00:00:42 elliott: Shoot ... I don't think I can make VAX work :( 00:00:46 Gregor: :( 00:00:47 :( 00:01:43 Butbut all the world's a VAX! 00:02:14 It's CISC but (apparently?) functions have to be 16-bit aligned! 00:02:42 data ExitVal = To DirVal | From DirVal; data DirVal = NW | NE | SW | SE; class Exit e where exitVal :: p e -> ExitVal; instance Dir d => Exit (To d) where exitVal x = To (dirVal x); ... class Dir d where dirVal :: p d => DirVal; instance Dir NW where dirVal _ = NW; ... 00:02:49 I can't really cope with function alignment not being the same as the smallest opcode. 00:02:52 Gregor: Just align everything? 00:02:58 i think something like that is needed to actually _test_ these directions 00:03:03 Gregor: It shouldn't hurt to align everything to n * size of smallest opcode for smoe n 00:03:04 some 00:03:42 hm why did i write p, i was going to use t... 00:03:55 data ExitVal = To DirVal | From DirVal; data DirVal = NW | NE | SW | SE; 00:03:58 i already have those 00:04:08 ah 00:04:09 without the reification stuff, though 00:04:11 but i can add that 00:04:36 and also a bunch of the datatypes need Dir dir => contexts for their quantified types 00:04:51 otherwise they might be type erased entirely 00:04:56 oerjan: this is so you can "ascend SW" or whatever right? 00:05:59 Gregor: The crap? 00:06:09 no, this was primarily so i can look at a gadt value and actually find out which dirs it _uses_ inside 00:07:10 oerjan: ah. 00:07:14 * Gregor reappears. 00:07:15 -!- augur has joined. 00:07:17 oerjan: mightn't a type family fit better? perhaps not 00:07:20 elliott: It's not the codegen that's the problem. 00:07:25 elliott: It's pro/epilogue detection. 00:07:32 Gregor: Alright 00:07:38 i dunno, if you see how... 00:07:46 i'm not so familiar with them. 00:07:50 Gregor: The crap? // the crap what? 00:08:50 -!- CakeProphet has joined. 00:08:50 -!- CakeProphet has quit (Changing host). 00:08:50 -!- CakeProphet has joined. 00:09:39 16-bit alignment for functions? 00:09:40 -!- Slereah has joined. 00:09:41 *16* bit? 00:10:39 Ayup :P 00:10:42 -!- Slereah_ has quit (Ping timeout: 244 seconds). 00:12:22 Still, even without VAX, I feel that x86[_64], SPARC, MIPS[el], ARM, PPC, Alpha is a pretty good list. 00:26:45 -!- CakeProphet has quit (Ping timeout: 258 seconds). 00:31:56 -!- augur has quit (Remote host closed the connection). 00:44:54 oerjan: go on... :P 00:49:12 -!- hagb4rd has quit (Ping timeout: 240 seconds). 00:51:33 -!- tiffany has quit (Quit: nyu~). 00:57:24 oerjan: :( 00:58:34 the problem is, i'm actually fed up of this already 00:58:45 (in case that wasn't obvious) 00:58:48 * elliott creys 01:01:48 oerjan shall forever be known as the person who sided with the finns and swedes 01:02:14 wut 01:02:21 deewiant and vorpal 01:02:34 JUST SAYING 01:05:21 -!- pikhq_ has joined. 01:07:38 -!- pikhq has quit (Ping timeout: 252 seconds). 01:11:01 oerjan has now committed seppuku. 01:11:16 brains... 01:11:24 hi 01:11:27 -!- Sgeo|web has joined. 01:11:30 hi brains 01:11:49 Hi 01:11:52 oerjan: i will now pester monqy for the algorithm instead to make you feel bad. 01:11:58 monqy: hi algorithm 01:12:02 what algorithm :( 01:12:03 hi 01:12:04 fancy 01:12:36 Ok, two different people in two different chatrooms greeted with pluralization... 01:12:50 Although I assume "brains" is just a part of context of prior conversation 01:13:08 monqy: east algorithm 01:13:16 oerjan is too scandinavian to tell me the secrets 01:13:18 east algorithm 01:13:50 what is east algorithm what are you talking about ;_; ;_;; _;;;;; 01:14:06 iv'eb eeeen too busy python to pay attenteion :( 01:14:59 well basically you need to go up in a quadtree, then in the right direction to end up east of where you were. the trouble is all the extra scaffolding of data which makes even those steps insanely tricky 01:15:55 oerjan: i'm going to give it a shot myself, and if i fail i will die HOPE YOU FEEL GOOD 01:17:26 yay 01:17:31 i mean, aww 01:17:51 oerjan: um out of curiosity is the "p" used in exitVal/dirVal just standing in for Proxy or similar? 01:17:58 or do you actually plan to apply them to MetaNodes or something 01:18:02 Sgeo|web: BRAINS... 01:18:49 elliott: as i said, i really wanted it to be t, similar to in your Neighbors* types 01:19:06 right 01:19:07 elliott: and yes, the idea was to just be able to put in any of those 01:20:27 oerjan: otoh it fails slightly because you have "p (To NW)" and you apply dirVal to it but dirVal is expecting "p NW"... 01:20:42 oh hm 01:21:10 oerjan: it's easy enough to make a proxy out of it though 01:21:22 a (b c) -> Proxy c :p 01:21:27 yeah. whatever you think easiest. 01:21:28 or just use scoped type variables 01:21:50 erm, scoped type variables are tricky with gadts 01:22:00 it seemed to me 01:22:19 it doesn't capture those that appear from pattern matching 01:22:25 er I meant 01:22:28 instance (DirectionType dir) => ExitType (To dir) where 01:22:28 exitVal = To $ directionVal (Proxy :: Proxy dir) 01:22:33 not replacing the whole mechanism with them 01:22:43 ah 01:22:48 it avoids a silly "a (b c) -> Proxy c" function to make the types work out 01:25:09 time to write "ascend :: MetaContext -> MetaContext"... 01:25:29 -!- augur has joined. 01:26:27 to find out how that changes the structure, it is necessary to know how many of the neighbors of the source and destination node lead to other real cursors 01:27:03 ugh. 01:27:24 oerjan: btw, I'm not sure a class actually /is/ needed, because everything ends up containing Neighbours, and you can pattern-match on those to determine which direction is being used >:) 01:27:39 in other words: /me adds class contexts to everything 01:27:48 elliott: not leafs, i thought... 01:27:54 well probably not :P 01:29:19 because if you are leaving a node with > 2 such neighbors, it gets turned into a MetaNode, while if it has 1, it gets merged into that one's MetaPath, while if it has 0, it and its neighbors get turned into a PlainCursor. 01:29:32 er, * >= 2 01:29:53 oh, excluding neighbors in the direction you are going, of course 01:31:23 ah and of course if there is another real cursor on the node _itself_, it always becomes a MetaNode. 01:31:45 * elliott is getting very suspicious of these very-similar data types that seem to give rise to a bunch of special cases. 01:31:58 heh 01:32:36 a MetaNode either contains a real cursor or is sort of a crossing whose signPost tells you how to get to other real cursors 01:33:45 i am also kind of suspicious of how the "zipper of tree of zippers" thing is sort of tied to the zippers themselves 01:33:56 hm? 01:34:14 well, it seems like you should be able to do that completely separately from the zipper operations themselves. 01:35:18 oerjan: also, I think SignPost is broken. there's nowhere to stash the IpData when switching cursor 01:35:51 wasn't it Either IpData Exit? 01:36:06 yes, but presumably you have to change it to the relevant Exit when switching cursors... 01:36:13 since you use the SignPost to figure out whether you're at a given cursor or not 01:36:16 and if not, how to get to it 01:37:32 well when you leave a node carrying a cursor, you change that node's signPost from containing the ipData to containing the exit, of course; at the same time you change the node you're going to to contain the ipData 01:37:55 oh 01:38:00 that makes sense :P 01:38:57 it may be that the entire SignPost gets moved to the new node, if the old one is packed into a MetaPath while the new one became unpacked from one 01:39:08 it seems like the leaf data should contain the IpId 01:40:19 well... i guess cursors that aren't in the move will always be on leaves 01:40:21 -!- pikhq has joined. 01:40:44 oerjan: well otherwise I have to scan the entire signpost just to find the value that's Left 01:41:13 erm, you just lookup your current ip, surely 01:41:32 there can be more than one ip at the same spot, too 01:41:36 oerjan: how 01:41:39 so more than one Left 01:41:41 LeafCtx :: (Direction dir) 01:41:41 => LeafData 01:41:41 -> MetaCursor (From dir) 01:41:41 -> SignPost 01:41:41 -> MetaContext 01:42:14 the SignPost can contain more than one ip key whose value is Left ... 01:42:16 -!- pikhq_ has quit (Ping timeout: 244 seconds). 01:42:26 corresponding to two ips in the same cell 01:42:30 oerjan: well, okay. I was going by well when you leave a node carrying a cursor, you change that node's signPost from containing the ipData to containing the exit, of course; at the same time you change the node you're going to to contain the ipData 01:42:51 oh, only for the ip key you are actually _moving_ 01:43:24 /cursor 01:43:53 oerjan: right. and how am i suppose to determine which one that is? 01:44:08 oh right, i forgot about that... 01:44:13 lol 01:44:18 you should take the ip as an argument to ascend 01:44:28 ok 01:46:28 * elliott idly wonders about a data structure that acts as a set of bytestrings; given a hash of one of the bytestrings, you can get the bytestring out, and given a hash that matches none of the bytestrings in the set, you can determine that it isn't present; but (a) it is unfeasible (brute-forcey) to get a valid hash out of the data without knowing one, and (b) it is unfeasible to access any bytestring without knowing its hash 01:46:47 I was thinking you could encrypt each bytestring with its hash as the key, but you need a way to go from the hash to that data, and that ends up storing the hash itself... 01:47:01 ooh: you could implement it as a lookup table from a hash of the hash to the bytestring encrypted with the hash 01:47:11 then doing (a) or (b) requires reversing the hash function 01:48:09 i'm sure this must have been investigated 01:48:33 probably, but i have no idea what i would google for 01:49:19 -!- derdon has quit (Remote host closed the connection). 01:49:32 -!- CakeProphet has joined. 01:49:32 -!- CakeProphet has quit (Changing host). 01:49:32 -!- CakeProphet has joined. 01:49:38 hi. 01:50:00 okay so when I convert the integer 5 to a float via c-style casting in C++ 01:50:08 that is 5.0 correct? 01:50:14 or is it something weird? 01:50:20 elliott: hm, all the difficulty in implementing that is in implementing one-point sets, i think 01:50:27 then you can just make a list of them 01:50:38 -!- copumpkin has changed nick to luke-sr. 01:50:53 assuming you don't consider iterating through all too expensive 01:50:54 -!- luke-sr has changed nick to copumpkin. 01:52:20 oerjan: that is indeed too expensive. 01:52:29 CakeProphet: yes 01:52:46 oh hm of course you could simply use the hash to look the right one up in a hashtable 01:53:21 oh hm i didn't read (a) properly 01:53:34 oerjan: you could do that and not store the keys 01:53:45 so it's O(n) in how many hashtable collisions there are to look up 01:53:54 oh wait 01:53:58 yeah that works 01:53:59 um wouldn't (a) make it impossible to _construct_ your data structure? 01:54:12 elliott: ah okay, see I thought differing representations of would mess things up, but... I guess not? 01:54:13 from the data you want to put in it 01:54:22 *of floats and ints 01:54:32 oerjan: well, with my proposed solution: table[hash(hash(data))] = encrypt(data, key=hash(data)) 01:55:52 or is it that c-style case in C++ does nice automatic conversion, with reinterpret_cast being an actual C-like cast? 01:55:56 help I am C nub. 01:56:02 s/case/cst/ 01:56:05 wehriuheriuhwUEHR UW EHRU HWUEIRH YU23H 78H 01:57:42 -!- pikhq has quit (Read error: Connection reset by peer). 01:57:58 -!- pikhq has joined. 01:58:42 it's times like this that I wish my data structures course was in Haskell. :P 01:59:33 > 5 `div` 2 01:59:34 2 01:59:36 > 5 / 2 01:59:37 2.5 02:01:22 elliott: hey I'm making a hash table too!!! :D :D :D :# 02:01:32 oerjan: does that not work? 02:01:46 @hoogle ($!!) 02:01:46 Control.DeepSeq ($!!) :: NFData a => (a -> b) -> a -> b 02:01:54 -!- copumpkin has quit (Quit: Computer has gone to sleep.). 02:02:24 elliott: oh, i don't trust myself to judge that 02:02:40 oerjan: heh 02:02:56 CakeProphet: Would you learn purely functional queues, or do structures that could only be used in ST? 02:03:12 Sgeo|web: good question. depends on the structure. 02:03:24 hash table would be IO/ST, trees would be pure. 02:03:58 I don't think you could cram Haskell + data structures into one course though. not effectively anyway 02:04:10 you'd want an intro Haskell course followed by a data structures course. 02:05:56 well, not all trees would be pure I guess. but then... hmmm 02:07:00 well really the data structures you'd want to use in Haskell are not the ones you'd want to use in C++, I think. 02:07:13 can you do tree rotations in pure Haskell efficiently? 02:12:07 O(log n) 02:12:54 to insert/delete and rebalance 02:18:43 Whose Line is it Anyway season 1 report: Absolute winner is Wayne Brady with an impressive 3,005,530.5 points. Absolute average episode winner is also Wayne with 187,845.7 points per episode. Logarithmic winner is Colin Mochrie with 130.4. Logarithmic average episode winner for performers appearing in >1 episode is also Colin with a commanding 6.5 log-points per episode. 02:19:19 ah but what about the sinusoidal winner 02:20:58 oerjan: ah okay. 02:23:35 Ryan Stiles was the declared winner in 12 episodes, with Colin coming in second at 9 episodes. 02:24:40 Gregor: ....WHY 02:24:42 STOP 02:24:51 Gregor: go obsess about something else. 02:25:03 * oerjan swats CakeProphet -----### 02:25:16 there are better things to invest time into 02:25:27 CakeProphet: My JIT works on nine platforms! 02:25:38 Gregor: uh, okay? 02:25:50 IT COULD WORK ON ALL THE PLATFORMS 02:26:04 CakeProphet: I taught my dog to teach my cat to behave! 02:26:08 "shachaf: Sometimes things are complicated because the domain is complicated. Other times things are complicated because edwardk." 02:26:20 (new Haskell Weekly News) 02:26:21 oerjan: ? 02:26:25 Oh. 02:27:13 Ugh, I got quoted twice. 02:27:47 hahahahaaaaaa 02:27:48 elliott: why is your quote so serious. 02:28:06 i have half a mind to `delquote it 02:28:15 so for my data structures project I have to provide pseudocode algorithms in my documentation for any function that's more than 15 lines. 02:28:24 but none of my functions are more than 15 lines. bahahahahahaha 02:28:46 CakeProphet: EXCELLENT 02:28:57 oerjan: sory 02:29:26 I can't imagine any of these functions being more than 15 lines 02:29:34 unless you're just bad at programming or bad at abstraction. 