00:12:06 `addquote that's where I got it rocket launch facility gift shop 00:12:07 7| that's where I got it rocket launch facility gift shop 00:12:29 GregorR, ? 00:12:43 why was that worth adding as a quite 00:12:43 GregorR: Hahahaah 00:12:44 quote* 00:12:45 The concept of a rocket launch facility having a gift shop is weird :P 00:12:55 GregorR: Oh I thought you meant like it=sex. 00:13:00 Because that was funny because it references sex. 00:13:01 You see? 00:13:02 GregorR, Kennedy Space Center gift shop exists 00:13:04 I just checked 00:13:05 ehird: Also good :P 00:13:30 OK, 'struth ... 00:13:31 ehird, you have a dirty mind 00:13:39 But that's a SPACE rocket launch facility :P 00:13:40 AnMaster: Thank you. 00:13:52 Erm, opposite, landing, but whatever. 00:14:02 ehird, I didn't mean that as a compliment (sp?) 00:14:04 Finefine 00:14:08 `revert 12 00:14:08 AnMaster: That's your problem. 00:14:09 Done. 00:14:13 NO 00:14:15 TAKE THAT BACK 00:14:20 `addquote that's where I got it rocket launch facility gift shop 00:14:21 7| that's where I got it rocket launch facility gift shop 00:14:25 .... 00:14:26 `revert `revert 12 00:14:36 :< 00:14:41 `revert 13 00:14:42 Done. 00:15:00 ehird, it is GregorR's bot. Respect what he decides. 00:15:09 `addquote that's where I got it rocket launch facility gift shop 00:15:10 8| that's where I got it rocket launch facility gift shop 00:15:12 notnx 00:15:16 `listquotes 00:15:16 It's the world's most hackable bot, don't respect anybody for anything. 00:15:16 No output. 00:15:17 `revert 12 00:15:18 Done. 00:15:18 Or something 00:15:21 `addquote that's where I got it rocket launch facility gift shop 00:15:23 7| that's where I got it rocket launch facility gift shop 00:15:25 AnMaster: we can go at this all day. 00:15:25 `revert 12 00:15:26 Done. 00:15:28 `addquote that's where I got it rocket launch facility gift shop 00:15:29 7| that's where I got it rocket launch facility gift shop 00:15:30 `revert 12 00:15:30 Done. 00:15:34 Argh, not this again X_X 00:15:34 ehird, all night you mean 00:15:41 GregorR, agreed 00:15:47 Fine. I'll stop. 00:20:44 -!- BeholdMyGlory has quit (Remote closed the connection). 00:27:26 ehird: wat 00:27:48 see above what i said 00:28:08 i saw. i dont get it 00:28:30 err what 00:29:08 augur, you need sleep 00:29:14 or something 00:29:21 what?? 00:29:36 AnMaster: why? 00:29:39 he typed some stuff in ocaml, i guess. i dont know what he was trying to show me 00:29:48 augur: no, above that 00:30:04 AnMaster: dude, its evening in the us 00:30:10 ehird, oh not the last bit 00:30:16 oh, above that? 00:30:37 augur: yes. 00:30:46 which part? 00:31:15 also, regarding the spaces in place of parens: russell and whitehead had some sort of interesting system for avoiding parens, apparently 00:31:18 using dogs 00:31:19 .. 00:31:20 dots 00:31:32 to show precedence relationships 00:31:52 augur: immediately above 00:32:06 zombie anuses? 00:32:06 augur: it was a joke about you being gay or something I think 00:32:13 how so? 00:32:27 23:47 GregorR: People for the ethical treatment of zombie anuses? 00:32:27 23:47 ehird: Hot. 00:32:29 23:47 ehird: augur: 00:32:37 augur: It was just a one-off joke. 00:32:40 hurf durf anal sex etc. 00:32:46 i see 00:32:54 bye 00:34:02 augur, now our token gay member. 00:34:16 i thought i was always your token homosexual 00:35:49 ... People care about that? 00:36:25 Thought that at least on the Internet we were sane, reasonable people. 