the result of the natural recursion is already sorted, so all we have to do is put (first lob) in the right place
06:56 (quit) lucian: Ping timeout: 246 seconds
06:57 (join) lucian
06:58 Lajla: Senjai, the last one does something different
06:58 Lajla: (list (sortlob (cdr lob)) returns a list inside a list.
06:59 Lajla: And it also adds the (car lob) element to the front.
06:59 Senjai: Yea
06:59 Senjai: Lajla: http://codepad.org/ML2D4oNM
07:00 (quit) lucian_: Ping timeout: 252 seconds
07:01 Lajla: Senjai, does it give an error?
07:03 Senjai: Lajla: that code doesn't
07:03 Senjai: It's the answer key, the insertion sorting method
07:05 Senjai: "Terminology This particular program for sorting is known as insertion sort in the programming literature."
07:06 Senjai: http://codepad.org/mufRk54p
07:06 Senjai: It eliminates that empty problem, checkes empty twice
07:11 (join) lucian_
07:12 (quit) lucian_: Remote host closed the connection
07:13 (quit) lucian: Ping timeout: 260 seconds
07:13 (quit) Senjai: Ping timeout: 245 seconds
07:16 (join) Senjai
07:16 Senjai: Wow
07:16 Senjai: Just blue screened of death
07:17 Senjai: Lajla: still here?
07:17 Senjai: Ugh.. I hate new drivers..
07:17 Lajla: Senjai, yeah
07:17 Lajla: Are you allowed to use 'inner functions'?
07:17 Senjai: How do you mean?
07:18 Senjai: like lambda?
07:19 DT``: define inside a define.
07:20 DT``: (define (fun x) (define (fun2 x y) ...) (fun2 x 1))
07:20 Lajla: Senjai, also, it seems you have your <= inverted again
07:21 Senjai: Lajla: We can use local.
07:21 Senjai: DT``: we use (local [(define ...)])
07:21 Senjai: I havent read up on it yet but we can use it
07:24 Senjai: DT``: Lajla do either of you know about cross products of types comments? or about prefix functions? http://codepad.org/lcpiVx8J (the whole lecture notes, I missed that class)
07:24 Lajla: Senjai: http://codepad.org/rytnww4p
07:27 Senjai: Lajla: that is so cool, nice and neat
07:27 Lajla: Senjai, one of the things C does not support
07:27 Lajla: inner functions
07:27 Lajla: The function isn't visible outside of it.
07:27 Lajla: stops you from breaking code or conflicting names
07:28 Senjai: Lajla: not true
07:28 Senjai: you can use static and force the function to exist in every instance the function is called
07:28 Senjai: or use {} scoping
07:28 Lajla: that's a method in C++
07:28 Senjai: static can also force variable data to be saved/updated within any instance of te function being called.
07:29 Lajla: And you still cannot define a function inside a function
07:29 Senjai: er.. yea
07:29 DT``: Lajla, you can in  GNU C.
07:29 Lajla: DT``, I read so yeah
07:29 Senjai: Lajla: yes you can :) int myfunc() { int yourfunc() { } return myfunc();}
07:29 Lajla: I wonder how that works with passing function pointers
07:30 Lajla: Senjai, not in standard C though
07:30 Lajla: But GCC supports that nonstandard extension
07:30 Senjai: Oh god, callbacks are just as bad as recursions.. lol.. Lajla Im talking about C++, I'm not fluent in C
07:31 Senjai: Then there is good ol goto :)
07:43 Senjai: Hey guys, Im gonna head in, get a few hours sleep, Lajla gonna be around later?
07:43 Lajla: Senjai, don't know
07:44 Lajla: But I'm usually there
07:44 Senjai: How often are you usually on/
07:44 Lajla: as I am sitting comfortably on a couple of hours of sleeps everty few days
07:44 Lajla: Often
07:44 Senjai: Cool, do you want to exchange emails by any chance?
07:44 lithpr: LORD HAVE MERCY I'M FINALLY THROUGH THE FOURTH CHAPTER OF THE RACKET GUIDE
07:45 Senjai: lithpr, are you doing the exercises?
07:45 lithpr: some
07:45 lithpr: i'm doing a lot of fidling wiht the code though, trying to understand
07:46 Senjai: Lajla: before I go, check this out, this is my profs version of what you did http://codepad.org/U9jWFLfa
07:46 Lajla: Senjai, I'd rather not give emails if you don't mind.
07:47 Lajla: Senjai, you mean losing 'local'?
07:47 Senjai: Lajla: no worries, thats fine
07:47 Senjai: You just defined the function in the function, my prof used local
07:47 Senjai: is there a difference?
07:47 cpach: are there exercises in the racket guide?
07:48 Lajla: I'm not sure how local works. =P
07:48 DT``: Senjai, local is a syntactic sugar for let and letrec.
07:48 Lajla: As I said, I don't use racket, as in, the language, that must.