02:30:26 -!- yours_truly has joined. 02:31:07 -!- Nisstyre has quit (Read error: Connection reset by peer). 02:32:30 `quote shachaf 02:32:33 620) elliott: GHC bug? Come on, it's the parentheses. The more parentheses you add, the closer it is to LISP, and therefore the more dynamically-typed. \ 666) Real Tar is GNU tar. You just ignore whichever features don't make you feel superior enough. \ 704) VMS Mosaic? I hope that's not Mosaic ported to VMS. Hmm. It's Mosaic ported to VMS. 02:32:46 Those quotes are all horrible. 02:33:07 You should get kmc in this channel. kmc has good quotes. 02:33:09 `quote kmc 02:33:10 c-style cast, dynamic cast, static cast, reinterpret cast, const_cast. 02:33:11 686) COCKS [...] truly cocks 02:33:21 Well, in theory. 02:34:10 oh I see, C just does automatic conversion of number types. 02:34:28 `addquote You should get kmc in this channel. kmc has good quotes. `quote kmc 686) COCKS [...] truly cocks Well, in theory. 02:34:30 720) You should get kmc in this channel. kmc has good quotes. `quote kmc 686) COCKS [...] truly cocks Well, in theory. 02:34:33 I thought it was a runtime NOP 02:34:36 And now, you have good quotes :P 02:35:05 :-( 02:37:04 What is with Haskell and people named Elliott? 02:37:08 shachaf: we have had kmc in this channel, he refuses to entertain us though 02:37:27 Sgeo|web: what 02:37:48 elliott: There's you, Conal, Some dude named Trevor, some guy I met IRL 02:37:49 Sgeo|web: Elliott is the name of elitist jerks who like elitist jerk languages. hurr hurr hurr 02:37:56 elliott: Sgeo|web is trying to say "What is it with elliott?" in a polite way. 02:38:15 Sgeo|web: there's also elliottcable! 02:38:20 who i doubt actually knows any haskell. 02:38:30 but is in #haskell occasionally and ragequits whenever people start pinging me 02:38:45 also Elliott Smith who is actally not a programmer and is dead. 02:38:51 Gregor: hm that 720 quote has some parsing ambiguity due to the recursion 02:38:59 elliott: Right now neither you nor he are in #haskell. 02:39:01 -!- augur has quit (Remote host closed the connection). 02:39:11 -!- CakeProphet has changed nick to Haskell. 02:39:36 hi guys. 02:39:41 this nick was registered in 2008 apparently. 02:39:55 Haskell's last name is probably Elliott 02:40:00 I should go to #freenode and get it. 02:40:19 oh, and Kallisti 02:41:06 because that's a name I like and stuff. 02:42:05 -!- Haskell has changed nick to CakeProphet. 02:42:24 I can only be Haskell for so long. The awesome blows my mind. 02:44:00 is doubling a hash table a reasonable resize function? I'm gonna go with "sure" 02:44:33 CakeProphet: Better to enlarge it by one byte. 02:44:40 More space-efficient that way. 02:44:58 yes indeed. 02:45:02 good call. 02:45:38 I was actually worried more about time than space. 02:45:41 imagine that. 02:45:55 -!- copumpkin has joined. 02:46:06 Gregor: hm that 720 quote has some parsing ambiguity due to the recursion // 'struth 02:46:31 `quote 1 02:46:33 1) I used computational linguistics to kill her. 02:46:55 `quote 2 02:46:57 2) EgoBot just opened a chat session with me to say "bork bork bork" 02:47:16 heh heh heh 02:47:46 honestly I'm not sure why you would expect great works of comedic genius from an #esoteric quote db 02:49:24 wait, how could you _not_? 02:50:02 oerjan: ha. ha. ha ha. ha ha ha. ha ha ha ha ha. 02:52:43 -!- yours_truly has changed nick to Nisstyre. 02:53:14 > intercalate(".")$fix((" ha":).scanl(++)(" ha")) 02:53:16 " ha. ha. ha ha. ha ha ha. ha ha ha ha ha. ha ha ha ha ha ha ha ha. ha ha h... 02:53:37 the Fibonacci laugh. 02:53:53 only expressed for the most Fibonacci humorous things. 02:54:35 oerjan: also how does that even work. 02:56:41 > fix((1:).scanl(+)1) 02:56:43 [1,1,2,3,5,8,13,21,34,55,89,144,233,377,610,987,1597,2584,4181,6765,10946,1... 02:56:50 > scanl (++) (" ha") "lo" 02:56:51 Couldn't match expected type `[GHC.Types.Char]' 02:56:51 against inferred ty... 02:57:02 i just modified that one in the obvious way 02:57:14 right I forget what scan does. 02:57:25 > scanl (++) (" ha") ["lo"] 02:57:27 [" ha"," halo"] 02:57:39 > scanl(+)[1..] 02:57:40 Overlapping instances for GHC.Show.Show ([[t]] -> [[t]]) 02:57:40 arising from a ... 02:57:42 oops 02:57:46 > scanl(+)0[1..] 02:57:46 [0,1,3,6,10,15,21,28,36,45,55,66,78,91,105,120,136,153,171,190,210,231,253,... 02:58:46 ah okay. 02:58:55 the usual laziness black magic. 02:59:14 > scanl(f)x[1..]::Expr 02:59:15 Couldn't match expected type `SimpleReflect.Expr' 02:59:15 against inferred ... 02:59:18 oops 02:59:26 yeah I got sccanl 02:59:28 -c 02:59:36 :t scanl 02:59:38 forall a b. (a -> b -> a) -> a -> [b] -> [a] 03:00:06 scanl is like foldl, but lists all the intermediate results 03:00:33 > scanl(f)x[1..]::[Expr] 03:00:34 [x,f x 1,f (f x 1) 2,f (f (f x 1) 2) 3,f (f (f (f x 1) 2) 3) 4,f (f (f (f (... 03:01:05 > scanr(f)x[1..] 03:01:06 [f 1 (f 2 (f 3 (f 4 (f 5 (f 6 (f 7 (f 8 (f 9 (f 10 (f 11 (f 12 (f 13 (f 14 ... 03:01:25 scanr gets no love. 03:01:52 > scanr(f)x[1..5] 03:01:53 [f 1 (f 2 (f 3 (f 4 (f 5 x)))),f 2 (f 3 (f 4 (f 5 x))),f 3 (f 4 (f 5 x)),f ... 03:01:56 > scanr(f)x[1..3] 03:01:57 [f 1 (f 2 (f 3 x)),f 2 (f 3 x),f 3 x,x] 03:02:33 scanl adds calls, scanr removes calls. 03:02:43 s/calls/applications/ 03:03:07 @src scanr 03:03:08 scanr _ q0 [] = [q0] 03:03:08 scanr f q0 (x:xs) = f x q : qs 03:03:08 where qs@(q:_) = scanr f q0 xs 03:03:30 @src scanl 03:03:31 scanl f q ls = q : case ls of 03:03:31 [] -> [] 03:03:31 x:xs -> scanl f (f q x) xs 03:03:55 weirdly different in style. 03:04:15 CakeProphet: only for laziness 03:04:19 > scan f x undefined 03:04:20 Not in scope: `scan' 03:04:23 well scanl can give the first element without any recursion 03:04:25 > scanl f x undefined 03:04:26 [x*Exception: Prelude.undefined 03:06:02 > reverse (scanl (+) 1 [1..5]) == scanr (+) 1 [1..5] 03:06:03 False 03:06:13 ah 03:06:16 the initial value. 03:06:29 > reverse (scanl1 (+) [1..5]) == scanr1 (+) [1..5] 03:06:30 False 03:06:34 yeah. 03:06:40 -!- augur has joined. 03:06:57 > reverse (scanl1 (+) [1..5]) == scanr1 (+) (reverse [1..5]) 03:06:58 True 03:08:42 ah 03:19:27 oerjan: http://upload.wikimedia.org/wikipedia/commons/6/65/Dannevirke_welcome_sign.jpg 03:19:30 is this what you look like? 03:20:34 hmm, my double-hashing solution is so ugly :( 03:20:41 of course not, vikings didn't have horns on their helmets! 03:20:45 elliott: double-hasing is gross. 03:21:04 elliott: SIMPLY USE SOME FORM OF NON-LINEAR PROBING. 03:21:13 also i'm clean shaven. 03:21:17 CakeProphet: i don't think you know what i'm talking about 03:21:23 oerjan: no, you look like dijkstra 03:21:34 elliott: double hashing as in... double hasing as in... hash tables. right? 03:21:39 what else could you be talking about? 03:22:40 In een alternatief universum, Ørjan lijkt Dijkstra. 03:23:08 bork bork bork 03:23:33 Ørjan 03:23:36 man, I want to be called that 03:23:41 i have a hunch the lijkt isn't quite correct 03:24:24 copumpkin: I'm famous. 03:24:30 oh? 03:24:31 therefore my name is awesome. (?) 03:24:43 famous for prophesizing cakes? 03:24:52 http://en.wikipedia.org/wiki/Adam_Curtis 03:24:54 I'm this guy. 03:25:08 oh, excellent 03:25:14 * elliott likes to think CakeProphet really is Adam Curtis. 03:25:27 elliott: you're in luck because I am. 03:25:34 lots of fucking brits in here 03:25:41 (I am too) 03:25:42 * CakeProphet is American. 03:25:51 CakeProphet: it says you're from the UK right on your wikipedia page. 03:25:55 quit foolin' around. 03:25:56 then you should correct your wikipedia page 03:25:59 itym hexhamers 03:26:08 elliott: [citation needed] 03:26:08 oh wait 03:26:17 copumpkin: you're not from hexham are you 03:26:21 nope 03:26:31 I'm from london! 03:26:36 oerjan: btw here's what pseqs look like now: http://hpaste.org/54147 03:27:06 elliott: so am I like the Michael Moore of the UK? 03:27:18 or like... not quite as controversial? 03:27:42 i don't think the UK, as a country, could actually support a figure describable as "the michael moore of the uk". 03:27:49 i don't think that's even mathematically possible 03:28:18 well, give me an analogy 03:28:58 i don't think i can analogise to a world so alien ;D 03:28:59 i don't think i can analogise to a world so alien ;D 03:29:00 oops 03:29:17 yes I find it difficult to wrap my head around the UK. 03:29:23 mainly because it's a large mass of land. 03:29:35 Adam Curtis seems more respectable perhaps? 03:29:38 or respected 03:33:05 "Suggested a parallel between the rise of Islamism in the Arab world and Neoconservatism in the United States in that both needed to inflate a myth of a dangerous enemy in order to draw people to support them." 03:33:09 tee hee hee 03:33:23 (that's the description for The Power of Nightmares) 03:33:28 (on Wikipedia) 03:33:57 _so_ easy to godwin that 03:35:42 elliott: apparently he is the /only/ famous Adam Curtis. 03:36:13 CakeProphet: s/he is/I am/ 03:36:32 note that I'm not actually Adam Curtis. 03:36:35 but I am Adam Curtis. 03:36:40 elliott: I'm sorry to disappoint you. 03:36:49 wrong 03:36:58 completely wrong 03:37:04 you're actually gleeful 03:38:58 elliott: I occasionally get weird reactions to my name. 03:39:13 not because it's the same as some kind of famous guy in the UK (no one in the US knows who that guy is) 03:39:22 but because it's apparently two first names or something. 03:40:55 i doubt adam curtis is that well known in the UK either :P 03:41:53 elliott: oh I recall from a previous conversation you saying that he was kind of well-known by people that know things. 03:42:07 or something similar 03:42:24 i guess among people who know things about documentaries? 03:42:56 do you know things about documentaries? 03:44:09 okay well this is good. this means I can go to the UK. 03:44:48 if my name were, say, Tony Blair I'd be more reluctant to go to the UK. 03:50:08 elliott: so let me get this straight 03:50:16 there are more than two major parties in the UK? 03:50:21 Better than "WTF how do you *say* that" as a reaction. 03:50:21 * CakeProphet mind blown. 03:50:33 CakeProphet: three. 03:50:39 it is smaller than the other two. 03:50:39 elliott: yes 03:50:56 and will probably never get elected again after this government 03:51:02 elliott: it's almost like you can track the path I took on Wikipedia! 03:51:30 er wait you mean Conservative Party? I thought it was like... "the big one" 03:51:33 Seriously, how many Americans do you expect to say "Worcester" even vaguely reasonable? 03:51:50 Reasonably, even. 03:53:07 CakeProphet: no, i mean the lib dems 03:53:18 the current government is a coalition between the conservative party and the liberal democrats 03:53:38 ah okay 03:53:56 I find it kind of crazy that the Labour party is socialist 03:53:57 I mean 03:53:59 socialist in name. 03:54:14 like, in the US, socialist is like a synonym for "bad". 03:54:16 That's because you're used to American politics, where "socialist" is a curse word. 03:54:42 CakeProphet: it's really not. 03:54:48 not New Labour 03:55:23 ah 03:55:31 aka Tony Blair? 03:55:38 What should weird you out is the BNP. 03:55:39 CakeProphet: basically they were useless for a few decades and then Tony Blair came along and they started calling themselves New Labour and became centre-right and got elected. 03:55:44 AKA "National Socialists". 03:55:53 elliott: whut? 03:55:56 that's weird as fuck. 03:56:10 pikhq: I think I've heard of those guys 03:56:15 CakeProphet: what, a political party getting popular and becoming crappier? 03:56:16 CakeProphet: Something rather similar happened to the Republican party, actually. 03:56:18 SHOCKING!!!! 03:56:24 "hey guys let's be conservative now!" 03:56:38 CakeProphet: no it's more like "let's look more like the conservatives because they keep winning" :P 03:56:44 pikhq: well.. was the republican party ever "socialist"? 03:57:00 CakeProphet: reminder that lincoln was a republican 03:57:09 yes yes I know all that. 03:57:14 CakeProphet: Not much *socialist*, but certainly very liberal on certain issues. 03:57:17 CakeProphet: In Canada, our opposition party is socialist 03:57:23 it's in their constitution, even 03:57:36 i don't think the usa has ever experienced any kind of popularity for socialism 03:57:40 nope. 03:57:48 too capitalist pigdog. 03:57:49 too much baggage 03:58:07 too much captialsit gipgod 03:58:14 the first sentence of their constitution, even 03:58:28 holy shit were we all surprised last election 03:58:47 elliott: It did once. We called it the "New Deal". 03:58:53 coppro: "WHEREAS the Provinces of Canada, Nova Scotia and New Brunswick have expressed their Desire to be federally united into One Dominion under the Crown of the United Kingdom of Great Britain and Ireland, with a Constitution similar in Principle to that of the United Kingdom:"? 03:59:13 elliott: he means the party's constitution. 03:59:17 elliott: the party's 03:59:23 pikhq: i dislike fdr but completely forget why 03:59:25 coppro: oh :P 03:59:32 that makes more sense 03:59:33 as opposed to like 03:59:41 WHEREAS we want a socialist opposition party 03:59:47 elliott: Probably the only notable action of his that's dislikable: Japanese internment camps. 03:59:48 and WHEREAS god help us if they get into government 03:59:50 BE IT HEREBY RESOLVED 04:00:02 elliott: he kind of started easing himself into a position of supreme authority. 04:00:06 or tried to. 04:01:19 3. It shall be lawful for the Queen, by and with the Advice of Her Majesty's Most Honourable Privy Council, to declare by Proclamation that, on and after the passing of this Act, the Provinces of Canada, Nova Scotia, and New Brunswick shall form and be One Dominion under the Name of Canada; and on and after that Day those Three Provinces shall form and be One Dominion under that Name accordingly.(4) 04:01:29 coppro: possible scam: check whether the queen actually declared that 04:01:59 4. Unless it is otherwise expressed or implied, the Name Canada shall be taken to mean Canada as constituted under this Act.(5) 04:01:59 wow, they preempted us making a nomic called canada and using the cosntitution's references to "canada" to take over! smart 04:02:32 elliott: I don't think the US constitution does anything like that. 04:02:42 maybe you should devise a coup 04:02:43 CakeProphet: yeah but you guys don't follow yours 04:02:58 i wonder if that's the one goedel found 04:03:00 elliott: ha. ha. ha ha. ha ha ha. ha ha ha ha ha 04:03:06 it's funny because it's true. also sad. :( 04:03:14 CakeProphet: ha ha ha ha ha ha ha ha. 04:03:24 elliott: The issue is, Canada doesn't have a *single* constitution. 