00:36:34 "People for the Ethical Treatment of Zombie Anuses" doesn't really come across as gay per se to me X-P 00:36:37 sane? here? 00:36:48 reasonable, occasionally 00:37:03 lousy sense of humor, always 00:37:11 lol 00:37:55 GregorR: well that would be homosexual necrophilia 00:38:14 and maybe coprophilia too, since that was mentioned earlier 00:38:22 Depending on the kinkiness-level involved, "ethical treatment" may be disputable in any case :P 00:38:23 oerjan: its not necrophilia 00:38:25 theyre UNdead 00:38:27 not dead 00:38:43 would that be anecrophilia, then? 00:38:47 Ah. Missed the context. 00:38:53 zecrophilia? 00:39:03 zombrophilia. 00:39:21 this. is. greek. no funny portmanteaus here, please. 00:39:43 proper prefixes and suffixes, than you 00:39:50 *thank 00:40:26 what's zombro? 00:40:59 ah still a portmanteau 00:42:54 Are we talking of zoophilia? 00:43:27 i don't think zombies are considered animals (more than humans), are they 00:43:34 no no 00:43:39 zombrophilia 00:43:43 with zombies 00:43:45 Hm 00:43:52 Zombro makes me think of sombreroes 00:43:59 me too 00:44:10 that r is so out of place 00:44:26 Can we make out while wearing giant sombreroes? 00:45:04 zombreros? 00:45:12 Zombie sombreroes 00:45:22 Don't put them on top of your brain fool! 00:45:34 hm do they make giant sombreros with two head whatchamacallits? 00:45:47 head holes? 00:45:52 A sombrero built for two? 00:45:55 possibly 00:45:57 yes 00:46:04 Isn't that romantic 00:46:08 If impractical 00:46:31 like those gloves i saw (on reddit?) a while ago 00:47:14 Yes, a mitten built for two 00:47:20 smittens, iirc 00:47:24 I posted them on Secret area of VIP quality 00:47:28 Because they love mittens 00:49:21 "two headed sombrero" does not google :( 00:50:36 like those gloves i saw (on reddit?) a while ago <-- link 00:52:28 Slereah_: apparently so, "VIPPER is the guy who hijacks that thread and changes the subject to MITTENS." 00:53:19 Does /prog/ have mittens? 00:53:39 Mittens are kind of a 4-ch and SAoVQ thing, not sure 4chan gets much mittens 00:54:07 4-ch != 4chan ? 00:54:21 this stuff is confusing 00:55:51 AnMaster: the company site is http://www.smittens.biz/Smittens/Home.html, i don't recall the actual reddit link 00:56:00 -!- FireFly has quit ("Later"). 00:56:54 that link had a cheesy picture of a couple using them 00:56:58 iirc 00:58:39 ah http://www.crunchgear.com/2009/01/21/smittens-worlds-dumbest-mittens/ 01:01:17 Hey, don't diss mittens >:| 01:01:29 hes not dissing mittens 01:01:32 hes dissing smittens. 01:01:39 click the link, faggot 01:02:35 Smittens are mittens, augur 01:02:40 Part of the mitten brotherhood 01:02:45 no, theyre not 01:02:55 Are you smitten with mittens on kittens? 01:03:01 don't dis people. like this there adolf friend of mine... 01:03:52 GregorR: he might have been bitten 01:05:23 No one wants to be bitten by smitten kittens with mittens. 01:05:54 Except Joe Biden 01:06:06 Biden bitten by smitten kittens with mittens 01:06:11 ......... /me has no idea what that's supposed to refer to :P 01:06:22 You realize "Biden" doesn't rhyme with "bitten", right? 01:06:22 you may be hitt'n on something there... 01:06:47 Best I could do on such short notice 01:08:22 biden and bitten are close when written 01:08:45 And yet, pronounced so differently :P 01:08:53 long i in biden? 01:09:04 Yes 01:09:17 Baiden 01:09:18 * oerjan may never have actually heard the name 01:09:49 You're forbidden to sit in, because your -ittens don't fit in. 01:10:26 well in that case i'm quittin' 01:34:57 -!- inurinternet has quit (Connection timed out). 02:10:11 -!- inurinternet has joined. 02:19:10 -!- Warrigal has quit ("leaving"). 03:02:50 -!- upyr[emacs] has quit (Remote closed the connection). 