07:48 DT``: inner define is a syntactic sugar for let and letrec.
07:48 Lajla: or letrec is for inner definitions
07:48 Lajla: No one knows
07:49 Lajla: But (letrec ((name value) ...) body) is identical to (let () (define name value) ... body) I believe
07:50 DT``: (let ((name #<undefined) ...) (set! name value) ... body))
07:50 DT``: *#<undefined>
07:50 Lajla: That too.
07:51 Lajla: I once implemented letrec with a horrendously convoluted variadic fixedpoint operator.
07:51 Lajla: It took me like 2 days to construct that thing.
07:51 DT``: variadic fixedpoint operator?
07:51 Lajla: It needed to converge even in a call-by-value setting.
07:51 Lajla: Yeah
07:51 DT``: is that something like loeb?
07:51 Lajla: What is loeb?
07:52 DT``: I don't know, really.
07:52 DT``: wait.
07:53 Lajla: It was an operator such that (fix f x ...) evaluated to (f (fix (f x) ...)) I believe
07:55 DT``: I don't get the ... part.
07:55 Senjai: ... means other code there..
07:55 Senjai: like etc.
07:55 DT``: What would ((fix f g h) x) do?
08:05 Lajla: It would evaluate to ((f (fix f (f g) (f h))) x) if I recall correctly.
08:08 (join) tauntaun
08:09 lithpr: (struct posn (x y))
08:09 lithpr: stdin::4765: struct: illegal use of signature form in: (struct posn (x y))
08:10 lithpr: that's copy-pasta from the guide
08:10 DT``: > (struct posn (x y))
08:10 DT``: >
08:10 DT``: what #lang are you using?
08:10 DT``: *which
08:11 lithpr: well, i'm using the racket repl in emacs, it throws an error when i try to set a language with #
08:12 lithpr: #lang racket
08:12 lithpr: stdin::4822: read: #lang not enabled in the current context
08:13 lithpr: > reference to undefined identifier: racket
08:13 lithpr: i actually asked about this last night, but no one replied
08:14 DT``: the can't-define-structs-from-emacs problem is a known problem (I read it somewhere, but I don't remember now).
08:14 lithpr: okay
08:14 DT``: you can't set a language from the repl.
08:15 DT``: you have to invoke racket with ``racket -il language''
08:15 DT``: > racket -ilq typed/racket
08:15 DT``: Welcome to Racket v5.1.
08:15 DT``: > 2
08:15 DT``: - : Positive-Fixnum
08:15 DT``: 2
08:16 lithpr: okay, i'm reading racket --help
08:16 lithpr: thank you
08:19 (join) corruptmemory
08:27 tauntaun: lithpr: my workaround is to use define-struct instead, even though it's deprecated.
08:27 tauntaun: (I think)
08:28 (join) mceier
08:29 (join) masm
08:43 (quit) Senjai: Ping timeout: 245 seconds
08:43 lithpr: ahhh, racket -i
08:49 lithpr: no, that wasn't it- i was using mzscheme
08:49 lithpr: all good now :)
08:51 (quit) MayDaniel: Read error: Connection reset by peer
08:59 (quit) mario-goulart: Read error: Operation timed out
09:00 (join) mario-goulart
09:14 (join) MayDaniel
09:14 (quit) mceier: Quit: leaving
09:45 (join) dnolen
10:54 (join) mceier
10:55 (quit) tauntaun: Quit: Ex-Chat
10:55 (join) tauntaun
10:57 (join) mheld
11:05 (join) mheld__
11:07 (quit) mheld: Ping timeout: 240 seconds
11:07 (nick) mheld__ -> mheld
11:27 (quit) dnolen: Quit: dnolen
11:37 (join) jonrafkind
11:45 (quit) mheld: Ping timeout: 260 seconds
12:15 (join) anRch
12:24 (quit) amoe_: Quit: leaving
12:38 (quit) anRch: Quit: anRch
12:42 (join) mheld
12:51 (quit) lithpr: Remote host closed the connection
12:51 (join) anRch
13:14 (quit) corruptmemory: Ping timeout: 255 seconds
13:15 (join) lithpr
13:19 (nick) Lajla -> GreenPride
13:25 (quit) coyotama|2: Read error: Connection reset by peer
13:28 (join) coyo
13:28 (quit) coyo: Changing host
13:28 (join) coyo
13:50 (quit) tauntaun: Ping timeout: 240 seconds
14:01 (quit) MayDaniel: Read error: Connection reset by peer
14:03 (join) tauntaun
14:11 (quit) tauntaun: Quit: Ex-Chat
14:18 (quit) tfb: Ping timeout: 260 seconds
14:28 (join) MayDaniel
14:35 (quit) anRch: Quit: anRch
14:41 (join) mheld__
14:44 (quit) mheld: Ping timeout: 255 seconds
14:44 (nick) mheld__ -> mheld
14:44 (nick) rapacity_ -> rapacity
14:51 (join) RyanRN
14:54 (join) lucian
15:26 (quit) RyanRN: Ping timeout: 264 seconds
15:46 (join) lucian_
15:48 (quit) lucian: Ping timeout: 250 seconds
15:59 (nick) lucian_ -> lucian
16:24 (quit) lucian: Read error: Operation timed out
16:31 (join) lucian
16:31 (quit) em: Ping timeout: 240 seconds
16:33 (join) em
16:34 (join) lucian_
16:36 (quit) lucian: Ping timeout: 260 seconds
16:38 (nick) lucian_ -> lucian
16:44 (join) zimio
16:45 zimio: heeeey
16:45 zimio: does R5S5 has for loops?