04:03:29 There's like three Constitution Acts. 04:03:32 coppro: yes let's laugh, and then cry. 04:03:32 oh, suber reckons it was the constitution's self-amendment clause itself 04:03:41 but I hope goedel would be smarter than that 04:03:43 Sorry, 2. 04:03:57 What's this about Goedel? 04:04:04 elliott: saw a reference to this which is sort of the list version of PSeq: http://hackage.haskell.org/packages/archive/thrist/0.2/doc/html/Data-Thrist.html 04:04:13 ha ha ha ha ha ha ha ha ha ha ha ha ha. 04:04:14 elliott: the problem with the constitution is that it's too liberal for neoconservatives. 04:04:20 oerjan: hm i think i've seen this 04:04:29 pikhq: I've told you about the supreme court act, right? 04:04:29 oerjan: the list version is easy ofc, it's doing it with a finger tree that's hard 04:04:37 CakeProphet: The problem with everything but a king by divine right is that it's too liberal for neoconservatives. 04:04:42 coppro: Don't *think* so. 04:04:43 elliott: you need /warrants/ to search and seize property? I mean come on. 04:05:15 http://opendylan.org/~gabor/Thrist-draft-2008-07-18.pdf sigh, not found 04:05:18 elliott: TERRORISTS ARE PLOTTING OUR END THIS VERY MINUTE. 04:05:20 CakeProphet: You need a /majority/ to pass a law? Come on. 04:05:35 pikhq: So the constitution is defined by the Constitution Act, 1982, non-exclusively. It says "The following are part of the constitution" but it isn't exhaustive. The Constitution also says that the Constitution can only be amended with regards to the composition of the Supreme Court by unanimous consent of the provinces 04:05:57 oerjan: http://hackage.haskell.org/packages/archive/thrist/0.2/doc/html/Data-Thrist-Monad.html 04:05:58 accordingly, no one knows if the Supreme Court Act is in the constitution or not, and so no one knows if it can be amended by Parliament or only by unanimous consent 04:06:09 coppro: *Ouch*. 04:06:16 oerjan: i suspect that's similar to the relation of arrows and monads... 04:06:38 I mean, the nonexhaustiveness is sometimes useful 04:06:54 for instance, parliamentary privilege is taken to be a constitutional power, even though it's not written out in any statute 04:07:10 oerjan: ha! they did thrist zippers. 04:07:12 same with independence of the three branches 04:07:14 elliott: does the UK have a supreme court thing? erase my ignorance. 04:07:22 CakeProphet: all our judging is done by a rabbit 04:07:29 ah yes. 04:07:32 it all makes sense now. 04:07:36 CakeProphet: As of a couple years ago, they have a Supreme Court. 04:07:38 CakeProphet: It does now. 04:07:47 Prior to that, they had the House of Lords. 04:07:48 appointed? 04:07:51 so like... 04:07:55 and for a while it wasn't the entire House 04:07:58 just a committee 04:07:59 Thrist? 04:08:04 Tony Blair or whoever appointed all the judges at once? nice move. 04:08:10 oh 04:08:10 okay. 04:08:11 Well, technically it *was*, but most of the house just didn't care. :P 04:08:47 oerjan: hmm, these things subsume vectors, I think 04:08:50 oerjan: as in, fixed-length 04:09:03 actually FDRs "court-packing" bill would not have been a bad idea. 04:09:15 oerjan: data Vec a n n' where Elem :: a -> Foo a n (Succ n) 04:09:21 CakeProphet: http://en.wikipedia.org/wiki/UK_Supreme_Court#Appointments_process 04:09:26 oerjan: then PSeq (Vec a) Z len 04:10:06 coppro: blah blah blah bureaucracy 04:10:58 CakeProphet: It seems that they seeded it with a selection of Law Lords from the House of Lords. 04:11:55 pikhq: damn, you're right 04:12:20 oerjan: hm this guy's blog makes the interesting point that they could be interesting to build applicative structurse and the like out of, because you can pull them apart as well as run them 04:12:26 might make for an interesting esolang, especially the Seq version 04:14:36 I think it's kind of weird that presidents are restricted to two terms now. 04:15:00 Blame FDR. :P 04:15:03 I understand the rationale but I'm not really sure it makes any sense at all. 04:15:20 pikhq: blame the congress under FDR, more like. 04:15:48 at least it's less ridiculous than the russian mess 04:15:49 Also, personally I think it's weird that *only* presidents are restricted to two terms. 04:16:04 2 *consecutive* terms. *groan* 04:16:16 long live the Bush Dynasty. 04:16:25 oerjan: oh hm the thrist blog also points out that Show for these things is hard 04:16:26 pikhq: why is consecutive so weird? 04:16:37 oerjan: because consider (5,id) <| (id, 42) <| empty 04:16:44 pikhq: also wasn't the point to avoid the president dying while in office thanks to constant reelection 04:16:45 oerjan: :: PSeq (,) Int Int 04:16:57 oerjan: you don't have forall a b. Show (a,b) obviously, you need a Show constraint on a and b 04:17:03 pikhq: okay so I think term restrictions for congress makes sense, but not for the president? does that make sense? 04:17:06 but with that, you can't do it because of the above 04:17:23 CakeProphet: Not really. 04:17:30 oerjan: you need a pair that bundles the Show constraints basically :( 04:17:48 pikhq: hmmm, perhaps it doesn't. 04:17:49 pikhq: consecutive? 04:18:07 Sgeo|web: Obama, Obama, Bachman, Obama. 04:18:10 coppro: Actually, the point of a two term limit was that it was tradition, so might as well make it binding after FDR had the audacity to violate it. 04:18:22 elliott: yeah. btw i think the MetaPath has a similar need for the Exit class 04:18:31 CakeProphet: Um, I'm pretty sure that's not possible 04:18:35 s/audacity/popularity/ 04:18:37 Sgeo|web: Is in Russia! 04:18:37 oerjan: i just added it to all the types 04:18:42 CakeProphet: Also true. 04:19:40 Sgeo|web: how is that not possible if the only restriction is a maximum of two consecutive terms? 04:19:53 the third is not consecutive. 04:20:12 it just hasn't happened. 04:20:13 Because the restriction doesn't in fact say consecutive, as far as I'm aware? 04:20:24 Sgeo|web: *It is in Russia*. 04:20:25 uh, yes it does. 04:20:37 oh, hmmm 04:20:41 The restriction in the US is two terms total. 04:20:47 why did I think it was consecutive? :P 04:20:59 CakeProphet: surely after Bachmann there won't be enough left of the USA to get another president. 04:21:01 oerjan: i suppose I could just use http://hackage.haskell.org/packages/archive/prelude-extras/0.1.0.2/doc/html/Text-Show-Extras.html#t:Show2 04:21:04 *-n 04:21:05 oerjan: Basically. 04:21:12 oerjan: but actually no, Show2 isn't enough 04:21:18 oerjan: because of that hidden type problem 04:21:23 oerjan: honestly Bachman is one of the better GOP candidates. sad right? 04:21:28 oerjan: Probably our major territorial claim would be the moon. And that only on arguable technicalities. 04:22:08 pikhq: as in "there's a treaty that says there is no claim in space"? 04:22:25 oerjan: but our flags there!!! 04:22:36 oerjan: As far as the US is concerned, a "treaty" is "just a law" 04:22:45 We can literally take back treaties. 04:22:50 Because, well... Fuck you all. 04:23:09 (no, I don't understand the rational behind this) 04:23:15 pikhq: yes, this logic continues to make sense once the US is gone (except for the moon, except the moon isn't US territory by treaty) 04:23:17 oerjan: aha! 04:23:34 CakeProphet: It continues to make sense if *most* of the US is gone. 04:23:37 oerjan: Thrist (Monad' m) a b is actually (Kleisli m a b) 04:23:41 Let's say we're left with Puerto Rico. 04:23:45 oerjan: so the arrow relation is very strong 04:23:53 pikhq: oh well, i suppose i vaguely taking back treaties _is_ permitted by international law 04:23:55 pikhq: ah yes, our strong military center. 04:23:58 *vaguely recall 04:24:16 CakeProphet: Not so much "military center" as it is "likely to be ignored in global thermonuclear war" 04:24:37 elliott: you mean Thrist (Kleisli m) ? 04:24:40 "This time I shall prove my claim that in Haskell any Control.Arrow instance can be rewritten as a Thrist data structure." heh, neat 04:24:44 oerjan: no, 04:24:45 data Monad' :: (* -> *) -> * -> * -> * where 04:24:45 Feed :: Monad m => m b -> Monad' m a b 04:24:45 Digest :: Monad m => (a -> m b) -> Monad' m a b 04:25:28 elliott: er, a Thrist cannot be just Kleisli because Thrists can be picked apart again 04:25:30 pikhq: hmmm, when can we submit to our robot overlords? I'm getting impatient with these imperfect humans. 04:25:45 oerjan: oh well I mean, is /run/ as (Kleisli m a b) 04:25:57 oerjan: in some sense a thrist seems like inspectable composition 04:26:04 CakeProphet: First we need robot overlords. 04:26:36 * Sgeo|web would like inspectable functions 04:26:39 >.> 04:26:53 Sgeo|web: Ew. 04:27:00 Sgeo|web would like to break parametricity. 04:27:24 And with it sane typing, security in an object-capability system, ... 04:27:36 parametricity? 04:28:19 At one point, I have thought about how one might go about sandboxing Picolisp 04:29:00 pikhq: surely robot overlordship is only NP-hard. 04:30:47 oerjan: "So I am not in bad company announcing that I am preparing a paper about thrists which are the moral equivalents of free categories." 04:30:51 oerjan: *sigh* nothing new under the sun 04:31:21 elliott: well i already vaguely knew about thrists 04:31:56 -!- GreaseMonkey has joined. 04:31:57 -!- GreaseMonkey has quit (Changing host). 04:31:57 -!- GreaseMonkey has joined. 04:33:00 "Finally, the category-theoretic twist: think of the parameter p as the morphisms of a category C (with C's objects being the morphisms' domains and ranges) then Thrist p is essentially the free category of C, often written as C*." 04:33:03 gotta love that "essentially" 04:33:38 didn't i use the same word above for PSeq :P 04:34:10 i think so :) 04:39:27 uh, in C++ 04:39:42 pikhq: your 27th amendment is hilarious 04:39:48 how do I say "it's okay to change members of this referenced object, but not the entire thing itself"??? 04:40:57 can't do that? okay. 04:41:03 ask coppro 04:41:16 coppro: lmao at that one 04:42:24 brilliant scam there 04:44:08 ugh C++ WHY ARE YOU STUPID. 04:44:29 why are you using c=+ 04:45:01 monqy: class 04:45:04 surely you know this by now. 04:45:06 ahh / haha 04:45:17 yeah I have had and will have classes that use c++ too 04:45:27 but presently the only one involving programming is python / it's just as bad 04:45:38 monqy: you don't get to laugh about C++ until you program in it. 04:45:47 hey, I've programmed in it! 04:46:10 okay, well then you should be crying in sympathy. 04:46:11 I wrote an x86 compiler in it, albeit a simple compiler for a simple language 04:46:14 isntead of laughing. 04:46:17 im laughcrye 04:46:52 coppro: Gaaah. 04:47:19 wait so string literals are strings now and not chars? 04:47:23 er char *s 04:47:32 uhweriuhwdihwierhiwuer 04:47:46 -!- oerjan has quit (Remote host closed the connection). 04:47:53 coppro: Neat scam at least. 04:48:04 does it depend on where you put them 04:48:12 warning: deprecated conversion from string constant to ‘char*’ 04:48:19 monqy: no I thought they were always char arrays. 04:48:41 I forget how c++ does it 04:50:32 CakeProphet: They're const. 04:51:07 pikhq: what is the good way to care about this? 04:51:18 C++ is merely *noisier* about the UB than C in this regard. 04:52:15 -!- calamari has joined. 04:52:21 oh look I think the code they provided us is broken 04:52:23 imagine. that. 04:52:44 -!- DCliche has joined. 04:54:11 actually no it has something to do with assigning a string to another string. 04:56:14 -!- Darth_Cliche has quit (Ping timeout: 260 seconds). 04:56:32 pikhq: what might cause assignment of a string to a string reference to segfault? 04:57:09 CakeProphet: Any number of things, most of which involve someone doing nasty things to the STL class. 05:00:22 ah and for my earlier question I believe I want const& 05:00:26 instead of const Type & 05:00:42 -!- MDude has changed nick to MSleep. 05:01:57 okay well the string /exists/ 05:02:16 ah 05:02:25 but the string I'm assigning to does not :P 05:02:33 debugger ftw 05:03:25 elliott: VAX 05:03:28 (Booya) 05:03:28 pikhq: okay so when I say blah = new Type[size]; 05:03:30 Gregor: Wow. 05:03:32 Gregor: How? 05:03:42 elliott: Very, very poorly :P 05:03:42 pikhq: that doesn't automatically construct a default instance of Type for every element in the array 05:03:46 ? 05:03:48 (And I may have to remove it if it makes the code too big ...) 05:03:57 Gregor: But how did you do it :P 05:04:09 elliott: I actually search for the epilogue rather than just assuming it's right before the next function. 05:04:15 -!- DCliche_ has joined. 05:04:16 lol 05:04:33 Haven't tested if it breaks everything else yet. 05:04:35 Oh, also: clang 05:07:52 CakeProphet: Yes, that *is* the semantics. 05:07:53 -!- DCliche has quit (Ping timeout: 260 seconds). 05:08:25 -!- calamari has quit (Quit: Bye). 05:08:35 pikhq: you mean the construct-each-element thing is the semantics? 05:08:42 sorry it's kind of ambiguous . 05:08:46 CakeProphet: Yes. 05:08:54 hm, okay. well now I know shit all about what's wrong. 05:09:05 actually I know the same amount about that, but now have no idea. 05:09:08 on what to do. 05:14:25 aaaaah. 05:14:38 there's a fun bug in their function. 05:14:45 that they gave us. 05:15:12 pikhq, elliott: I've more-or-less run out of architectures. 05:15:20 Gregor: MMIX, still! 05:15:24 elliott: DAMN IT 05:15:27 Forgot :P 05:15:55 Gregor: Z80? 05:16:03 Gregor: PIC? (<-- JOKE) 05:16:24 Gregor: But yeah, Z80 is a good 'un. 05:16:27 Gregor: PA-RISC, too. 05:16:30 No 16-bit. 05:16:46 ..................oh 05:16:49 so like 05:16:50 PA-RISC could be doable but I'd need a sim (at this point virtually all RISCs are a shoo-in) 05:16:58 you know how hash tables have a size and like 05:17:02 Gregor: Z80 has 32-bit addressing, I'm pretty sure. 05:17:06 your hash function can give you indices larger than your table? 05:17:16 yeah, I forgot to mod the index. 05:17:16 CakeProphet: No shit. 05:17:45 Gregor: Hmm, OK, maybe not, but it's not thaaaat 16-bit. 05:17:50 PDP-11? 05:17:52 :P 05:18:01 Gregor: SIMH emulates a Z80: "SIMH emulates MITS Altair 8800 computer with Intel 8080, Zilog Z80 or Intel 8086 processors." 05:18:05 pikhq: Not enough RAM. 05:18:07 I mean, sure, it's 16-bit, but it's the PDP. 05:18:24 elliott: Orly? 05:18:31 pikhq: So sayeth Gregor 05:18:39 * shachaf uses a 256-bit CPU. 05:18:44 Gregor: SuperH? 