03:14:24 -!- oerjan has quit ("Good night"). 03:14:44 -!- bsmntbombdood_ has changed nick to bsmntbombgirl. 03:58:19 GregorR: I am of the opinion that I should listen to classical other than the works of Nobuo Uematsu. 03:58:23 Suggestions? 03:58:43 (beyond critiquing my usage of the term 'classical'?) 04:01:30 All classical is other than the works of Nobuo Uematsu :P 04:01:52 Also, I'd recommend starting with Romantic and Baroque and working your way inwards. 04:02:40 Also, GREGOR HATE BEING SWEATY 04:04:51 Specific suggestions. 04:05:04 I suggest that you log in to AIM with a non-shitty client. 04:05:20 You're talking to someone who is barely aware of the terms "Romantic" and "Baroque", and wants to broaden his horizons. 04:05:27 Define 'non-shitty'? 04:05:35 One which can receive sent files. 04:05:42 Ah. 04:05:56 Let's see if I *have* one. 04:30:55 -!- Warrigal has joined. 04:31:08 Yay, I am now connected from lunch.normish.org. 04:31:22 Be jealous; I command you. :-P 04:33:00 pikhq: start with vivaldi! 04:33:06 or hendel 04:34:56 handel 04:35:34 whatever. he sucks anyway. 04:35:40 also, if you're going to name baroque composers and omit bach, you're a prick 04:35:52 and jeeze, don't omit the late renaissance 04:35:56 nono, this is intentional 04:36:04 palestrina, orlando di lasso, these are all wonderful 04:36:09 you would start with vivaldi and go on to bach 04:36:16 starting with palestrina is already a bit odd 04:36:28 lord, why 04:36:37 well 04:36:52 the reason i want to start with vivaldi is so that then you can switch to bach and appreciate how awesome bach is 04:37:10 palestrina is too different from modern stuff to be at all comparable, imo 04:37:30 who cares about comparable, it's amazing music 04:38:36 there's tons of amazing music out there 04:39:09 it makes sense to start with the more mainstream stuff though 04:39:11 there's not tons of music of palestrina's caliber 04:40:30 what about every other big name 04:40:50 not all that many of 'em 04:41:08 there's like 15 composers who're better known than palestrina, are you saying any of them are worse? 04:41:11 except schubert 04:41:30 schubert is amazing 04:41:41 and there are plenty of people better known than Palestrina who are worse 04:41:46 let me see 04:41:49 richard strauss 04:41:53 i would say there are a LOT of composers better known than palestrina 04:41:54 really awful 04:41:57 CESSMASTER: well come on. 04:42:09 first, you really really suck 04:42:10 who else is famous and worse than palestrina 04:42:13 strauss is awesome 04:42:15 Nice thing about having eclectic tastes is being introduced to whole new genres, rather than just new groups, artists, etc... 04:42:23 strauss is bombastic substance-free german nonsense 04:42:24 Mozart is nice. 04:42:27 he's awesome just for that zarathustra intro 04:42:57 strauss is like wagner, but with no point 04:43:22 that's just german romanticism. You can't blame him for belonging to a movement. 04:43:48 there's plenty of post-wagner stuff that's not awful like strauss 04:44:14 mahler, for instance 04:44:31 pikhq: http://www.youtube.com/watch?v=rKZr3ExeXUc 04:45:58 schoenberg's stuff at the ass-end of german romantic tonality is also quite good 04:53:32 You guys and your 'knowing this genre'. 04:58:12 Malinowski is obscure but fun. 04:58:58 Obscure enough that I should probably give his first name. 05:00:17 * Warrigal reboots 05:05:15 Stephen. 05:12:59 -!- amca has joined. 05:50:37 Listening to music with the sound muted is so strange. 06:03:39 Like putting on a blindfold and still seeing. 