00:00:27 also somewhat terrifying 00:00:59 ever so slightly 00:01:00 olsner: yep :D sevilla, jerez, cádiz, gibraltar, marbella. 00:02:20 boily: cool, fun? 00:04:36 it was! lots of sun, old architecture, tapas... 00:09:49 -!- not^v has quit (Read error: Connection reset by peer). 00:10:00 -!- ^v has joined. 00:11:02 time to achieve sleep. 'night all! 00:11:12 (btw, where is Taneb?) 00:11:20 -!- boily has quit (Quit: COMBINATORIAL CHICKEN). 00:17:35 -!- yorick has quit (Remote host closed the connection). 00:32:23 -!- mihow has joined. 00:32:36 -!- Froox has joined. 00:33:58 -!- Frooxius has quit (Read error: Connection reset by peer). 00:39:50 -!- Phantom_Hoover has quit (Read error: Connection reset by peer). 00:53:22 -!- scoofy has quit (Ping timeout: 264 seconds). 01:00:29 -!- not^v has joined. 01:12:14 > (\s -> length $ takeWhile (!= s) $ iterate scramble s) "CHICKEN!" 01:12:16 Not in scope: ‘!=’ 01:12:16 Perhaps you meant one of these: 01:12:16 ‘!’ (imported from Data.Array), ‘M.!’ (imported from Data.Map), 01:12:16 ‘IM.!’ (imported from Data.IntMap)Not in scope: ‘scramble’ 01:13:13 > (\s -> length $ takeWhile (/= s) $ iterate scramble s) "CHICKEN!" 01:13:15 Not in scope: ‘scramble’ 01:13:51 @let scramble = concat . (_last %~ reverse) . transpose . chunksOf 2 01:13:53 Defined. 01:14:05 > (\s -> length $ takeWhile (/= s) $ iterate scramble s) "CHICKEN!" 01:14:06 0 01:14:20 18:13:27 > (\s -> length $ takeWhile (/= s) $ tail $ iterate scramble s) "CHICKEN!" 01:16:46 > (\s -> length $ takeWhile (/= s) $ tail $ iterate scramble s) "CHICKEN!" 01:16:48 3 01:18:52 > (\s -> (+ 1) . length . takeWhile (/= s) $ tail $ iterate scramble s) "potato" 01:18:54 5 01:19:01 > (\s -> (+ 1) . length . takeWhile (/= s) $ tail $ iterate scramble s) "arbitrary phrase" 01:19:03 5 01:19:19 > (\s -> (+ 1) . length . takeWhile (/= s) $ tail $ iterate scramble s) "abc" 01:19:21 2 01:19:25 > (\s -> (+ 1) . length . takeWhile (/= s) $ tail $ iterate scramble s) "bcd" 01:19:26 2 01:20:01 > (\s -> (+ 1) . length . takeWhile (/= s) $ tail $ iterate scramble s) $ take 8 [1..] 01:20:04 4 01:20:25 length "CHICKEN!" 01:20:31 > length "CHICKEN!" 01:20:33 8 01:21:37 > (\s -> (+ 1) . length . takeWhile (/= s) $ tail $ iterate scramble s) (take <$> length <*> id $ "CHICKEN!") 01:21:39 4 01:21:50 > (\s -> (+ 1) . length . takeWhile (/= s) $ tail $ iterate scramble s) (take <$> length <*> const [1..] $ "CHICKEN!") 01:21:52 4 01:22:29 > ((\s -> (+ 1) . length . takeWhile (/= s) $ tail $ iterate scramble s) . flip take [1..]) <$> [1..] 01:22:32 [1,1,2,3,3,5,6,4,4,9,6,11,10,9,14,5,5,12,18,12,10,7,12,23,21,8,26,20,9,29,30... 01:23:22 Oh. We got that earlier. :( 01:49:58 Time to complain about Haskell on Reddit. 01:50:10 (And by Haskell, I mean one common example of a GHC language extension) 02:23:38 Sgeo: Which extension? 02:26:10 DeriveGenerics 02:26:23 Although I guess the Generics mechanism is closer to the actual issue 02:26:28 And I guess that's not an extension? 02:26:43 http://www.reddit.com/r/haskell/comments/2douzn/problem_with_popular_ghcgenerics_example/ 02:33:19 -!- BlueProtoman has joined. 02:41:04 -!- conehead has quit (Quit: Computer has gone to sleep). 02:44:09 -!- diginet has quit (*.net *.split). 02:44:10 -!- KingOfKarlsruhe has quit (*.net *.split). 02:44:10 -!- olsner has quit (*.net *.split). 02:44:11 -!- not^v has quit (*.net *.split). 02:44:11 -!- shikhout has quit (*.net *.split). 02:44:11 -!