05:18:47 Gotta address that RAM, man. 05:18:51 Gregor: As seen in the Sega Genesis! 05:18:57 RISC, 32-bit. Should be trivial. 05:19:00 IA64! 05:19:03 Gregor: And supports two endiannesses! 05:19:04 pikhq: sim 05:19:09 pikhq: (That doesn't suck foot) 05:19:19 Probably impossible anyways. IA64 makes x86 look normal. 05:19:26 Gregor: http://forums.evolutionm.net/ecuflash/220884-superh-emulator.html 05:19:32 elliott: SuperH is the one also known as sh[34], right? 05:19:41 Gregor: Oh, yeah 05:19:55 http://omnimaga.org/index.php?topic=8015.0 ;; wow googline superh emulator produces such unreliable sourecs. 05:19:55 elliott: Yeah, I had trouble making a working cross-compiler for it. 05:19:56 sources. 05:20:02 Gregor: Fair enough 05:20:03 elliott: qemu has sh4 :P 05:20:26 * elliott has reached the "Other" section of http://en.wikipedia.org/wiki/List_of_instruction_sets :P 05:20:35 Gregor: Are you sure you can't try the 6502? 05:20:40 Gregor: You would be a hero. 05:21:12 Definitely MMIX. 05:21:13 lol 05:21:23 I can't quite figure out the OS situation of MMIX >_> 05:21:29 How do I make a cross compiler with a C library? <_< 05:21:36 Gregor: Dude, you WROTE JSMMIX. 05:21:41 What did you do to get that running? 05:21:49 er, no const& is not what I want. 05:21:57 elliott: Switched to MIPS. 05:22:04 elliott: But more to the point, it never had an OS. 05:22:11 Gregor: I distinctly recall a bash shell before you switched. 05:22:17 elliott: Uh, no. 05:22:22 elliott: Not even close to close to that. 05:22:35 Oh, okay then. 05:22:45 Gregor: Oooh... 05:22:55 Gregor: Subleq. Before you laugh, there's a sort-of-C->Subleq compiler. 05:23:24 "sort-of"? 05:23:26 https://en.wikipedia.org/wiki/PicoJava 05:23:30 Gregor: http://esolangs.org/wiki/Higher_Subleq 05:23:36 Gregor: More C than you might expect. 05:23:42 Gregor: (Just skip straight to the examples) 05:23:55 (note: may require a C compiler to be written) 05:24:37 Gregor: "The GNU Compiler Collection includes an MMIX back-end for its C/C++ compilers, contributed by Hans-Peter Nilsson and part of the main GCC distribution since late 2001. As of January 2011, the MMIX back-end to GCC continues to be actively developed and maintained by volunteers. 05:24:37 Installation instructions for GCC + MMIX tools by Hans-Peter Nilsson. 05:24:37 §3.17.26. MMIX Options for GNU GCC version 4.5.2 (GNU GCC Web site). 05:24:37 §9.25. MMIX-dependent Features for GNU as from GNU binutils version 2.21, the assembler back-end for GNU GCC (GNU Binutils Web site). 05:24:39 The above tools could theoretically be used to compile, build, and bootstrap an entire FreeBSD, Linux, or other similar operating system kernel onto MMIX hardware, were such hardware to exist." 05:24:48 Gregor: Not sure what to do for libc though >_> 05:24:48 -!- sebbu2 has quit (Ping timeout: 240 seconds). 05:24:53 Aha 05:24:59 Gregor: newlib does mmix: http://cygwin.com/ml/newlib/2001/msg00577.html 05:25:00 as of 2011 05:25:01 erm 05:25:02 2001 05:25:13 But can I actually see its output X-P 05:25:25 Gregor: 05:25:25 MMIXware – Donald Knuth’s MMIX-SIM simple (behavioral) simulator, MMIXAL assembler, test suite, sample programs, full documentation, and MMMIX architectural (pipeline) simulator (gzipped tar file). 05:25:25 MMIXX – An X11-based graphics package contributed by Andrew Pochinsky of MIT’s Center for Theoretical Physics which, when combined with the MMIXware sources above, augments the MMIX virtual machine with a 640×480 pixel, true-color ‘virtual display’ (for UNIX/Linux). 05:26:00 I dont' mean "can I simulate MMIX" 05:26:09 I mean "when I printf, will that go to real stdout" 05:26:14 assignment operators in C++ are the hugest pain in the ass... 05:26:17 newlib's default "system" is to stub out write. 05:26:41 Gregor: Fair enough :P 05:26:50 Gregor: http://www-cs-faculty.stanford.edu/~knuth/mmix-news.html 05:26:52 Gregor: Definition of simple I/O, the runtime environment, and the simulator's online/offline iteraction commands (45KB) (final version, 31 August 2011) 05:26:56 You could PROBABLY patch in a write? 05:26:56 Oh well, to be investigated tomorrow. 05:27:01 Gregor: But I bet it already has a write(). 05:27:04 There's only one MMIX machine :P 05:27:07 Probably. 05:27:13 Either way, oh well, to be investigated tomorrow. 05:27:21 "The current target date for the ultimate volumes is the year 2020, so there is plenty of time to do the conversion." -- ever the optimist. 05:27:26 (The silly thing is it's practically guaranteed to work) 05:27:36 Wow. One of the IOCCC entries was a Lisp quine. 05:27:52 And a subset-of-Lisp interpreter in C sufficient to interpret itself. 05:27:59 pikhq: Link :P 05:28:05 http://ioccc.org/2005/mikeash/mikeash.c 05:28:08 Common Lisp. 05:29:37 according to GCC, I have invalidlty used a member. 05:30:30 Naughty naughty. 05:30:52 pikhq: mikeash! I have exchanged approx. ~50 words with that man. 05:31:47 Approximately approximately. 05:31:50 Yes. 05:32:08 (They were all about Classic Mac OS binutils support, to my knowledge.) 05:33:15 http://www007.upp.so-net.ne.jp/eiji-y/vmmmix/vmmmix.html <-- hey look, somebody ported Linux to MMIX 05:34:16 Limmix. 05:34:32 Gregor: You mean almost MMIX :P 05:34:34 http://www007.upp.so-net.ne.jp/eiji-y/vmmmix/problem.html 05:34:40 'struth 05:37:20 print ind 05:37:21 $4 = -2939 05:37:23 uh.... 05:37:29 what? 05:40:31 my guess: the hash function they gave us can produce negative values. ??????? 05:41:06 Is there something wrong with negative values for hashes? 05:41:14 uh... 05:41:19 for hash table /indices/, yes. 05:41:19 hashes are just piles of bits 05:41:31 CakeProphet: then just interpret them as unsigned numbers? 05:41:40 copumpkin: Racist! Hashes are Unicode code points! 05:43:53 print key.data() 05:43:53 $9 = 0x612288 "abbreviate" 05:43:59 print location(key) 05:44:00 $10 = -615167939 05:44:02 nice. overflow. 05:44:20 GOOD JOB /PROFESSORS OF COMPUTER SCIENCE/ 05:45:04 s/int/unsigned int/ 05:49:12 > 3000 `mod` 3000 05:49:13 0 05:50:07 just making sure I'm sane. 06:14:06 > 1 `div` 0 06:14:07 *Exception: divide by zero 06:14:27 > length [1 `div` 0, 1 `div` 0, 1 `div` 0, 1 `div` 0, 1 `div` 0] 06:14:28 5 06:27:17 # time gcc -O3 full.c 06:27:17 gcc: Internal error: Illegal instruction (program cc1) 06:27:33 elliott: Epdeet 06:28:55 1145.44 real 1102.56 user 31.51 sys 06:32:12 -!- CakeProphet has quit (Ping timeout: 252 seconds). 06:39:32 -!- DCliche_ has quit (Quit: You are now graced with my absence.). 06:42:03 -!- pikhq_ has joined. 06:42:25 -!- pikhq has quit (Ping timeout: 258 seconds). 06:46:57 Wow. The total amount spent on NASA is $471 billion. 06:47:30 For comparison, DOD gets $685 billion a year. 06:47:34 Imma be depressed now. 06:48:39 it's not about money, it's about taking things that people know about and cutting them so people know we're serious about saving money 06:48:42 like the presidential limo 06:49:12 presidential mini 06:49:43 copumpkin: And not actually being serious, of course. 06:50:06 If we *were* the military could reasonably get cut way the heck down. 06:50:17 yeah 06:50:27 Perhaps, y'know, not being half of military spending internationally. 07:26:05 -!- CakeProphet has joined. 07:26:05 -!- CakeProphet has quit (Changing host). 07:26:05 -!- CakeProphet has joined. 07:26:51 so as consolation for the scandinavians 07:27:33 I got roughly 1/3rd less fuel efficiency on this tank of gas than my previous. 07:27:51 All because of your maniac speed? 07:27:53 proof that traveling quickly expends more energy than traveling slowly. 07:28:03 Neat that it has that dramatic of an effect. 07:28:14 pikhq_: well I've been driving faster in general. 07:28:50 that calculation is just an estimate based on my current miles on this tank vs. fuel gauge position 07:28:56 when I fill up tomorrow I can get an exact number. 07:28:57 It also makes sense; the air resistance goes up, what, quadratically based on current speed? 07:29:07 no clue :P 07:32:18 pikhq_: oh wait yes I do 07:32:27 pikhq_: it also depends on surface area for drag. 07:33:34 well it depends heavily on a mystical dimensionless ratio called the drag coefficient 07:33:41 which I'm sure involves surface area somehow. 07:34:07 oh wait no surface area is part of the actual drag equation along with the coefficient I think. 07:34:18 but yes it's v^2 07:36:05 I suspect the drag coefficient for my Honda Accord is pretty significant. 07:36:25 You should do that golfball trick the MythBusters tried out to it. 07:36:55 what is that? 07:37:06 http://www.blogcdn.com/www.autoblog.com/media/2009/10/mythbustersdimple_opt.jpg -- it's not just a drag-reducing method, it's also a fashion statement. 07:37:49 mmmm drag coefficient for a 2002 Honda Accord Coupe is 0.32 07:38:02 I imagine that's similar to a 1998 Hinda Accord Coupe even though the body style is slightly different 07:38:15 fizzie: ha. 07:38:18 beautiful. 07:39:23 "The show's team completely covered a last-gen Ford Taurus with modelers clay and figured out that it would achieve about 26 mpg at a constant 65 mph. They then went about adding over 1,000 dimples to the car's exterior. To keep the experiment consistent, all 1,082 dimples removed from the clay exterior were put in a box and set in the back seat so that the car would weigh exactly the same as before dimpling. The theory is that, like a golf ball, the dimples wo 07:39:24 uld reduce the car's drag through the air, thus allowing it to travel the same distance at the same speed using less fuel. The result? Over 29 mpg." 07:39:50 hi 07:39:56 I can't quite recall what the original mpg figure for the car was before they covered it with that much clay though. 07:39:58 oh gosh there is a picture 07:40:01 what a beautiful picture 07:40:06 fizzie: did you see the computer fans with those dimples 07:40:07 they're so ugly 07:40:23 Nnno, I didn't. 07:40:25 http://www.geeky-gadgets.com/wp-content/uploads/2010/07/sharkoon-gg.jpg 07:40:28 behold 07:40:55 http://www.sharkoon.com/sites/default/files/products/tl6s08_0.jpg 07:40:58 It looks like it has a skin condition. 07:41:01 fizzie: bhehold even moreso 07:41:15 http://www.sharkoon.com/sites/default/files/products/big_sharkoon_silent_eagle_2.jpg dmiples 07:41:32 fizzie: wow, they got that much fuel efficiency from a Ford? I'm impressed. 07:41:35 dmiples 07:41:43 I want to golf ballize my honda now. 07:41:54 DMI, plz. 07:42:02 also it would increase my top speed! 07:42:04 The last picture is very artsy. 07:42:07 so perhaps I could hit 150. 07:42:52 but seriously they should make cars like that. 07:43:18 I imagine it's not even that expensive compared to the fuel savings over years of travel. 07:48:09 yes but 07:48:11 really ugly 07:51:14 elliott: oh no 07:51:15 Also I wouldn't trust any MythBusters numbers as-is; there seems to be something called "FastSkinz" (ugh), some sort of a dimpled plastic wrap for cars, which according to one test doesn't do squat. 07:51:23 American car culture will go into shock. 07:51:32 mybthusters 07:51:33 fizzie: I just read that :) 07:51:43 fizzie: fastskinz just makes me think of like 07:51:44 fastskinz 07:51:45 peeling human skin 07:51:49 and burning it off and like 07:51:57 it's still kinda soggy and burnt 07:52:01 elliott: It's made by SkinzWraps. 07:52:03 and you layer it over your car in the blistering heat 07:52:05 fizzie: maybe mybthusters did better at the airodieyamicks 07:52:06 and it sizzles slightly 07:52:10 as it starts to get redder and less pink 07:52:13 and fuses to your bonnet 07:52:21 and then it melts into the car and your car becomes part human and euuuugh 07:52:42 monqy i dont want to live any more 07:53:00 :( 07:53:13 cars are bad 07:53:20 -!- augur has quit (Read error: Connection reset by peer). 07:53:39 -!- augur has joined. 07:53:48 fizzie: fastskinz is literally just mass produced sheets that they apply to any car, it seems. 07:54:00 Mythbusters created one for a specific body shape. 07:54:04 perhaps there's a difference 07:54:37 There's more than one way to skin a car. 07:54:49 fizzie: did you envision that mental image well enough to 07:54:50 be disgusted 07:54:52 ha. ha. ha ha. ha ha ha. 07:54:53 only worsk with black car btw 07:54:59 elliott: I love peeling flesh. 07:55:04 :) 07:55:25 will you be my friend? 07:55:53 -!- Sgeo|web has quit (Ping timeout: 265 seconds). 07:57:09 flesh peels frend 07:57:19 bunrt peelt flesh cars frend 07:57:28 vroome vroome 07:57:31 "its fast" 07:58:08 why is C++ so bad. 07:58:39 because MSVC++ 07:58:48 no. 07:58:51 no that's not why. 07:58:56 because c++ 07:58:58 that's why it doesn't tend to compile 07:58:59 that has nothing to do with C++ actually. 07:59:04 no. 07:59:10 I use GCC 07:59:13 because typecasts 07:59:13 "The US Air Force devised a non-random scheme for promotion. Photograph: David Moir/Reuters" --Guardian 07:59:14 this is irrelevant 07:59:18 no not that either. 07:59:31 thanks usaf 07:59:32 one particularly irksome thing is that string vs. char * stuff is stupid and annoying. 07:59:33 i forget how you're supposed to cast stuff properly in C++ in a way that it stops bitching at you 07:59:46 CakeProphet: C++ is so bad because it has no design direction and is far too big, complicated, and self-contradictory for any possible design direction 07:59:46 hth 07:59:56 elliott: yes thank you. 07:59:58 that is it. 08:00:05 if you really want an explicit string class you should just stick with java 08:00:10 uh 08:00:11 GreaseMonkey: what 08:00:15 .. 08:00:17 :) 08:00:18 or something that isn't C++ 08:00:22 I'll let elliott handle this. 08:00:23 list of possible C++ criticisms: MSVC++, typecasts, no explicit string cast 08:00:27 that is the whole list 08:00:31 s/cast/class/ 08:00:36 GreaseMonkey: do you not know the context of this discussion.... 08:00:39 also hasn't everyone used std::string for like the last 08:00:43 decade??? or more? 08:00:44 GreaseMonkey: it's kind of: essential to: understand it 08:00:48 so there's an explicit string class for you 08:00:50 elliott: apparently not. 08:01:04 well i use C hence why i tend not to give a shit about using "string" 08:01:20 yeah USUALLY you don't use features of other languages 08:01:20 elliott: ifstreams constructor parameter is char *... also its << operator won't take strings I have to allocate an array with associated overflow problems. 