06:10:46 -!- CESSMASTER has quit ("☃☃☃☃☃☃☃☃☃☃☃☃☃☃☃☃☃☃☃☃☃"). 06:52:06 -!- MigoMipo has joined. 07:13:47 -!- immibis has joined. 07:23:04 -!- MigoMipo has quit ("QuitIRCServerException: MigoMipo disconnected from IRC Server"). 07:34:09 -!- coppro has quit (Read error: 110 (Connection timed out)). 07:59:59 -!- clog has quit (ended). 08:00:00 -!- clog has joined. 08:06:46 There is the wlan in the conference place. 08:08:42 Unfortunately I think I'm going to have to listen to the presentations and not just IRC here. 08:12:06 -!- asiekierka has joined. 08:12:08 Hello 08:14:38 Hello asiekierka. 08:14:50 I think I'm crazy 08:15:02 I want to make an extended version of Windows 3.11 08:15:11 I think I should get and work on Windows NT 3.51 instead though 08:15:25 Why? It even runs Firefox (mostly), unlike Windows 3.11! 08:27:16 http://codu.org/imgs/win3plusplus.png 08:27:47 The main problem with most programs is its lack of a registry, which ironically makes F/OSS programs much more likely to work than proprietary. 08:41:33 Oh, yeah, that just makes me SO want to run NT 3.51 alongside XP on my PC 08:41:40 It would need a bunch of bugfixes 08:41:43 and drivers that don't exist! 08:41:45 Why? 08:41:49 1) No HD Audio driver 08:41:58 2) No GeForce driver (but we can get around that) 08:42:01 3) No USB driver 08:42:10 4) No PCI-E ethernet card driver! 08:52:54 I think it could work well on my 100mhz Pentium laptop 08:53:01 it's not as laggy as NT4 but still runs a bunch of apps 09:11:20 [Global Notice] Hi all, I'm going to restart services -- you'll be without NickServ, ChanServ, MemoServ etc for a few moments! Sorry for the inconvenience and thank you for using freenode. 09:11:25 from christel 09:13:39 they're back 09:13:54 oh 09:44:47 why did you repeat the global notice? we all got it... 09:45:03 anyone know how to set a prompt in a MOO? 10:13:17 -!- AnMaster_ has joined. 10:24:39 -!- AnMaster has quit (Connection timed out). 10:29:13 -!- immibis has quit (Read error: 104 (Connection reset by peer)). 11:43:13 -!- Sgeo has joined. 11:51:19 -!- GreaseMonkey has quit ("HydraIRC -> http://www.hydrairc.org <- Nobody cares enough to cybersquat it"). 13:07:17 hooray! another mystery solved with the help of wireshark 13:28:30 -!- amca has quit ("Farewell"). 13:36:25 -!- upyr[emacs] has joined. 13:52:18 Gracenotes, what mystery? 13:52:20 -!- AnMaster_ has changed nick to AnMaster. 14:02:52 -!- oerjan has joined. 14:05:06 -!- Judofyr has joined. 14:06:35 -!- KingOfKarlsruhe has joined. 14:45:24 -!- oerjan has quit ("leaving"). 15:18:48 -!- Judofyr_ has joined. 15:22:15 -!- MigoMipo has joined. 15:24:43 -!- Judofyr_ has quit ("raise Hand, 'wave'"). 15:27:13 -!- asiekierka has quit (Read error: 110 (Connection timed out)). 15:30:41 -!- Judofyr has quit (Read error: 113 (No route to host)). 15:32:54 -!- CESSMASTER has joined. 15:44:05 -!- oklopol has joined. 15:45:02 i joined this channel 15:56:37 hi 15:57:05 helloes 15:59:01 -!- inurinternet has quit (Connection timed out). 16:01:52 well this was fun 16:01:57 i must go now 16:02:07 sicp is a good book 16:02:33 actually gets pretty interesting in chapter 4 16:02:42 or not 16:02:48 yeah i think it does 16:02:55 pretty sure i do. 16:03:41 i just guess it shouldn't, because i've seen the exact same thing done multiple times 16:03:45 okokokokokokokokokokokoko 16:03:50 okokokokokokokokokokokokokokokokoko 16:04:20 (chap4 = interpretation) 16:04:33 there's a chain in front of me 16:04:41 -!- inurinternet has joined. 16:04:51 it's like made of metal 16:04:59 i don't know what "metal" is in spanish 16:06:29 this off it -> 16:06:32 -!