- atehwa has quit (*.net *.split). 02:44:12 -!- Sprocklem has quit (*.net *.split). 02:44:12 -!- subleq has quit (*.net *.split). 02:44:14 -!- newsham has quit (*.net *.split). 02:44:14 -!- HackEgo has quit (*.net *.split). 02:44:14 -!- Taneb has quit (*.net *.split). 02:44:15 -!- b_jonas has quit (*.net *.split). 02:45:44 -!- ion has quit (Excess Flood). 03:13:20 -!- ion has joined. 03:13:20 -!- not^v has joined. 03:13:20 -!- shikhout has joined. 03:13:20 -!- diginet has joined. 03:13:20 -!- atehwa has joined. 03:13:20 -!- Sprocklem has joined. 03:13:20 -!- KingOfKarlsruhe has joined. 03:13:20 -!- subleq has joined. 03:13:20 -!- olsner has joined. 03:13:24 -!- ^v has changed nick to 21WAAZFQ3. 03:13:25 -!- not^v has changed nick to ^v. 03:13:35 -!- newsham has joined. 03:13:36 -!- HackEgo has joined. 03:13:36 -!- Taneb has joined. 03:13:36 -!- b_jonas has joined. 03:13:57 -!- 21WAAZFQ3 has changed nick to ^4. 03:40:59 -!- Sprocklem has quit (Ping timeout: 272 seconds). 03:42:01 -!- Sprocklem has joined. 04:48:07 -!- Patashu has quit (Remote host closed the connection). 04:48:42 -!- Patashu has joined. 04:54:52 -!- BlueProtoman has quit (Ping timeout: 245 seconds). 05:10:43 -!- bb010g has quit (Quit: Connection closed for inactivity). 05:44:56 -!- Lorenzo64 has joined. 05:57:44 -!- shikhin has joined. 06:01:34 -!- shikhout has quit (Ping timeout: 272 seconds). 06:36:08 -!- ^4 has quit (Ping timeout: 260 seconds). 06:54:47 -!- sebbu has quit (Ping timeout: 245 seconds). 07:06:19 -!- FireFly has quit (Excess Flood). 07:07:31 -!- FireFly has joined. 07:37:15 -!- MindlessDrone has joined. 08:48:47 -!- J_Arcane has quit (Quit: Konversation terminated!). 08:50:22 -!- J_Arcane has joined. 08:51:02 -!- sebbu has joined. 08:51:40 -!- sebbu has quit (Changing host). 08:51:40 -!- sebbu has joined. 08:52:57 -!- J_Arcane has quit (Client Quit). 08:54:58 -!- J_Arcane has joined. 09:13:58 -!- AnotherTest has joined. 09:14:22 -!- sebbu has quit (Ping timeout: 240 seconds). 09:18:20 -!- sebbu has joined. 09:18:30 Morning 09:18:36 -!- oerjan has joined. 09:18:57 -!- sebbu has quit (Changing host). 09:18:57 -!- sebbu has joined. 10:16:26 -!- J_Arcane has quit (Quit: Konversation terminated!). 10:17:58 -!- lifthrasiir has quit (Ping timeout: 264 seconds). 10:23:55 -!- J_Arcane has joined. 10:44:40 -!- lifthrasiir has joined. 10:51:54 -!- lifthrasiir has quit (Ping timeout: 246 seconds). 10:58:28 -!- shikhin has quit (Remote host closed the connection). 11:01:38 -!- MoALTz has quit (Ping timeout: 250 seconds). 11:04:08 -!- shikhin has joined. 11:09:43 -!- lifthrasiir has joined. 11:18:57 -!- yorick has joined. 11:27:17 -!- FreeFull has quit (Ping timeout: 272 seconds). 11:28:40 -!- FreeFull has joined. 11:38:48 -!- boily has joined. 11:57:49 -!- shikhout has joined. 12:01:10 -!- shikhin has quit (Ping timeout: 264 seconds). 12:11:06 -!- shikhout has changed nick to shikhin. 12:16:22 -!- Patashu has quit (Ping timeout: 240 seconds). 12:22:53 -!- Phantom_Hoover has joined. 12:32:05 -!- Phantom_Hoover has quit (Remote host closed the connection). 12:47:38 -!- Phantom_Hoover has joined. 12:55:00 -!- boily has quit (Quit: SECONDARY CHICKEN). 13:20:41 great 13:20:59 my windows partition now boots into system restore, which faffs around for a while and then says nothing can be done 13:24:48 :( 13:26:57 sounds like an analogy of my life 13:26:57 helpfully it also provides no meaningful diagnostic information 13:27:49 I spent this morning trying to fix broken graphics drivers in Debian only to discover that there's no fglrx-driver for jessie. 