16:48 (quit) lithpr: Remote host closed the connection
16:50 (join) Nanakhiel
16:51 (join) lithpr
16:51 (join) DT-sama
16:52 (join) Demosthenex
16:52 Nanakhiel: zimio, define 'for loop'
16:52 Nanakhiel: It has a do ""loop""
16:52 (quit) Demosthenes: Ping timeout: 250 seconds
16:52 (quit) GreenPride: Ping timeout: 250 seconds
16:52 (quit) DT``: Ping timeout: 250 seconds
16:52 (join) tauntaun
16:52 (quit) mattmight: Ping timeout: 250 seconds
16:52 zimio: oh thanks
16:53 (join) mattmight
16:53 Nanakhiel: http://pauillac.inria.fr/cdrom_a_graver/www/bigloo/manual/r5rs-5.html
16:53 Nanakhiel: look for 'do is an iteration construct'
16:57 (join) misterm
17:01 (join) RyanRN
17:01 (quit) lucian: Ping timeout: 250 seconds
17:10 (quit) MayDaniel: 
17:10 (join) MayDaniel
17:12 (quit) RyanRN: Read error: Connection reset by peer
17:13 (join) RyanRN
17:23 (quit) lithpr: Read error: Connection reset by peer
17:25 (join) lithpr
17:25 (quit) RyanRN: Ping timeout: 276 seconds
17:26 (join) RyanRN
17:29 (quit) RyanRN: Client Quit
17:29 (join) lucian
17:35 (join) RyanRN
17:35 (quit) RyanRN: Client Quit
17:49 (part) shofetim: "ERC Version 5.3 (IRC client for Emacs)"
17:51 (quit) zimio: Read error: Connection reset by peer
17:52 (quit) masm: Quit: Leaving.
18:11 lithpr: is it possible to distribute a program developed in racket to someone who doesn't have racket installed?
18:11 DT-sama: raco exe
18:12 lithpr: ty
18:19 (join) corruptmemory
18:36 (quit) lucian: Remote host closed the connection
18:55 (quit) MayDaniel: Read error: Connection reset by peer
19:04 (join) jim__
19:05 (quit) corruptmemory: Ping timeout: 246 seconds
19:06 (join) common_tragedian
19:16 (join) dnolen
19:17 (join) mheld__
19:19 (quit) mheld: Ping timeout: 255 seconds
19:19 (nick) mheld__ -> mheld
19:50 (join) evhan
20:05 (quit) dnolen: Quit: dnolen
20:14 (quit) mye^: Ping timeout: 255 seconds
20:15 (quit) tauntaun: Quit: Ex-Chat
20:25 (join) Senjai
20:26 Senjai: Hey all
20:38 (quit) mceier: Quit: leaving
20:54 Senjai: Ayone around?
20:55 jonrafkind: do you have a question?
20:55 Senjai: I will :P but I'm also just gauging whos around.
20:56 Senjai: I'm studying racket for a comp sci class, I'm sure as I work through the book Ill have questions :P BST's last night were a pain..
20:58 (join) coyotama|2
21:01 (quit) coyo: Ping timeout: 276 seconds
21:01 (join) tauntaun
21:15 (quit) tauntaun: Quit: Ex-Chat
21:18 (nick) zakwilson_ -> zakwilson
21:26 (quit) evhan: Quit: leaving
21:27 (join) evhan
21:29 (quit) evhan: Client Quit
21:36 (join) evhan
21:48 (quit) jonrafkind: Ping timeout: 255 seconds
21:57 (quit) common_tragedian: Remote host closed the connection
22:24 (join) tauntaun
22:31 (join) common_tragedian
22:36 (quit) common_tragedian: Remote host closed the connection
23:05 (quit) tauntaun: Quit: Ex-Chat
23:07 (quit) jim__: Quit: Leaving
23:07 (join) corruptmemory
23:31 (quit) mheld: Quit: mheld
23:34 (join) tauntaun
23:35 tauntaun: I'm trying to understand better how the word "reflection" is used in the context of other programming languages.  Does Scheme's 'apply' endow the language with a capacity for reflection?  Or would that be another word?
23:44 Senjai: tauntaun: sorry dude, the channel can be a bit dry sometiems, I don''t know thee answer but wait a bit someone might come by