- oklopol has quit ("WebIrc http://www.hiillos.org/irc/"). 16:08:57 -!- comex has quit ("Changing server"). 16:09:16 -!- comex has joined. 16:15:46 -!- kar8nga has joined. 16:28:37 -!- KingOfKarlsruhe has quit (Remote closed the connection). 16:34:26 -!- kar8nga has quit (Remote closed the connection). 16:40:14 -!- sebbu2 has joined. 16:41:24 -!- ais523 has joined. 16:44:59 -!- sebbu has quit (Read error: 60 (Operation timed out)). 16:48:26 -!- GregorR-L has joined. 17:15:11 Intel has apparently canned production of 45nm Havendale chips -- planning to just skip to 32nm. 17:25:14 -!- inurinternet has quit (Read error: 104 (Connection reset by peer)). 17:31:13 -!- inurinternet has joined. 17:31:56 -!- inurinternet has quit (Client Quit). 17:57:00 Anybody written a virtual pet in some scripting language? 17:57:07 (Using sqlite3 for data storage) 17:58:25 where does this David slowed stuff come from? 17:58:38 NOBODY KNOWS 17:58:52 But David did slow his pace slightly as his ears! 18:15:03 -!- MigoMipo has quit ("QuitIRCServerException: MigoMipo disconnected from IRC Server"). 18:42:05 `run perl -wlne'END{print$n}eof&&$n++;/([^<]+)/i&&$n--' * 18:42:06 <HackEgo> No output. 18:42:10 <GregorR-L> >_> 18:51:00 -!- CESSMASTER has quit ("Computer has gone to sleep"). 19:05:04 <olsner> pikhq: source? 19:06:18 <pikhq> GregorR-L: Ah, esolangs. 19:07:12 <AnMaster> <GregorR-L> But David did slow his pace slightly as his ears! <-- ... , 19:07:26 <AnMaster> not ! 19:07:40 <GregorR-L> I also changed "slowed" to "did slow", so clearly my intention was not to quote it precisely. 19:07:55 <AnMaster> ah missed that 19:07:56 <AnMaster> sorry 19:13:45 <AnMaster> hash tables are prone to DDoS. But are balanced trees really immune to that? 19:14:58 <AnMaster> just wondering, since at least the implementation I'm using has severe issues with some operations I need. Fast in general, but when inserting a mostly sorted list it takes ages. Yes I realise it probably rebalances, but no there isn't really any alternative. 19:15:56 <AnMaster> as in, I can't control if the list is sorted or not. 19:21:25 <AnMaster> (and I only get one element at a time) 19:30:30 <Asztal> can anyone here visit http://webchat.freenode.net/ ? 19:30:37 <Asztal> I just get an encoding error 19:30:46 <Deewiant> Ditto 19:31:47 <fizzie> Lucky you; I just seem to be getting a blank screen. 19:32:06 <Deewiant> Well, I got some error about not having javascript before I enabled it 19:32:30 <fizzie> I got that too, then I said to noscript to temporarily allow it, and after that was when I got the empty page. 19:34:25 <Asztal> now I get a page, but it won't actually connect; the button does nothing... 19:34:39 <GregorR-L> It works, but can't connect for me. 19:34:51 <GregorR-L> [14:34] == *** (qwebirc) Looking up your hostname... 19:34:51 <GregorR-L> [14:34] == *** (qwebirc) Found your hostname. 19:34:51 <GregorR-L> [14:34] == Disconnected from server: Connection to IRC server failed. 19:35:22 -!- GregorR-WC has joined. 19:35:25 <Asztal> ah, back to the compression error, looks like GregorR-L is in the lead 19:35:26 <GregorR-WC> HEY LOOK I WINS 19:37:04 -!- GregorR-WC has quit (Client Quit). 19:39:12 -!- DeeWC has joined. 19:39:16 <Deewiant> Amazing. 19:39:38 <DeeWC> Looks like a decent UI, too. 19:39:42 -!- DeeWC has left (?). 19:41:31 <Warrigal> Actually, I know precisely where this David slowed stuff comes from. 19:42:15 <Warrigal> But I have to go, so I'll tell you later. 19:45:40 <ehird> I think I'll implement the lambda calculus with IO in OCaml. 