13:28:08 stupid kernel updates. 13:34:42 What do you need fglrx for? 13:57:04 opengl on my card is broken since the last kernel update. hell I don't even get proper framebuffer mode in the console. X still works, but Steam complains even starting and none of my games work because OpenGL is broken. 13:59:27 welcome to intel hell 13:59:28 * Melvar nods. 14:00:23 when i had a laptop with an intel chip there was some random driver update which made it crash if i ever used anything that tried to use shaders 14:00:30 they worked fine before that, even! 14:04:43 -!- conehead has joined. 14:19:08 Phantom_Hoover: It's a Radeon, actually, but that is still sort of the problem. 14:19:28 AMD cut support for anything prior to the 5xxx series, and I have a 4890. 14:19:29 welcome to anything-but-nvidia hell 14:20:01 Yeah. I was an ATI loyalist for nearly a decade, but AMD has more or less murdered that with this card. 14:23:08 If I switch from X to a non-X virtual console, my machine beeps and irrevocably futzes up the display. Welcome to nvidia hell. 14:23:11 (Admittedly they say it doesn't play nice with framebuffer devices, but I haven't managed to disable the UEFI one in any way. 14:36:51 I have a Radeon HD 4650, and it’s been running fine with the radeon driver ever since AMD stopped updating fglrx for it. 14:39:37 Except a few days ago, when trying to get a game running in wine got the card to lock up, panicking the kernel. But between the support cut and a few days ago, I had no discernably graphics card related problems. 14:39:38 fizzie: ) 14:40:07 Melvar: how long ago was the support cut? :p 14:46:16 Melvar: IF was on debian stable, it would've been fine, but there was a kernel update to testing, two actually in as many days, and ATI support broke with it. So unless someone patches it ... 14:47:57 elliott: I’m having a hard time finding it, but a few years certainly. 14:48:16 Melvar: it would be funnier if the support cut, e.g. two months ago. 14:48:23 "it worked for weeks" 14:48:30 that was how I originally read it and I'm not sure why. 14:51:09 -!- naor has joined. 14:54:16 -!- naor has left. 15:16:21 -!- oerjan has quit (Quit: leaving). 15:48:24 -!- Vorpal has joined. 15:51:20 -!- Phantom_Hoover has quit (Remote host closed the connection). 15:52:55 -!- conehead has quit (Quit: Computer has gone to sleep). 16:06:54 -!- conehead has joined. 16:07:36 -!- Phantom_Hoover has joined. 16:33:29 -!- tromp_ has joined. 16:33:54 -!- tromp_ has quit (Read error: Connection reset by peer). 16:34:24 -!- tromp_ has joined. 16:35:48 -!- tromp has quit (Ping timeout: 250 seconds). 16:37:33 -!- Froox has quit (Ping timeout: 240 seconds). 16:39:01 -!- Frooxius has joined. 17:19:47 -!- not^v has joined. 17:53:04 -!- yorick has quit (Remote host closed the connection). 17:57:24 -!- shikhout has joined. 18:00:17 -!- shikhin has quit (Ping timeout: 245 seconds). 18:03:20 -!- shikhout has changed nick to shikhin. 18:28:28 -!- conehead has quit (Ping timeout: 255 seconds). 18:35:21 -!- BlueProtoman has joined. 19:05:04 -!- conehead has joined. 19:49:00 -!- Phantom_Hoover has quit (Remote host closed the connection). 19:50:34 -!- sebbu has quit (*.net *.split). 19:50:34 -!- newsham has quit (*.net *.split). 19:50:34 -!- HackEgo has quit (*.net *.split). 19:50:34 -!- Taneb has quit (*.net *.split). 