08:01:21 in a language 08:01:26 like perl ties don't appear much in my haskell code 08:01:27 and whenever i have to typecast something in C++ that refuses to accept my casts, i just pull out the extern "C" block 08:01:38 elliott: dude but they should 08:01:44 ties should be integral to Haskell programming. 08:01:52 CakeProphet: uh its like .cstr() or something to convert an std::string to const char * 08:01:57 not hard :P 08:02:06 CakeProphet: I don't think ifstreams have << because... they're input streams 08:02:09 or do you mean >> 08:02:12 elliott: yes 08:02:29 .data() seems to work as well 08:02:29 c++ confusion 08:02:32 that's not the issue. 08:02:49 CakeProphet: "A global function with the same name exists in header . This global function provides a similar behavior, but with standard C++ string objects instead of c-strings: see getline (string)." 08:02:51 then again, i'm someone who would rather shoot myself in the foot than fuck around with unnecessary OO-overhead 08:03:08 grease "hardcore" monkey 08:03:08 CakeProphet: or: 08:03:09 CakeProphet: http://www.cplusplus.com/reference/string/operator%3E%3E/ 08:03:19 you have no idea how hard it was to type monkey properly 08:03:24 no idea 08:03:31 greasemonqy 08:03:35 ;_; ;_; ;_; 08:03:37 elliott: oh hi. 08:03:43 http://www.cplusplus.com/reference/iostream/istream/operator%3E%3E/ 08:03:45 so yeah they 08:03:46 can do strings?? 08:03:50 oh hi oh 08:04:42 elliott: another problem with C++: its documentation requires me to read too much. 08:04:47 I miss stuff!! 08:05:23 08:06:15 elliott: oh data isn't null-terminated 08:06:17 cstr is 08:06:22 c_str rather 08:07:37 elliott: but anyway, a legitimate complaint about C++: const 08:07:49 good idea, bad reality. 08:08:49 const is also in C :p 08:10:52 ah yes 08:10:57 this makes things much better. 08:25:47 so what happens when you try to compare an unsigned int with a signed int? 08:25:50 in C. 08:31:09 you get a warning 08:32:06 -!- sebbu has joined. 08:44:23 you *might* get a warning 08:54:42 -!- monqy has quit (Quit: hello). 08:59:05 http://esoteric.voxelperfect.net/wiki/Bath_Room_Blow_Jobs 08:59:15 all it really wants to do is link us to wikipedia 09:36:51 http://upload.wikimedia.org/wikipedia/commons/8/8f/Elman_srnn.png 09:48:28 -!- aloril has quit (Ping timeout: 260 seconds). 10:01:28 -!- aloril has joined. 10:24:36 -!- GreaseMonkey has quit (Quit: The Other Game). 10:43:08 -!- ais523 has joined. 11:05:10 -!- ais523 has quit (Ping timeout: 245 seconds). 11:11:45 -!- derdon has joined. 11:23:38 -!- Phantom_Hoover has joined. 11:27:00 hi Phantom_Hoover 11:30:07 Hello. 11:30:08 Phantom_Hoover: You have 2 new messages. '/msg lambdabot @messages' to read them. 11:47:54 -!- artoct6 has joined. 11:50:12 -!- ais523 has joined. 11:53:49 hi ais523 11:53:54 hi 11:54:01 I'm teaching at 12, so won't have much time to talk much 11:54:19 and actually, I've been here since last time I was here, just the connection here is really unreliable 11:54:27 (except when I was printing stuff or eating lunch) 11:57:33 What are you teaching? 12:02:25 -!- ais523 has quit (Ping timeout: 240 seconds). 12:15:06 The art of running away, it seems. 12:15:10 -!- Patashu has quit (Quit: MSN: Patashu@hotmail.com , Gmail: Patashu0@gmail.com , AIM: Patashu0 , YIM: patashu2 , Skype: patashu0 .). 12:16:37 -!- Phantom__Hoover has joined. 12:18:01 -!- Phantom_Hoover has quit (Ping timeout: 240 seconds). 12:21:22 * Phantom__Hoover → cake day 12:21:23 -!- Phantom__Hoover has quit (Client Quit). 12:28:32 An apt pupil, I see. 12:38:54 APT Guy? 12:56:02 fizzie: Have you ever considered becoming an enterprise? 12:58:29 And then, to boldly go where no nickname has gone before. Also, to detach the saucer, that's also a thing that gets done, ain't it. 12:58:47 fizzie: Yes; once at light speed. 12:58:58 I kid not. 12:59:05 Sorry. 12:59:06 s/light/warp/ 12:59:27 "Though highly unadvised, it was possible to separate the saucer from the stardrive section at high-warp velocities. Used by Jean-Luc Picard during the first encounter with the Q, the saucer was separated at a warp speed of 9.5. According to Data, it was highly impractical, but possible with no margin for error. The slightest mishap would have made this attempt deadly. It was also required to clear the saucer section from the stardrive section to 12:59:27 ensure safety, because as soon as separation was over, the saucer section would start to lose speed, causing danger to the stardrive section. However, Picard was able to successfully separate the saucer in order to protect the families of the Enterprise from the immense power of Q. (TNG: "Encounter at Farpoint") " 12:59:27 -!- poia has joined. 12:59:34 `@ poia ? welcome 12:59:41 poia: Welcome to the international hub for esoteric programming language design and deployment! For more information, check out our wiki: http://esolangs.org/wiki/Main_Page 12:59:43 -!- artoct6 has quit. 13:01:07 -!- poia has quit (Client Quit). 13:27:58 -!- aloril has quit (Ping timeout: 240 seconds). 13:40:42 -!- Ngevd has joined. 13:41:25 -!- aloril has joined. 13:45:43 elliott: Works on SH4 (had my x-compiler building at night :P ) 13:45:57 Gregor: lol 13:46:54 Gregor: How about... one of Intel's RISC processors? I THINK one of them has their own architecture :P 13:47:04 TBH RISC is sooooo boring though. 13:47:09 RISC is easy, yeah. 13:47:19 Gregor: http://en.wikipedia.org/wiki/Intel_iAPX_432 13:47:23 DO IT. 13:47:24 z/Series does /not/ work, by the way :P 13:48:11 Software running on the 432 does not need to explicitly deallocate objects that are no longer needed, and in fact no method is provided to do so. Instead, the microcode implements part of the marking portion of Edsger Dijkstra's on-the-fly parallel garbage collection algorithm (a mark-sweep style collector). The entries in the system object table contain the bits used to mark each object as being white, black, or grey as needed by the collector. 13:48:15 Gregor: Holy shit this thing is the best??? 13:48:54 Architecture with OO memory. Wow 13:49:21 Garbage-collected OO memory :P With built-in IPC 13:50:17 Gregor: http://en.wikipedia.org/wiki/Intel_i860 The other VLIW Intel processor! 13:50:36 The other horrible failure of a VLIW Intel processor! 13:51:20 Gregor: Are you not even gonna test on a PDP btw? :P 13:51:41 PDP, VAXen, and x86 = the world, several times over. 13:52:20 -!- boily has joined. 13:56:07 -!- ais523 has joined. 13:56:28 hi ais523 13:57:17 hi elliott 13:57:29 Phantom_Hoover: I was teaching some first-year Java students recursion 13:57:52 (elliott: it is /so annoying/ to see a question you can easily answer, and not be able to answer it because your Internet connection is refusing to work) 13:57:58 heh 13:58:45 Gregor: Test on the TI TMS320C64x, it's the non-failure non-Intel VLIW. (It's the DSP they have in OMAP3/OMAP4 and therefore in gazillions of mobile phones.) 14:00:19 ais523: what's the recommended cure if you start thinking about how wonderful ursala's data model is? 14:01:00 elliott: actually using it 14:01:11 ais523: heh 14:01:22 ais523: but i only like one bit of it! ok, I like the other bits too, but in a totally different way. 14:01:32 which bit do you like? 14:02:22 ais523: the data model 14:02:27 I just said that! 14:02:35 oh, I meant actually using the data model 14:02:40 not actually using the language 14:03:04 how do you "use" a data model? that's rather under-the-hood stuff 14:03:35 hmm… I'm beginning to think that the whole static vs. dynamic typing argument is missing the point 14:03:42 in that what the two sides are arguing aren't mutually exclusive 14:03:47 -!- Ngevd has quit (Quit: Goodbye). 14:04:01 (^ only tangentially related) 14:04:22 ais523: /me thinks http://cdsmith.wordpress.com/2011/01/09/an-old-article-i-wrote/ is relevant 14:04:28 and, hmm, you could write a language using the model 14:04:45 a conditional is dynamic typing 14:10:25 fizzie: Gregor: "The i860 did see some use in the workstation world as a graphics accelerator. It was used, for instance, in the NeXTdimension, where it ran a cut-down version of the Mach kernel running a complete PostScript stack. However, the PostScript part of the project was never finished so it ended up just moving color pixels around. In this role the i860 design worked considerably better, as the core program could be loaded into the cache 14:10:25 and made entirely "predictable", allowing the compilers to get the ordering right." 14:11:46 Gregor's quest for portability is rather madder than mine 14:11:59 I was just planning to check Linux-x86, Linux-ARM, and Windows-x86 14:12:13 possibly Linux-x86_64 too, I guess 14:12:22 ais523: you're not writing a JIT 14:12:32 but I'm planning to run a bunch of different compilers, as it's more a compiler thing that I'm exploiting than an arch thing 14:12:40 a massively portable JIT with no platform-specific code is more impressive than a C program just working on a bunch of architectures 14:13:16 right, indeed 14:13:33 and I think Gregor testing on a bunch of archs is correct, it'll lead to a more hilarious hints file 14:14:18 ais523: which will then be massively elided by the judges? :P 14:14:49 nah, they have no motive to elide it but to save a tiny amount of bandwidth 14:15:20 ais523: except that the hints files always include judge's comments and "/selected/ author's comments" 14:15:37 which never make sense as a complete hints file to send in because they're missing tons of information, usually things the judges have already stated in their own words earlier 14:15:38 elliott: but why would they deselect an arch list? 14:15:46 ais523: readability? 14:16:29 it's not too hard to mentally skip over an arch list 14:17:02 I don't understand why we have editors when I have Ctrl+F 14:17:37 elliott: well, I don't typically use editors for /reading/ files 14:17:45 relatedly, why doesn't less do syntax highlighting? 14:18:05 there really needs to be a way to indicate bad jokes as separate from being dumb 14:27:49 -!- Ngevd has joined. 14:30:39 Sorry, I had to walk my gran 14:30:50 I still need to rewrite my comments to be more amusing. 14:30:59 Ngevd: I ... what? 14:31:11 She's got a bad leg 14:31:39 hmm, what's the chance that elliott and Ngevd have met each other in real life but didn't recognise each other? 14:31:47 ais523: too high 14:31:49 I think people only use the word "walk" as transitive verb when they're referring to pets :P 14:31:49 Quite likely 14:32:02 Hexham's not a big place 14:32:29 # time gcc -O1 full.c 14:32:29 1297.88 real 1246.13 user 34.41 sys 14:32:43 Gregor: Wow, what :P 14:32:48 elliott: VAX 14:32:52 Noice. 14:32:56 PDPDPDPDPDPDPDP 14:33:08 Gregor, how's your IOCCC entry going? 14:33:13 Ngevd: I'm talking about it right now :P 14:33:34 Ngevd: x86[_64], SPARC, MIPS[el], ARM, PowerPC[64], Alpha, VAX 14:33:41 Hmm... 14:33:45 I could compile it... 14:34:00 On ALMOST ALL THE COMPUTERS I OWN 14:34:17 How about ALL THE 32-bit computers you own? 14:34:50 Ngevd: What computer do you own that isn't one of those? An Amiga? 14:35:00 Commodore? Atari? I can't think of anything not really old. 14:35:13 Wii? DS? not sure if they even apply 14:35:37 DS is ARM. 14:35:48 In fact, two different, incompatible ARMs. 14:35:54 Running simultaneously. 14:36:14 Wii is PPCish. 14:36:19 Um, I've got computer that I don't know the architecture of 14:36:23 Almost certainly PPCish enough to run my JIT. 14:36:25 Almost certainly PPCish enough for Gregor's JIT to work :P 14:36:25 And an XBox 360 14:36:28 HOLY SHIT Gregor 14:36:30 THOSE SENTENCES 14:36:31 PERFECTLY 14:36:34 MATCHED UP 14:36:35 FOR 14:36:36 :P 14:36:39 FOUR WORDS 14:36:40 JESUS 14:36:45 THAT WAS LIKE 14:36:46 QUANTUM 14:36:48 ENTANGLEMENT 14:36:49 OF WORDS 14:36:56 So, entanglement 14:36:58 Ngevd: Eks bawks: Also PPCish 14:37:47 Relatively old iPod Nano? 14:37:49 What's PS3? Oh, Cell. 14:37:56 So... PPCish. 14:38:06 Wow, SUN cc's option to not align memory is -misalign. 14:38:13 elliott: Yeah, all the modern game systems are PPCish. 14:38:21 Ngevd: ARM: "The iPod Nano uses a PortalPlayer PP5021C "system on a chip" with dual embedded 80 MHz ARM 7TDMI processors." 14:38:28 Well, my oldish computer uses ARM 14:38:39 That I doubt, unless it's a netbook. 14:38:55 Or a Samsung SoC in the second-generation Nano. 14:38:57 It's got two stickers on it saying "ARM" 14:39:02 ARM-based, of course. 14:39:09 Gregor: Get Deewiant to test it on OpenSolaris, then you can say you've tested it on REAL UNIX. 14:39:14 REAL SHIT UNIX 14:39:24 elliott: I'm actually on Purdue's Solaris right now :P 14:39:28 elliott: Hence my SUN cc comment. 14:39:37 Gregor: But that's not Open! 14:39:40 No wait, they say AMD 14:39:48 AMD, ARM, what's the diff 14:39:49 Gregor: Also, surely "Sun cc". 14:40:03 Which is x87 14:40:07 *86 14:40:08 elliott: It's the one that comes in the SUNcc package. 14:40:14 "Sun C compiler" seems to be what Sun called it, but really, SUNcc? 14:40:20 Gregor: That's internals :P 14:40:42 Gregor: At the MOST, I'll settle for SUN CC. 14:40:53 Sunc. 14:41:08 Gregor, make it able to be compiled on my digital watch and I'll be amazed 14:41:14 Gregor: Sunc is the premier compiler for the Itanic. 14:41:20 Ngevd: What model :-P 14:41:34 elliott, some sort of Casio 14:41:37 Has a compass 14:41:52 I bet that describes ~1,000,000 models of watch. 14:41:55 I'll make it run on the compass. 14:42:02 Using only magnetism for computation. 14:42:35 How about z80? I think TI still makes e.g. the TI-84+ calculator based on that, so it's not "really old". Of course it's not 32-bit, and the computerness is arguable. 14:42:45 Can it run on GBA SP? 14:42:53 Gregor: You should find a way to make it run on reality itself, answering such pressing questions as "is reality a RISC?", "what do reality's function prologues look like?", and "can reality segfault?". 14:42:57 GBA is an ARM too. 14:43:05 fizzie: I told Gregor Z80, he said waaah address space. 14:43:08 All the world's an ARM. 14:43:09 I said it's big enough. 14:43:12 Gregor said nothing. 14:43:18 IT SNOT. 14:43:25 Gregor: How can you even use more than like 16 megs of ram 14:43:38 Z80 address space is 64K though. 14:43:46 eZ80 goes up to 16M. 14:43:48 elliott: Allocating things as [W] where W = 9<<16 :P 14:44:09 Gregor: That'll work for the 16 meg one, no? 14:44:53 elliott: And the other thing I allocate as 9<<25 :P 14:45:02 Nowait, I changed that to 9<<16 too ... 