19:45:45 <ehird> as a compiler. 19:46:32 <ehird> 16:35:49 <pikhq> ... People care about that? 19:46:32 <ehird> 16:36:25 <pikhq> Thought that at least on the Internet we were sane, reasonable people. 19:46:44 <ehird> pikhq: why would that stop us making jokes about gayness? 19:46:51 <ehird> i make jokes about plenty of things i'm fine with 19:47:16 <pikhq> ehird: That's because you're only one step removed from 'lawl cockfag'. 19:47:38 <ehird> pikhq: amusing but no. 19:47:42 <ehird> i just have silly phases. 19:47:54 <pikhq> Lawlz. 19:48:49 <ehird> god, I hate touchscreens 19:49:03 <comex> ehird: so why the iphone 19:49:28 <ehird> because i hadn't used it before buying it and put my faith in the holy jobs to make it work for me 19:49:30 <ehird> which was stupid 19:49:44 <ehird> well i used it for like 3 minutes 19:49:51 <ehird> on a smudgy display copy 19:49:52 <GregorR-L> But ... but ... it's Apple! D-8 19:50:08 <ehird> the iphone does touchscreens a lot better than any other device i've seen but it's still shit 19:50:34 <ehird> the current technologies are fundamentally (a) incredibly prone to get smudgy as fuck, (b) hard to manipulate smoothly and (c) completely untactile 19:50:37 <ehird> so fuck that shit 19:51:05 <comex> I hear the 3GS is better at smudginess 19:51:28 <ehird> how 19:51:35 <ehird> are you contractually obligated to clean it thoroughly every week? 19:52:21 <comex> the screen is oleophobic 19:52:35 <ehird> it doesn't like japanese oreos? 19:52:38 <comex> I have no idea how much of an effect this actually causes 19:52:39 <ehird> :D 19:52:44 <GregorR-L> ehird: lol 19:52:50 <ehird> [[According to Electricpig, the iPhone 3G S will be Oleophobic - which does not mean “scared of cookies”. ]] 19:52:51 <ehird> damn 19:52:53 <ehird> I'm so unoriginal 19:53:08 <ehird> "Apple coated the new iPhones screens' in a special material that repels the oil on your hands that leaves greasy fingerprints." 19:53:12 <ehird> that could be cool for keyboards and mice 19:54:16 <ehird> oh, and 19:54:18 <ehird> Original: June 29, 2007[1] 19:54:18 <ehird> 3G: July 11, 2008[2] 19:54:19 <ehird> 3G S: June 19, 2009[3] 19:54:25 <ehird> Buy an iPhone and you get to replace it every year! 19:54:33 <comex> ++ 19:54:47 <comex> anyway I can't decide whether I want the 3GS or Pre 19:54:50 <comex> not that I'm likely to get either one 19:54:51 <comex> but 19:54:52 <ehird> comex: G1 bitch 19:54:52 <comex> ;p 19:54:54 <comex> no 19:54:56 <comex> G1 is java shit 19:55:29 <ehird> comex: meh. it's got an open source linux-based OS, it's got a real fucking keyboard, and it's not locked down at all so you can fuck with everything and distribute shit yourself 19:55:40 <comex> only if you have the dev version 19:55:44 <comex> and also 19:55:48 <comex> the linux fork is crap 19:55:49 <comex> and also 19:55:49 <comex> JAVA 19:55:53 <ehird> comex: rooting the non-dev one takes like 3 seconds 19:55:56 <comex> admitted 19:56:00 <ehird> comex: also the dalvik vm is way less shit than hotspot for this 19:56:04 <comex> *more 19:56:06 <comex> it doesn't JIT 19:56:06 <comex> at all 19:56:10 <comex> last I heard 19:56:15 <ehird> comex: so? anyway, you can run C shit 19:56:17 <ehird> it just isn't supported 19:56:31 <fizzie> The best thing is that Android uses their own VM (what was it called again, delvik?) so they can't use the Jazelle "java-acceleration" tricks the ARM cpu would have. 19:56:32 <ehird> write your prog in $blah and write a ui in jython or something 19:56:47 <ehird> fizzie: Dalvik, mr Im Using AnMaster's IRC Client So I Have No Backlog. 