19:50:35 -!- b_jonas has quit (*.net *.split). 19:50:36 -!- diginet has quit (*.net *.split). 19:50:36 -!- KingOfKarlsruhe has quit (*.net *.split). 19:50:36 -!- olsner has quit (*.net *.split). 19:50:36 -!- conehead has quit (*.net *.split). 19:50:37 -!- ion has quit (*.net *.split). 19:50:37 -!- ^v has quit (*.net *.split). 19:50:37 -!- atehwa has quit (*.net *.split). 19:50:37 -!- subleq has quit (*.net *.split). 19:50:47 -!- sebbu has joined. 19:50:47 -!- newsham has joined. 19:50:47 -!- HackEgo has joined. 19:50:47 -!- Taneb has joined. 19:50:47 -!- b_jonas has joined. 19:53:04 -!- diginet has joined. 19:53:05 -!- KingOfKarlsruhe has joined. 19:53:05 -!- olsner has joined. 19:53:48 -!- conehead has joined. 19:53:48 -!- ion has joined. 19:53:48 -!- ^v has joined. 19:53:48 -!- atehwa has joined. 19:53:48 -!- subleq has joined. 20:18:22 -!- Phantom_Hoover has joined. 20:20:13 -!- Phantom_Hoover has quit (Remote host closed the connection). 20:43:26 -!- conehead has quit (Read error: Connection reset by peer). 20:59:29 -!- oerjan has joined. 20:59:57 -!- Phantom_Hoover has joined. 21:12:51 -!- MindlessDrone has quit (Quit: MindlessDrone). 21:25:58 -!- AnotherTest has quit (Quit: ZNC - http://znc.in). 21:27:01 -!- Phantom_Hoover has quit (Remote host closed the connection). 21:29:04 -!- zzo38 has left. 21:29:07 -!- zzo38 has joined. 21:29:35 -!- augur_ has quit (Quit: Leaving...). 21:29:51 -!- augur has joined. 21:34:44 > 23514624000 :: Int32 21:34:46 2039787520 21:37:42 :t digitToInt 21:37:44 Char -> Int 21:37:57 > 23514624000 :: Int64 21:37:59 23514624000 21:38:34 > 2091059712 :: Int32 21:38:36 2091059712 21:40:26 -!- Phantom_Hoover has joined. 21:40:37 -!- Phantom_Hoover has changed nick to Guest31250. 21:41:15 -!- Guest31250 has quit (Client Quit). 21:41:32 -!- Phantom__Hoover has joined. 21:41:33 -!- Phantom__Hoover has quit (Remote host closed the connection). 21:42:43 -!- conehead has joined. 21:44:53 -!- Phantom__Hoover has joined. 21:45:05 -!- Phantom__Hoover has quit (Client Quit). 21:45:25 -!- Phantom__Hoover has joined. 21:45:41 fuck me, virgin media have managed to ban all the torrentz proxies i can find 21:45:51 sucks 21:46:04 Phantom__Hoover: https://www.torproject.org/projects/torbrowser.html.en#downloads 21:46:35 tor seems p. heavy-duty... 21:46:56 you click it, you open it, it's firefox, you use torrentz, job done? 21:47:24 there is no reason to go around looking for website-specific proxies when tor does it better and just as easily 21:47:51 oh, it's changed a lot from when i last used it 21:48:12 -!- Phantom__Hoover has quit (Read error: Connection reset by peer). 21:51:51 can you use bittorrent through tor? 21:51:55 i have no idea how tor works, i think 21:52:06 Bike: Yes, but you shouldn't. 21:52:12 -!- Phantom__Hoover has joined. 21:52:12 darn 21:52:35 Tor lets you essentially tunnel TCP streams through it. 21:52:57 Using BitTorrent on it is just rude, because it really strains the network. 21:53:11 well i'm not planning on that obv. 21:53:34 Yeah. For that sort of thing you should just get some sort of VPN provider. 21:54:01 pikhq: it's not just rude 21:54:03 it's not anonymous 21:54:12 https://blog.torproject.org/blog/bittorrent-over-tor-isnt-good-idea 21:54:28 Oh, haah. 21:54:40 Makes perfect sense that it's doing that, but yeah. 21:54:50 So, it's rude *and* defeats the point. 21:54:52 Fun stuff. 