14:45:16 Gregor: Just do malloc(SLIGHTLYLESSTHANW) if it fails :P 14:45:44 OK, I PROBABLY need about 3MB. 14:45:51 So eZ80 will work! 14:45:56 http://en.wikipedia.org/wiki/Van_der_Grinten_projection Wow, this is definitively the worst map projection /ever/. 14:46:09 Holy crap agreed 14:46:16 Being vandalised by xkcd morons >_> 14:46:19 Solaris mmap doesn't like -1 as protection modes :P 14:46:45 Ngevd: It has all the advantages of Mercator with none of the advantages of Mercator! 14:47:08 Sun CC seems to generate the most retarded function epilogue imaginable ... 14:47:13 I like the projection that's like those triangles 14:47:17 2fed4: 81 c7 e0 08 ret 14:47:17 2fed8: 81 e8 00 00 restore 14:47:18 2fedc: 81 c7 e0 08 ret 14:47:18 2fee0: 81 e8 00 00 restore 14:47:18 2fee4: 00 01 00 00 illtrap 0x10000 14:47:18 2fee8: 00 01 00 00 illtrap 0x10000 14:47:20 2feec: 00 01 00 00 illtrap 0x10000 14:47:22 2fef0: 00 01 00 00 illtrap 0x10000 14:47:24 2fef4: 00 01 00 00 illtrap 0x10000 14:47:32 I'm not sure what C compilers there are for eZ80, though, other than Zilog's own best-name-ever "eZ80ZDS0100ZCC" compiler. 14:47:52 eZ80ZDS0100ZCC might just be the best name for a compiler, ever. 14:48:00 Ngevd: Dymaxion was a strange dude. 14:48:17 fizzie: HOW IS IT ACTUALLY CALLED eZ80ZDS0100ZCC 14:48:39 elliott, Dymaxion's a brand name. You mean Buckminster Fuller 14:48:41 Well, it's a component for the ZDS, the Zilog Developer Studio. I suppose they don't use the full name much. 14:48:48 But that's what the title of the manual is. 14:49:20 ais523: suffer through painful irl laughter at that name with me, please 14:49:24 I'm not sure what C compilers there are for eZ80, though, other than Zilog's own best-name-ever "eZ80ZDS0100ZCC" compiler. 14:49:39 elliott: oh right, just realised what you were referring to 14:49:43 I am indeed IRL laughing 14:49:55 although it's not that painful, and I managed not to be drinking water at the time 14:50:04 Admittedly the prose only refers to it as "eZ80 C-Compiler", but that's more of a description than a name. I guess it's a matter of semantics. 14:50:11 i'm going to call @'s compiler @kjsdf'sd;sa;d'\glsd;ghkshlang 14:50:18 sorry need some numbers in there 14:50:19 i'm going to call @'s compiler @kjsdf'sd;sa;d'\glsd;ghkshlang2347839423 14:50:57 fizzie: I bet the developers working on it anthropomorphise it with a name as catchy as it has. 14:51:06 "Oh, little old eZ80ZDS0100ZCC's broken again..." 14:54:53 Hey, AVR32 is a modern 32-bit instruction set you don't have in your list. And it runs Linux and all. 14:55:23 -!- copumpkin has quit (Quit: Computer has gone to sleep.). 14:55:49 Gregor: ^ 14:58:24 Gregor: That, and I guess you could consider m68k too. I mean, sure, it's quite obsolete, but VAX. 14:58:49 I suggested that upon-a-time. 14:59:53 http://en.wikipedia.org/wiki/Special:Contributions/Petergray8 14:59:55 worst person 15:00:01 http://en.wikipedia.org/w/index.php?title=User_talk:ClueBot_Commons&diff=prev&oldid=461104563 15:00:05 absolute worst :'( 15:00:18 who the hell creates a wp account to vandalise anyway :P 15:01:57 Possibly he considers it not vandalism. There are many kinds of folks. 15:04:32 14:46, 17 November 2011 95.149.230.3 (Undid revision 461104069 by Petergray8 (talk)) (Tag: section blanking) 15:04:32 13:00, 17 November 2011 Petergray8 (http://www.xkcd.com/977/) 15:04:32 12:57, 17 November 2011 ClueBot NG (Reverting possible vandalism by Petergray8 to...) 15:04:32 12:57, 17 November 2011 Petergray8 (http://www.xkcd.com/977/) 15:04:42 What I think he's thinking now: IT KEEPS HAPPENING. 15:14:25 -!- MSleep has changed nick to MDude. 15:19:22 I have no problem with Cluebot arbitrarily removing out-of-place xkcd links 15:20:01 -!- derrik has joined. 15:24:01 -!- copumpkin has joined. 15:24:06 "Avid Reader of XKCD and believe anything stated there should be paralleled here as well." --Petergray8, describing himself, a bad person 15:39:00 -!- derrik has quit (Ping timeout: 252 seconds). 15:39:37 Gregor: STATUS REPORT 15:40:12 I went to look at the Fentiman's offices today 15:40:23 But their floor was being painted 15:41:59 -!- derrik has joined. 15:45:22 crosstool-ng seems indecisive about whether it can make an m68k target ... 15:58:32 -!- ais523 has changed nick to ais523\unfoog. 16:05:13 ais523\unfoog: Was there ever a foog clan? 16:05:25 I don't think so 16:05:38 btw, you can watch me play on termcast.org, I'm termcasting 16:05:48 That might help me stay awake. 16:06:06 , he says, tuning into Sokoban. 16:06:18 it's /that/ Sokoban level I'm in 16:06:22 the one that everyone hates 16:06:31 but I have the solution mostly memorised by now 16:07:01 TAEB523 is also termcasting 16:07:51 What clan is e in? 16:07:55 clan bot 16:08:01 founding and only member 16:08:26 I tried to persuade xaiph to join, but she doesn't understand ssh 16:08:29 ais523\unfoog: How lonely. Can I join? I guarantee I am not distinguishably better at playing NetHack than a bot :-) 16:08:40 but, umm, you aren't a bot 16:08:54 So you're saying it's a speciesist clan? 16:09:02 elliott can pass the turing antitest 16:09:13 `quote turing.*Sgeo 16:09:14 yep 16:09:16 445) Will anyone be irritated if I tend to disconnect and reconnect a lot? [...] we _almost_ have an established policy that bots will be banned it they do that. which means we might have to administer a turing test to sgeo, and that could get ugly. 16:09:25 ais523\unfoog: I protest? 16:09:25 I don't know what I did there 16:09:40 You should be able to do the Sokoban levels by just vaulting over the pits. 16:09:43 a turing antitest would be lost by humans really easily 16:09:56 elliott: you can do the first one like that; the others, though, you get dragged down by air currents 16:10:01 A Turing antitest would be some long equation with a lot of digits that's really easy to compute. 16:10:23 or for NetHack bots, something along the line of "calculate the distance between the two staircases, you have 0.001 of a second, go!" 16:10:28 heh 16:11:05 (TAEB523 would probably lose that one, incidentally; it doesn't calculate distances in steps) 16:11:25 (and uses really heavy caching to optimise distance lookups, and the cache takes several seconds to build) 16:12:09 ais523\unfoog: TAEB523 is Planar, presumably? 16:12:20 yes 16:12:40 I like how I can tell whether you've noticed I've typed something or not because you stop moving in the termcast. 16:12:46 all the Planar bots have had 523 in their name 16:13:08 ais523\unfoog: Was that zombie invisible? 16:13:13 no 16:13:19 there were two of them, they didn't last long 16:13:23 Very lightly-coloured, by any chance? 16:13:28 I'm on a white-backgrounded terminal. 16:13:34 they're white 16:13:45 ais523\unfoog: aha! 16:13:50 NetHack sends foreground color codes but not background color codes 16:13:57 ais523\unfoog: you thought you could pass the turing test! 16:14:07 ais523\unfoog: good; I would have disconnected if it forced me to look at a black background 16:14:15 right now most of the screen is legible, rather than none of it 16:14:30 AceHack sets background too because you never set one without the other 16:14:40 but I'm thinking of allowing you to swap two color codes for terminal native background/foreground 16:14:59 ais523\unfoog: yes, that's one of the main things that annoys me about acehack. 16:15:14 the fact that it's possible to clash badly with background colours doesn't mean "impose /more/ of my preferences rigidly" is the correct solution 16:15:28 it's more a case of "impose preferences without which the game doesn't work" 16:15:36 arguably, the game should be fixed to work with a white background too 16:15:41 and I'm thinking about ways to do that 16:17:26 The worst part of termcast compared to NAO is that I can't annoy people by sending them in-game mail. 16:17:31 -!- augur has quit (Remote host closed the connection). 16:17:57 ais523\unfoog, you're walking outside the map. Stop it. 16:18:05 no I'm not 16:18:17 where do you think the map /is/? 16:18:27 ais523\unfoog: You punched holes in the map! 16:18:29 It's not natural! 16:18:44 wait 16:18:51 I think my terminal just desync'd somehow 16:18:59 there we go 16:19:03 I just redrew 16:20:14 I wonder what the collective name for snakes is. 16:20:28 I know the collective name for mice is "friendship", as in "a friendship of mice". 16:21:44 -!- derrik has left. 16:21:55 hmm, I wonder if I'll ever cast knock? 16:22:02 ais523\unfoog: hmm, are you playing at >80x24? 16:22:09 it's useful as I have no other method of unlocking things, but a valk's unlikely to get it castable for ages 16:22:13 no, I'm playing at 80x24 exactly 16:22:22 as is the norm for termcasts 16:22:28 oh, my terminal was 22 high 16:22:29 how strange 16:22:37 and hmm, BoH variant, I may be in trouble if I don't find an asc kit item soon (BoH isn't urgent) 16:22:41 can you redraw again? :P 16:22:46 done 16:22:47 thanks 16:23:12 shall I do this trapping two boulders (the way everyone does it) or trapping no boulders (which is about twice as long, but massively impressive)? 16:23:54 ais523\unfoog: I'm biased because watching people play Sokoban is not terrible interesting 16:23:59 but impressiveness is always good I suppose 16:24:02 -!- Darth_Cliche has joined. 16:24:06 s/terrible/terribly/ 16:24:36 You should do it my way (well, it was the other map), which is trapping like fifty boulders, ending up with too few, and killing one of your five kittens in the process of getting more. 16:24:50 hmm 16:25:39 oh no, I haven't messed this up, have I? 16:25:52 It... looks sort of messed up, but I'm no expert. 16:26:22 OK, I'll trap one 16:27:42 it's still better than the usual solution where two are trapped 16:27:47 I wonder what I remembered wrong? 16:27:58 How do I watch? 16:28:02 You forgot the five kittens. 16:28:03 Ngevd: telnet termcast.org 16:28:05 Ngevd: $ telnet termcast.org 16:28:08 then press the letter for ais523 16:28:45 Watching 16:31:38 heh, it did come in useful, as well 16:32:21 El-berhte/ehlbereth0EleberthElerberetheElber*(48thEeb*($H$)XthElbereth 16:35:37 oh come on, first time I do the Sokoban zoo without waking it, and there's a leprechaun blocking the door to the prize 16:37:01 whoops, did I just eyebump? 16:37:04 that could have gone a lot worse 16:38:56 -!- Phantom_Hoover has joined. 16:41:02 raven get out of the dungeon 16:41:05 you dont belong here.. 16:44:37 -!- augur has joined. 16:46:55 yay mines, full of nice hapyp gomes 16:47:00 gomes like Ngevd 16:47:24 elliott: I possibly won't even mercilessly slaughter them! 16:47:39 the thing about being lawful in NetHack, is that monsters believe you'll be law-abiding regardless of whether you actually are or not 16:47:43 ais523\unfoog does not know the delights of being at peace with the gnomes. 16:48:19 hmm, ais523\unfoog's gratuitous use of ctrl+move makes me feel better about my constant usage of it 16:48:37 why press a direction key twice when you can combine it with ctrl instead? 16:48:38 heh, I'd be using autoexplore sometimes if it existed in vanilla 16:48:49 elliott: farmoving is safer than spamming a key 16:48:58 as it stops when you're in trouble, unlike holding the key down 16:49:38 * elliott would like roguelikes a lot more if they didn't involve so much micromanagement of simple map navigation 16:49:58 I should just be able to glance at a tile and say "over there" and be walked there. :p 16:50:55 you can, that's what the _ command is for; it's just that specifying the tile needs cursor movement or a click 16:51:33 ais523\unfoog: well, yes; that's considerably more work than glancing :) 16:51:56 combine it with a gaze sensor? 16:52:09 Plus speech recognition!! Best roguelike! 16:52:36 speech recognition would get wearing quickly 16:52:54 Grunt recognition. 16:53:14 sure, Manlobbi, my pickaxe is outside 16:53:20 (don't worry, I'm not actually planning to take it out of the bag) 16:54:44 I find Minetown's design somewhat perplexing. 16:57:02 indeed 16:57:11 the portal is a /dev/null thing, I think it's just placed randomly on the level 16:57:43 bleh, I can't BCU it without buying it 16:58:12 ais523\unfoog: hmm, don't recall a portal :) but i've only been idly watching 16:58:20 it's the pink ^ 16:59:13 ah. what does it do? 16:59:40 it goes to the digdug challenge 16:59:42 which is stupidly luck-based 17:00:07 I'm wondering whether to risk trying it; I can't remember if failing it can still leave you permanently trapped (I'm guessing no) 17:00:27 failing it /does/ make the current game basically unwinnable until you go succeed at it on another server 17:00:55 and I really want a ?oCharging before I try it, and I won't have one of those IDed for ages 17:01:00 on the other hand, Dx:15 17:01:01 -!- CakeProphet has quit (Ping timeout: 245 seconds). 17:01:03 should I risk it? 17:01:08 ais523\unfoog: yes 17:01:25 (note: this reply does not consider your current situation; it is based purely on the structure of the question) 17:01:33 heh 17:02:06 OK, let's try it 17:02:42 you know it's good when it starts with an elbereth 17:03:09 I spawned next to a monster 17:03:18 I had no other real choice 17:03:29 @tell elliott test 17:03:30 You can tell yourself! 17:03:32 grr 17:03:39 -!- elliott has changed nick to qqq. 17:03:41 @tell elliott test 17:03:42 Consider it noted. 17:03:43 -!- qqq has changed nick to elliott. 17:03:45 a 17:03:46 elliott: You have 3 new messages. '/msg lambdabot @messages' to read them. 17:03:48 good 17:05:31 ais523\unfoog: i love distorted elbereths 17:05:56 elbeeth 17:06:24 gah, failed 17:06:32 stupid luckbased missions 17:06:42 did I just ruin anouther tournament game for you? 17:06:47 hmm, unless I can wrest it to death 17:07:00 nothing happens 17:07:08 there's a 1/121 chance something happens 17:07:12 and it's got to happen eventually 17:07:13 oh good 17:07:23 nethack needs control structures 17:07:26 heh, and it missed anyway 17:08:09 this is so funny 17:08:46 OK, /now/ I'm screwed 17:08:59 ais523\unfoog: try poking itto death 17:09:06 you don't win the challenge if you do that 17:09:10 they're rather easy to kill with HP 17:09:21 but that just makes sure you lose 17:10:08 oh wow, that was a really crazy offchance and it /actually happened/ 17:10:22 wait, what happened? 17:10:33 random unIDed !oPoly while wearing an unIDed =oPC 17:10:35 im not sure pooka is fooled 17:10:51 oh, I lost anyway due to being hit too muich 17:10:51 wh00ps 17:10:52 *much 17:11:05 sorry about your game 17:11:18 I'll have to go play on a different server to recover from that 17:11:22 brb, startscumming wizards 17:11:37 -!- augur has quit (Remote host closed the connection). 17:11:41 taeb523 is ugly on this terminal 17:13:10 elliott: I'm going to startscum wizards on clan EIT's server 17:13:35 * elliott watches 17:14:38 ?oCharging is quite rare 17:14:38 Unknown command, try @list 17:14:42 but it makes the challenge much easier 17:14:51 well, more likely to win 17:14:55 it's not really a skill-based challenge 17:15:29 ais523\unfoog: grr, that "would you like to return" message is missing a space and uses (y/n) in place of the proper [yn] 17:15:35 indeed 17:15:48 also, makes me press return after typing y, unlike every other prompt in NetHack 17:17:20 ais523\unfoog: that's more understandable, it takes more than a few chars of patch to fix that :) 17:17:27 ais523\unfoog: also, it has two spaces after the prompt... 17:18:04 bleh, why does ?oCharging have to be so rare 17:18:13 and /oDangerouslyInflate so low on charges 17:18:34 wand of dangerously inflate is definitely name of the century imo 17:18:58 that name is the only good thing about the digdug challenge ;) 17:19:12 I /suspect/ you aren't supposed to be able to trivialise the movement around the level with E-coining 17:19:18 wow, i just realised what @'s equivalent of database indexes, and the answer makes me very happy 17:19:25 s/indexes/indexes is/ 17:21:09 -!- Ngevd has quit (Ping timeout: 260 seconds). 17:21:36 woohoo 17:21:49 now I have to somehow survive to the challenge 17:22:00 given how awful the rest of that inventory is 17:22:06 I can have fun with +oFear after a while, at least 17:22:50 also, now I have to remember how to play wizard 17:23:45 kitten why are you fetching gems, you know they're worthless 17:23:56 -!- Ngevd has joined. 17:24:18 that's better 17:24:25 I tend to give cats female names, for whatever reason 17:24:38 I may rewrite Homestuck as a epic poem 17:24:56 ais523\unfoog: because kitten is not smart kitten is just happy??? 17:25:04 In the style of, for example, the Odyssey 17:25:07 and gems are pretty............ 17:25:24 "Tell me muse, of the tale of Sburb" 17:25:25 (i am projecting onto the kitten) 17:25:47 she's called Gemma 17:25:48 kitten why are you fetching gems, you know they're worthless 17:25:48 Phantom_Hoover: You have 3 new messages. '/msg lambdabot @messages' to read them. 17:25:54 This is the least aisy thing to say? 17:26:10 I think that may be didactic hexhameter 17:26:15 *hexhameter 17:26:20 -!- augur has joined. 17:26:20 AAARGH 17:26:26 Without the second h 17:26:35 That wasn't deliberate? 17:26:37 No 17:26:54 it's a great typo anyway 17:27:07 presumably you're used to typing "hexham" 17:28:01 lichens are worth 4 XP? wow! 17:28:12 this channel is hexham + helsinki + feather 17:28:15 And I meant dactylic, not didactic 17:28:20 feather is a country? 17:28:36 yes. 17:28:40 or rather, will haven be. 17:31:13 yay, XL 2 17:33:41 wow, only 31% fail on force bolt even with metal armour? 17:33:49 *29% 17:33:54 **39% 17:34:01 I suppose that goes with my intelligence of 19 17:35:11 -!- elliott has quit (Remote host closed the connection). 17:35:32 -!- Zuu has quit (Ping timeout: 244 seconds). 17:35:42 -!- elliott has joined. 17:35:42 -!- elliott has quit (Changing host). 17:35:42 -!- elliott has joined. 17:40:16 hmm, I suppose my gnome is home 17:42:00 -!- elliott has quit (Quit: Leaving). 17:44:47 So what OS name am I supposed to use in an MMIX triple? 17:44:59 2009 17:47:20 * Gregor proceeds to use mmix-unknown-mmix as if that makes any sense. 17:47:32 Nope, doesn't like that ... 17:47:38 mmix-unknown-mmixware? 17:47:44 Aha 17:53:00 *** Configuration mmix-unknown-mmixware not supported GCCCCCCCCCCCCCCCCC 17:53:07 (That was supposed to be "GCSEEEEEEEEEEEEEEE" 17:53:27 Try changing that to A-level 17:53:36 A sentiment which Ngevd will no doubt— dammit. 17:54:07 How do I target GCC for this >_< 17:56:29 mmix-mmixware apparently. 17:56:33 mmix-unknown-mmixware is no-go 18:01:33 -!- Ngevd has quit (Ping timeout: 260 seconds). 18:15:02 1. 4000000000000094: 27fefe10 (SUBUI) privileged instruction! 18:15:04 Damn you MMIX 18:15:30 How do I tell mmix "I am privileged, let me run privileged instructions" 18:15:52 should the program need privileged instructions? 18:17:02 ais523\unfoog: Arguably. I have no mmap, so I'm just plopping instructions into malloc'd space and then executing them. Probably doing that requires some privs. 18:18:46 I'll bet it would work too, if it wasn't for those meddling kids. 18:19:05 -!- Phantom__Hoover has joined. 18:19:29 Gregor: can't you mprotect the malloced area to be executable/ 18:19:40 ais523\unfoog: No mprotect either. 18:19:47 ouch 18:20:49 -!- Phantom_Hoover has quit (Ping timeout: 240 seconds). 18:22:04 Either that, or subtraction is a privileged operation :) 18:22:04 -!- Ngevd has joined. 18:22:46 The obnoxious thing is, looking at the code, I'm confident this would work. 18:22:49 But I can't test it. 18:48:06 -!- augur has quit (Remote host closed the connection). 18:48:44 -!- GreaseMonkey has joined. 19:01:26 -!- Vorpal has joined. 19:01:36 -!- CakeProphet has joined. 19:02:06 -!- monqy has joined. 19:04:01 -!- Phantom__Hoover has quit (Ping timeout: 240 seconds). 19:05:28 hello peons. 19:05:42 Who's peons? 19:09:56 whose peons 19:11:10 eons of peons muahahahahaha 19:12:33 We've the Fifteen best 3d tv Next year worlwide : and put each one of style a single spot intended for your current verdic-doing pleasure. 19:13:30 -!- chickenzilla has quit (*.net *.split). 19:13:30 -!- shachaf has quit (*.net *.split). 19:13:34 -!- shachaf has joined. 19:13:36 -!- z^ck has joined. 19:28:38 the knowledge that Brunt is also played by jeffrey combs is really hard to integrate 19:30:34 -!- augur has joined. 19:31:17 -!- Phantom__Hoover has joined. 19:31:21 -!- augur has quit (Read error: Connection reset by peer). 19:31:27 -!- augur has joined. 19:41:09 ]http://www.npr.org/blogs/health/2011/11/17/142414818/americans-are-fat-and-expected-to-get-much-fatter?sc=fb&cc=fp 19:41:14 this is the most shocking news I've read in a long time. 19:41:18 Americans are /fat/? 19:51:43 seriously why are we not cutting military spending this is obscene. 19:51:49 erwerjwjtwehthweuhwuierh america 19:54:25 I'm going to laugh if they rescind the automatic sequestration that's supposed to occur if the supercommittee doesn't reach an agreement. 19:54:28 laugh and then cry. 19:55:50 Defense Secretary Leon Panetta has warned Congress that automatic across-the-board cuts to the Pentagon's budget would "invite aggression from U.S. adversaries." 19:55:55 holy crap this is the most retarded thing I've ever heard. 19:55:58 Like who? 19:56:07 pikhq_: like who what? 19:56:36 I think the only people that would be actively *agressive* against the US would do well to handle a single guy from Texas. 19:56:38 -!- GreaseMonkey has quit (Quit: The Other Game). 19:57:05 Rather than, y'know, just pretty damned annoyed at the US because we *haven't* cut the military. :P 19:57:14 http://en.wikipedia.org/wiki/List_of_countries_by_military_expenditures 19:57:21 okay look at this 19:57:28 Per capita Switzerland has the highest 19:57:39 and then tell me that cutting $600 billion over a decade from that 19:57:49 will make us INCREDIBLY VULNERABLE TO ATTACK. 19:57:55 AAAAAH PANIC 19:57:56 etc 19:58:09 OH NO CHINA MIGHT ATTACK THEIR SINGLE LARGEST TRADING PARTNER 19:58:20 FOR REASONS UNBEKNOWNST TO ME 19:58:26 Or worse: FRANCE 19:58:47 It could be 1812 all over again! 19:58:55 Except with a different country! 19:58:56 but no really that's insan. 19:59:01 er 19:59:02 -r 19:59:04 .. 19:59:54 we could cut from the military, pay off the deficit, and still be spending well over /quadruple/ what the next highest country in military expenditure spends on its military. 20:00:18 (over time I mean) 20:00:22 (not all at once) 20:03:17 the news tends to cover how much money programs will make in a decade or whatever. 20:03:34 but it's really more about whether or not it will reverse the expenditure/revenue disparity. 20:05:06 We could also divert that funding to NASA, and have 50x'd the NASA budget. 20:05:23 And probably produce rather absurd economic growth from not throwing that money onto a burn pit. 20:06:01 THINK OF THE SCIENCE! 20:06:48 but people see "$600 billion over a decade" and then look at a 15 trillion dollar deficit and think "oh well that's really not going to help at all this is pointless" 20:07:10 they should present in terms of how much it changes cost vs. revenue. 20:07:29 pikhq_: oh also 20:07:31 if you were sane before 20:07:33 http://www.usdebtclock.org/ 20:07:36 now you're not. 20:07:47 characters in tv series should learn to check the timestamp before they get their hopes up... *no chance* this will work at 10 minutes into the episode 20:08:23 olsner: What's worse is that if you consider debt to GDP ratio, the US isn't that bad. 20:08:36 Erm. 20:08:37 CakeProphet: 20:08:38 lol 20:08:52 olsner: That would involve more meta than audiences are usually comfortable with. 20:08:53 olsner: oh really? 20:09:15 oh, I mean 20:09:18 pikhq_: 20:09:49 pikhq_: indeed 20:10:18 so yeah on the debt clock 20:10:26 look at the section titled "largest budget items" 20:10:37 CakeProphet: We're nearing 100%. 20:10:41 now pick the ones you'd be most comfortable cutting. 20:10:50 hint: it's the defense/wars category 20:10:56 CakeProphet: Japan is currently sitting at around 250%. 20:11:09 https://upload.wikimedia.org/wikipedia/commons/0/0c/Public_debt_percent_gdp_world_map_%282010%29.svg 20:12:05 What's really effing crazy is Japan is pretty close to the US national debt *in absolute terms*. 20:12:17 huh 20:12:18 so maybe 20:12:24 we should just make less of a deal about our debt? 20:12:30 and it will cease to matter? :P 20:12:38 It's a *little* bit of an issue, but yes, it really is overblown. 20:12:52 huh. 20:12:56 so basically 20:13:02 american citizens are bad at numbers 20:13:03 okay 20:13:04 good to know. 20:13:15 Most developed nations have rather high national debt, and have for ages. 20:13:39 so maybe... 20:13:49 politicans are kind of intelligent, or at least some of them are 20:13:50 and know this? 20:13:51 This is in part because the government loaning money is a major part of the basis of the worth of fiat currency. 20:14:19 Erm, borrowing. 20:16:29 pikhq_: well, it's a systematic problem that can easily become worse (the deficit) 20:16:32 so.. 20:16:40 something worth fixing now before it gets truly problematic. 20:16:50 Quite true. 20:17:06 The key to doing that is actually rather simple: don't have a budget deficit. 20:17:14 I no rite? 20:17:18 That would require military cuts. 20:17:22 * CakeProphet gasp 20:17:25 pikhq_: TERRORISTS 20:17:28 Hence why it's not happened. 20:18:03 pikhq_: WE NEED THE MILITARY BUDGET OF SEVERAL POWERFUL NATIONS COMBINED IN ORDER TO COMBAT NON-STATE GUERILLA FORCES. 20:18:08 or else we're /doomed/ 20:18:26 Heck. I'd imagine if we just got rid of military bases in countries that have been stable in the long term and aren't near nations that *aren't*, we'd have pretty much gotten rid of the budget deficit. 20:18:39 So. Goodbye, bases in Europe. 20:19:07 Goodbye, last vestiges of the US occupation of Japan. 20:19:26 Sadly can't do anything at all about South Korea; nominally that's still a war zone. 20:20:47 Also: say the fuck goodbye to Australia and Canada (why we have bases there is beyond me) 20:21:09 pikhq_: SO THAT WE CAN ASSIST OUR ALLIES IN THEIR TIME OF NEED. 20:21:15 Oh god, not politics. 20:21:28 when the nazi terrorists land on the shores of Canada will be ready to nuke them. 20:21:33 *we will 20:22:09 Phantom__Hoover: contribute or shush :P 20:22:37 nuke canada? sounds like the natural reaction yes 20:22:45 -!- oerjan has joined. 20:22:52 characters in tv series should learn to check the timestamp before they get their hopes up... *no chance* this will work at 10 minutes into the episode 20:22:57 This DS9 you were talking about? 20:23:25 Phantom__Hoover: yes, as always recently, about a season left of it 20:23:40 Which episode? 20:23:48 (I'm assuming it was some assault on the Dominion.) 20:24:04 pikhq_, wow, looking at that debt graph, the UK isn't as bad as I'd have thought. 20:24:22 no, this was 6x24 - Time's Orphan... molly goes into a time thingamajig 20:24:33 olsner: well it was kind of to highlight the lack of sense that happens when you think about war in terms of conventional/nonconventional warfare and nuclear warfare at the same time. 20:24:49 Looking at it further, I'm seriously doubting the usefulness of it as a metric. 20:24:50 but, yes that's the best course of action. 20:24:52 in that scenario. 20:24:59 olsner, ah, yes, the last O'Brien Must Suffer! 20:25:10 (Was there one in S7, I don't really remember any.) 20:25:16 Phantom__Hoover: It seems to be only loosely useful as a metric. 20:25:25 Phantom__Hoover: I doubt /your/ usefulness as a metric. 20:25:32 What's more important, IME, is ability and willingness to repay debt. 20:25:38 I'm not a metric. 20:26:09 I'm overflowing with x, y such that x /= y, Phantom__Hoover(x,y) = 0. 20:30:51 -!- Zuu has joined. 20:31:20 Phantom__Hoover: well, in dominion war plots, success could happen that early because there are ways to go after that... 20:31:33 but in your basic get problem/solve problem plot like this, you can only have failed solutions until the end of the episode 20:32:35 pikhq_: what would be bad is if lawmakers decide to cut medicare / social security instead of military spending. 20:32:46 CakeProphet: Also impossible. 20:32:50 That would hurt old people. 20:32:52 Old people vote. 20:33:09 pikhq_: yes, and those are the two main alternatives 20:33:11 so... hopefully 20:33:14 the choice is obvious. 20:33:15 Get rid of their social security and you get a lynch mob. 20:33:35 Of old people, mind, but a lynch mob is still nothing to fuck around with. 20:33:51 pikhq_: but apparently Rick Perry wants to cut education? wtf 20:34:10 has anyone won the presidency on an offer to cut SPENDING AWAY FROM THE FUTURE OF OUR CHILDREN 20:34:13 wtf. 20:34:38 education gets you smart people, and smart people are boring as well as scary in the eyes of stupid people 20:34:54 obviously the proper approach is to stop educating people 20:36:01 -!