21:55:38 Oh, *right*, tracker communication is usually UDP. 21:55:50 And uTP is tunneled over UDP as well. 21:55:55 it's understandable that the tor browser bundle defaults to scripts being on these days but it still makes me :/ 21:56:15 even if they've gone to some length to minimise the fingerprinting you can do with it 21:56:28 it's gotten a lot faster in the last few years, i see 21:57:20 btw, you should stick to https:// versions of sites (torrentz does it) on tor if at all possible, since exit relays can do whatever they want with your traffic 21:57:34 probably not a big deal for torrenting though 21:57:47 Combination of the network getting much faster and the Tor browser bundle doing some stuff to make HTTP work better over a high-latency link. 21:58:05 -!- ais523_ has joined. 21:58:06 but entering passwords unencrypted over Tor is sort of a bad idea. (admittedly, that applies to the internet at large, but Tor makes it easier to be the man in the middle) 21:58:22 pikhq: huh, what special stuff does it do? that sounds interesting 21:59:01 In short, uses Polipo as a proxy. Main notable thing that Polipo does is HTTP pipelining. 21:59:12 Which believe it or not *browsers still don't do*. 21:59:33 I seem to remember hearing a good reason for that, but I forget what it was 21:59:49 Misbehaving servers screw it up. 22:00:22 Which is why Polipo actually keeps data on whether or not a given server behaves. :) 22:00:47 it'd be nice to live in a world where the Tor network was low-latency and high-capacity enough to run all traffic over it, and it supported UDP 22:01:02 The other reason for using Polipo is that Polipo can just tunnel over SOCKS, which makes using Tor for HTTP stuff *really easy*. 22:01:16 Tor is already a SOCKS proxy 22:01:25 Yes, but Polipo is an HTTP proxy. 22:01:42 oh, um, I guess I don't know what you mean by the line I replied to 22:01:55 That can talk SOCKS instead of dialing out via normal TCP. 22:02:25 I.e. you can do browser -> polipo -> tor. 22:02:29 oh, I see. 22:02:52 Not *that* helpful for Firefox, but there's a lot of stuff that doesn't speak SOCKS but does speak HTTP proxy. 22:03:24 I take back my previous statement. it'd be nice to live in the world where the global network was anonymous, encrypted and authenticated by design. 22:03:31 Agreed. 22:04:23 There's a lot of incredibly stupid and pointless lacks-of-security in networking... 22:04:36 *cough*CA*cough* 22:04:59 ideally IP addresses would just be randomly-generated public keys, I think. 22:05:33 If designing from the ground up? Probably. 22:05:54 well, yes. how else are you going to get a network that's anonymous, encrypted and authenticated by design? 22:06:05 *nod* 22:06:50 Pity security's hard. 22:08:09 speaking of which, magnets are just hashes of their files, right? 22:08:19 Phantom__Hoover: plus a list of trackers, often 22:08:27 magnet URIs for BitTorrent? Basically, yeah. 22:08:49 Slightly more complicated than that, but it amounts to just a hash of the contents. 22:08:51 i know collisions are very unlikely under normal circumstances, but could you not construct two files with the same hash and use that to nefarious ends? 22:09:00 Phantom__Hoover: my ecommendation for getting torrents you found on torrentz is to use https://torcache.net/torrent/.torrent 22:09:04 Yes. Yes you could. 22:09:10 if you can construct two files with the same hash, you can break a lot more than bittorrent 22:09:24 Though the hash is, of course, really dang hard to break. 