- Phantom__Hoover has quit (Ping timeout: 240 seconds). 20:36:02 oh, replace "smart" with "educated" there, different things... 20:39:53 -!- Patashu has joined. 20:45:16 ais523\unfoog: c'est spam on le wiki 20:46:21 oerjan: deleted 20:47:45 -!- Phantom__Hoover has joined. 21:01:14 pikhq_: oh hey so I started talking about military spending on facebook and one of my friends who has a husband in the military was like "lol military spending, that's funny" 21:01:19 A CHALLENGER HAS APPEARED. 21:01:31 * CakeProphet releases the fact bombs. 21:02:11 CakeProphet: I'd imagine part of her reaction to that is because, well, our military spending is *moronic*. 21:02:19 Billions on a useless device? Okay! 21:02:31 pikhq_: er I meant to say /military spending/cut military spending/ 21:02:40 Supply soldiers basic amenities? Eh, if we feel like it, maybe. 21:03:06 pikhq_: she equates "cut military spending" with "leave soldiers who need vital equipment defenseless" 21:03:09 which was already do anyway. 21:03:19 s/was/we/ 21:12:10 pikhq_: also if we diverting funding from the military into NASA 21:12:17 not only would we get civilian technological advancements 21:12:20 but military as well. 21:12:26 or, we could anyway 21:12:31 provided NASA were sufficiently funded. 21:12:53 -!- copumpkin has changed nick to and. 21:13:22 and: hi 21:13:25 and, I resume the nick-change is related to other-channel shenanigans 21:13:37 `? welcome @and 21:13:39 welcome @and? ¯\(°_o)/¯ 21:13:39 yes 21:13:51 `? @and welcom 21:13:53 ​@and welcom? ¯\(°_o)/¯ 21:14:03 How do you do that again? 21:14:26 -!- and has changed nick to copumpkin. 21:14:26 lol 21:14:31 `@ copumpkin welcome 21:14:34 copumpkin: exec: 9: welcome: not found 21:14:39 `@ copumpkin ? welcome 21:14:41 copumpkin: Welcome to the international hub for esoteric programming language design and deployment! For more information, check out our wiki: http://esolangs.org/wiki/Main_Page 21:14:57 That's the one 21:15:13 failbot 21:15:24 failpumpkin 21:19:02 `echo '#!/bin/sh\n@ \"$1 ? welcome\"' >bin/welcome 21:19:04 ​'#!/bin/sh\n@ \"$1 ? welcome\"' >bin/welcome 21:19:09 oops 21:19:15 `run echo '#!/bin/sh\n@ \"$1 ? welcome\"' >bin/welcome 21:19:17 No output. 21:19:23 `cat bin/welcome 21:19:25 ​#!/bin/sh\n@ \"$1 ? welcome\" 21:19:32 argh 21:19:38 oh 21:20:05 oerjan: ha. ha. ha ha. 21:20:08 `run echo "#!/bin/sh\n@ "'"$1 ? welcome"' >bin/welcome 21:20:10 No output. 21:20:13 `cat bin/welcome 21:20:15 ​#!/bin/sh\n@ "$1 ? welcome" 21:20:25 `chmod +x bin/welcome 21:20:27 chmod: missing operand after `+x bin/welcome' \ Try `chmod --help' for more information. 21:20:31 argh 21:20:35 oh darn 21:21:42 `run (echo "#!/bin/sh"; echo '@ "$1 ? welcome"') >bin/welcome 21:21:44 No output. 21:21:49 `cat bin/welcome 21:21:50 -!- Sgeo|web has joined. 21:21:51 ​#!/bin/sh \ @ "$1 ? welcome" 21:22:03 `run chmod +x bin/welcome 21:22:05 No output. 21:22:10 `welcome whoever 21:22:12 whoever: Welcome to the international hub for esoteric programming language design and deployment! For more information, check out our wiki: http://esolangs.org/wiki/Main_Page 21:22:17 yay 21:22:20 `welcome The Pope 21:22:22 The: exec: 9: Pope: not found 21:22:27 BUG 21:22:31 argh 21:22:32 `welcome 'The Pope' 21:22:34 ​'The: exec: 9: Pope': not found 21:22:36 `run welcome 'The Pope' 21:22:38 The: exec: 9: Pope: not found 21:23:05 wtf 21:23:16 @ doesn't take multiple words and stuff 21:23:17 is why 21:23:17 that is 21:23:27 so it's not really a bug per se 21:23:30 ah 21:23:32 I just can't welcome the pope :( 21:23:41 `welcome Oprah 21:23:43 `cat bin/@ 21:23:44 Oprah: Welcome to the international hub for esoteric programming language design and deployment! For more information, check out our wiki: http://esolangs.org/wiki/Main_Page 21:23:44 ​#!/bin/sh \ if echo "$1" | grep -c ' ' >/dev/null; then \ echo -n "$1" | sed 's/ .*/: /' \ exec $(echo "$1" | sed 's/[^ ]* //') \ else \ echo -n "$1: " \ shift \ exec "$@" \ fi \ 21:23:45 yaaaaaay 21:24:04 oerjan: well it needs to know when the name ends and the command begins 21:24:23 restricting the name to one word (since it's supposed to be an IRC nick) is totally reasonable 21:24:40 `@ The Pope ? welcome 21:24:42 The: exec: 9: Pope: not found 21:24:47 hm... 21:24:49 `@ ThePope ? welcome 21:24:52 ThePope: Welcome to the international hub for esoteric programming language design and deployment! For more information, check out our wiki: http://esolangs.org/wiki/Main_Page 21:25:32 what's up with the >/dev/null? 21:25:36 oh 21:25:37 nevermind 21:25:37 :P 21:29:13 `@ ThePope welcome 21:29:15 ThePope: : Welcome to the international hub for esoteric programming language design and deployment! For more information, check out our wiki: http://esolangs.org/wiki/Main_Page 21:29:19 huh 21:29:22 :))) 21:29:33 oh wait duh :P 21:29:39 `@ ThePope welcome asshole 21:29:42 ThePope: asshole: Welcome to the international hub for esoteric programming language design and deployment! For more information, check out our wiki: http://esolangs.org/wiki/Main_Page 21:30:18 hm... 21:30:28 here we see the noble viking savage Oerjan perplexed by the workings of combining bash scripts. 21:31:06 he is likely playing with the large pointy horns atop his helmet. 21:31:12 because that is totally a thing vikings have. 21:31:17 very likely. 21:31:44 oh wait, horns? no way. horns are for cows 21:33:24 * olsner has a horned helmet 21:33:35 -!- ais523\unfoog has quit (Remote host closed the connection). 21:35:11 though it seems to be missing one of its horns, I suspect it became a drinking glass at some point 21:35:29 for viking... mead? 21:35:32 what do vikings drink. 21:35:38 mead is good 21:35:54 yes vikings love mead. 21:36:02 beer also good 21:36:08 wine is for girly men 21:36:16 oerjan: you've never had American light beer, I see. 21:36:33 CakeProphet: assorted alcoholic beverages of unspecified kind 21:36:46 vinland is irrelevant. _norse_ beer is good. 21:37:50 oerjan: http://en.wikipedia.org/wiki/File:Norse-world.png 21:37:58 nice worldview you have. 21:38:14 oerjan: norse beer... like ringnes? 21:38:39 Helluland. good land. 21:38:46 the swede seems to be brewing up a fight 21:41:01 hell of a land 21:43:40 oerjan: so what's it like being superior to all other races? 21:44:03 good, good 21:44:50 DOOGNIHGT 21:44:54 -!- Ngevd has quit (Quit: Goodbye). 21:44:57 yes, I imagine Hitler would have great respect for you if his empire still existed today. :> 21:45:53 `run which perl 21:45:54 ​/usr/bin/perl 21:48:06 `run cat `which perl` 21:48:09 ​ELF... 21:50:24 `run which perl test 21:50:26 ​/usr/bin/perl \ /usr/bin/test 21:50:31 ah 21:50:34 `run test 21:50:36 No output. 21:50:44 o_o? 21:51:53 i think test only acts by exit code... 21:53:15 ah yes 22:04:01 something is wrong 22:04:04 elliott is missing. 22:07:04 I'm exceptionally annoyed too. 22:09:23 `run (echo '#/usr/bin/perl -w'; echo '$_ = $ARGV[0]; if ($_) { s/ +/ @ /g; exec "bin/@", $_ . " ? welcome"; } else { exec "bin/?", "welcome"; }') >bin/welcome 22:09:26 No output. 22:09:45 `run chmod +x bin/welcome 22:09:46 -!- Slereah has quit (Ping timeout: 244 seconds). 22:09:47 No output. 22:09:50 `welcome test 22:09:53 ​/hackenv/bin/welcome: line 2: syntax error near unexpected token `{' \ /hackenv/bin/welcome: line 2: `$_ = $ARGV[0]; if ($_) { s/ +/ @ /g; exec "bin/@", $_ . " ? welcome"; } else { exec "bin/?", "welcome"; }' 22:09:56 argh 22:10:01 -!- Sgeo|web_ has joined. 22:10:23 `cat bin/welcome 22:10:24 Sgeo|web_: NEW HOMESTUCK OMG 22:10:25 ​#/usr/bin/perl -w \ $_ = $ARGV[0]; if ($_) { s/ +/ @ /g; exec "bin/@", $_ . " ? welcome"; } else { exec "bin/?", "welcome"; } 22:10:34 -!- Slereah_ has joined. 22:11:10 CakeProphet: Do you want me to start being your notifier, or are you somehow peeking ahead? 22:11:54 Because that update is old 22:12:22 -!- Sgeo|web has quit (Ping timeout: 265 seconds). 22:12:25 http://www.reddit.com/r/askscience/comments/mfpxi/if_starlight_is_extremely_old_if_someone_were_to/c30ladn?context=3 22:12:31 The ultimate obnoxious smiley. 22:12:50 @tell elliott If everything goes well a new super-resolution Skyrim pano will be ready during the weekend. I found the fov switch in the skyrim console :) 22:12:51 Consider it noted. 22:13:06 Vorpal, shut up about Skyrim you bastard the pain is too much 22:14:04 Phantom__Hoover, what? aren't you getting a new computer? 22:14:10 Yes. 22:14:17 well then 22:14:20 what is the issue? 22:14:38 elliott, displaying astonishing attention to detail, recommended a mini-ATX case. 22:14:56 Phantom__Hoover, he what 22:15:02 Phantom__Hoover, I hope you didn't buy that? 22:15:10 oops 22:15:26 Yes; it was an honest mistake, and I had to look quite hard to find that detail on the product page. 22:15:26 Sgeo|web_: I was actually just fooling you into thinking there was more 22:15:36 Phantom__Hoover, so you ordered it. ouch 22:15:37 I should have known that you hit refresh constantly waiting for updates 22:15:37 It was categorised with ATX cases, apparently. 22:15:42 and thus already know that there are no new updates 22:15:51 Vorpal, Scan seem to have a reasonable enough return policy. 22:16:28 But anyway, I ordered a replacement on Tuesday, only to discover today that my card was blocked after the original purchase. 22:16:39 Phantom__Hoover, that is good at least 22:16:44 Phantom__Hoover, what? 22:16:46 blocked? 22:16:48 what do you mean 22:17:17 In case of fraud; the purchasing history for that card is something like £8 £8 £8 £8 £750. 22:17:29 ouch 22:17:41 Phantom__Hoover, call the bank then? 22:18:12 But I'd already ordered the replacement by then, and Scan don't give a decent way to cancel the order, and it seems I ordered the last of that case in stock; elliott's absence prevents me from working around that. 22:18:13 -!- Vorpal has quit (Quit: ZNC - http://znc.sourceforge.net). 22:18:28 -!- Vorpal has joined. 22:18:32 gnh 22:18:53 Scan also haven't cancelled the purchase from the blocked card, so I have no idea whether it'll go through once that's cleared up. 22:18:56 -!- Patashu has quit (Ping timeout: 252 seconds). 22:18:58 okay that was stupid. Messed up by starting ipv6 tunnel on two computers at once 22:19:10 Best-case scenario, I get the case on Saturday. 22:19:21 Monday is not looking like an unlikely option. 22:19:28 oh well 22:19:38 still that is O(1) 22:19:46 or at most O(n) 22:19:55 depending on what you count 22:20:03 By that point, the motherboard, CPU and RAM will have been sitting assembled in an anti-static bag in an in-use utility room for nearly a week and my paranoia will have gone through the roof. 22:20:52 assembled? 22:21:03 Phantom__Hoover, don't you start with mounting the mobo in the case? 22:21:05 Plugged into each other. 22:21:07 then do all the other stuff 22:21:14 that is how I did it always 22:21:23 Vorpal, yes; I only realised the form factor was wrong after I'd put it all together. 22:21:45 just put it on your desk like a proper *desktop computer* and start using it until the case gets there 22:23:01 Leaving it assembled was basically the only option I could see at the time, particularly as I thought I'd only be waiting for a couple of days. 22:25:27 Robotfindskitten: Good game, best kitten 22:27:08 `run (echo '#/usr/bin/perl -w'; echo 'if (defined($_=shift)) { s/ +/ @ /g; exec "bin/@", $_ . " ? welcome"; } else { exec "bin/?", "welcome"; }') >bin/welcome 22:27:10 No output. 22:27:17 `run chmod +x bin/welcome 22:27:19 No output. 22:27:24 `welcome test 22:27:26 ​/hackenv/bin/welcome: line 2: syntax error near unexpected token `$_=shift' \ /hackenv/bin/welcome: line 2: `if (defined($_=shift)) { s/ +/ @ /g; exec "bin/@", $_ . " ? welcome"; } else { exec "bin/?", "welcome"; }' 22:27:34 _now_ what 22:27:47 oh of course 22:27:54 `run (echo '#!/usr/bin/perl -w'; echo 'if (defined($_=shift)) { s/ +/ @ /g; exec "bin/@", $_ . " ? welcome"; } else { exec "bin/?", "welcome"; }') >bin/welcome 22:27:56 No output. 22:27:59 `run chmod +x bin/welcome 22:28:00 No output. 22:28:04 `welcome test 22:28:06 test: Welcome to the international hub for esoteric programming language design and deployment! For more information, check out our wiki: http://esolangs.org/wiki/Main_Page 22:28:19 `welcome The Pope 22:28:21 The: Pope: Welcome to the international hub for esoteric programming language design and deployment! For more information, check out our wiki: http://esolangs.org/wiki/Main_Page 22:28:25 `welcome 22:28:28 Welcome to the international hub for esoteric programming language design and deployment! For more information, check out our wiki: http://esolangs.org/wiki/Main_Page 22:28:34 :) 22:31:34 Written by fizzie, I see. 22:33:27 what? 22:34:25 The: Pope. 22:34:49 `welcome Barack_Obama Oprah 22:34:51 Barack_Obama: Oprah: Welcome to the international hub for esoteric programming language design and deployment! For more information, check out our wiki: http://esolangs.org/wiki/Main_Page 22:34:59 good 22:35:02 we're prepared. 22:35:26 for the day when Barack Obama and Oprah simultaneously join. 22:35:27 `welcome CakeProphet 22:35:29 CakeProphet: Welcome to the international hub for esoteric programming language design and deployment! For more information, check out our wiki: http://esolangs.org/wiki/Main_Page 22:38:15 -!- augur has quit (Remote host closed the connection). 22:55:20 -!- boily has quit (Ping timeout: 252 seconds). 23:06:18 "TG: and was like shit yes i ALL KINDS of know how to use my web browser to download serveral files 23:06:21 " 23:17:45 Thought of a tooootally different way to JIT. 23:17:48 Time to do magic. 23:23:16 Gregor, oh, how? 23:23:57 * Phantom__Hoover → sleep 23:23:59 -!- Phantom__Hoover has quit (Quit: Leaving). 23:25:34 Vorpal: I'll wait to see if it works at all first :P 23:25:40 Gregor, fair enough 23:27:46 -!- Sgeo|web_ has quit (Ping timeout: 265 seconds). 23:28:50 Gregor, do you think it will take long? 23:29:01 Like, next hour? Tomorrow? Next week? 23:29:57 Several hours. 23:32:03 Gregor, oh well, going to put this computer to sleep and go to bed. @tell me if it worked or not, so I know if I should bother looking through logs for how it worked. 23:32:05 night → 23:33:39 -!- Vorpal has quit (Read error: Operation timed out). 23:42:37 I wonder if memcpy(a, NULL, 0) would consistently not segfault on all systems ... 23:48:38 void **b=&b; 23:48:42 By no logic should that work. 23:49:32 Gregor: this is C! you're in luck!