22:09:33 Especially because the BT info hash is a hash of hashes... 22:09:41 pikhq: I sometimes worry the security-is-impossible fearmongering makes us all give up on security. 22:09:52 elliott: well MD5 collisions are within the range of normal people now, aren't they? 22:09:54 elliott: Quite likely. 22:10:06 wow, bittorrent still uses md5? 22:10:09 ais523_: Mostly length extension attacks. 22:10:13 It never did. 22:10:16 elliott: no, but my point is 22:10:18 BT uses SHA-1. 22:10:28 that hashes once thought secure have a tendency to not be 22:10:30 ais523_: well, the idea is to move off a hash before there are practical breaks of it 22:10:31 well i know random hash collisions are almost impossible, and that engineering a hash collision for a predetermined file is just as impossible, but not what the situation was with two files you both picked 22:10:39 ais523_: protocols should plan ahead for changing the hash they use 22:10:54 Perversely BitTorrent totally has not. 22:11:20 Phantom__Hoover: if we know x, y s.t. hash(x) = hash(y) no matter how we derived x and y then that hash function is well beyond dead and buried, basically 22:11:40 incidentally, I sometimes use MD5 when I want a hash that doesn't have random collisions, and don't care about it being a cryptohash 22:11:45 huh 22:11:56 ais523_: For that I'd actually recommend using CRC32. 22:11:57 because it's the most likely to be in language stdlibs 22:12:04 At least then you don't pretend it has security. 22:12:25 pikhq: that's got like a 1 in 60000 chance of random collisions because of the birthday paradox 22:12:30 Phantom__Hoover: finding x, y s.t. H(x) = H(y) is finding a collision; finding x s.t. H(x) = h for some known h is a preimage attack; finding y =/= x s.t. H(y) = H(x) for known x is a second-preimage attack 22:12:57 collisions are easier than preimage attacks. preimage attacks are what you need to break passwords, say 22:13:01 ais523_: Well. True. *shrug* 22:13:14 but being able to do any of them in practice means we should have moved off that hash years ago 22:13:31 -!- Patashu has joined. 22:14:01 I think the current belief is that if a hash turns out to be even one bit less secure that people thought it was, it's likely going to be broken at some point and you should move on 22:14:41 Better paranoid than hacked. 22:14:42 that's why people are dubious about SHA-1, there are no practical or even impractical attacks generally known, but IIRC there is a way to brute-force it slightly faster than the theoretical maximum 22:14:53 http://valerieaurora.org/monkey.html has some nice tables that may be outdated by now 22:15:05 well, is definitely outdated; only goes to 2009 22:15:20 pikhq: I agree with that practice, basically if your hash isn't as strong as you thought it was, it means you don't understand it properly 22:15:27 and someone else might understand it better than you 22:22:38 -!- Phantom__Hoover has quit (Quit: Leaving). 22:22:49 -!- Phantom_Hoover has joined. 22:38:41 -!- ais523 has joined. 22:41:46 -!- jj2baile_ has changed nick to jj2baile. 22:48:17 -!- yorick has joined. 23:07:07 -!- ais523 has quit (Ping timeout: 272 seconds). 23:32:30 -!- oerjan has quit (Quit: Νύξ). 23:49:31 -!- ais523 has joined. 23:55:36 -!- callforjudgement has joined. 23:56:20 -!- ais523 has quit (Ping timeout: 260 seconds). 23:57:12 -!- callforjudgement has changed nick to ais523. 23:58:00 -!- shikhout has joined.