19:56:52 <comex> the pre also uses a linux-based OS, the apps are unobfuscated javascript even if the source isn't officially published, and can be rooted in 5 seconds 19:56:57 <ehird> Even For Things That Were Said Three Seconds Ago :P 19:57:08 <comex> and javascript >>>>> java 19:57:11 <ehird> comex: but it's not an open platform still. also, "EVERYTHING IS A WEB APP LOLZ" is totally shit 19:57:20 <ehird> it severely limits the cool shit youc an do. 19:57:20 <comex> ehird: execpt it's apparently really fast, even with lots of open apps 19:57:29 <comex> and I'm sure people will figure out how to run c stuff 19:57:47 <comex> see, I still prefer the iPhone, where apps are actually native code so you don't have to fight an interpreter 19:57:54 <comex> but at least with javascript, you know it's shit 19:57:56 <fizzie> ehird: I'm using LimeChat right now, and the Dalvik message had not appeared when I wrote that. Dialup, you know. 19:58:02 <ehird> comex: the g1 can run C shit. 19:58:05 <ehird> thus it can run anything 19:58:14 <ehird> yes the apis are java, just write a damn wrapper 19:58:21 <comex> java is just good enough that everyone pretends it's a real programming language 19:58:22 <ehird> fizzie: ha 19:58:25 <comex> so everything is slow 19:58:34 <ehird> comex: what you forget is that the G1 isn't java-based 19:58:35 <ehird> it's JVM-based 19:58:48 <comex> you want to use clojure? 19:59:03 <ehird> comex: no 19:59:08 <ehird> comex: you can use $any_jvm_language, or just write your fast logic in $native_lang and a UI in $jvm_lang 19:59:08 <ehird> etc 19:59:30 <ehird> so it doesn't really limit you, worst case? make a wrapper program for the jvm and call it from your c program 19:59:34 <ehird> if you really need to intwine that shit so much 20:00:48 <ehird> comex: also, the G1 has a real keyboard and a real joysticky thing and real buttons 20:00:51 <ehird> discussion over 20:01:14 <ehird> a real keyboard with real fucking modifier keys. 20:01:39 <ehird> ooh, the joystick is actually a trackball 20:01:44 <ehird> that's even better 20:02:11 <fizzie> The trackball got good reviews, I think. 20:02:45 <ehird> trackballs absolutely rock for games 20:02:57 <ehird> (not so much for general meesing...) 20:03:46 -!- kar8nga has joined. 20:04:59 <comex> ehird: 20:05:01 <comex> the pre has a real keyboard 20:05:18 <ehird> comex: so it does, but it's smaller 20:05:31 <ehird> comex: and lacks separate numeric/punctuation keys 20:05:40 <ehird> so a slight win for the G1 on that ponit 20:06:11 <ehird> the G1 has an autofocus camera and the pre doesn't seem to 20:06:17 <ehird> although this is really nitpicking :P 20:06:35 <ehird> (if I actually took photos it'd be nice because my hands are jittery as fuck) 20:08:16 <ehird> comex: wrt the pre "Also, the temptation to keep many apps running can outstretch the device's ability to handle them. I quickly learned to dread the dialog box informing me that I could not open a new app without removing some of my open cards." 20:08:45 <ehird> comex: ah, ew: 20:08:45 <ehird> "The keyboard's too small for thumb typing (welcome to fingernail typing) " 20:08:52 <ehird> awful 20:23:09 -!- CESSMASTER has joined. 20:23:48 <Sgeo> Is the Google phone good? 20:24:20 <fizzie> Google has said that before the end of 2009 there should be 18 Android phones available; the rumours have it that at least some of those have keyboards. Here's hoping. The Samsung I7500 they're actually almost-selling already is a touchscreen-only thing. 20:24:33 <ehird> Sgeo: Google phone = G1. 20:24:35 <ehird> fizzie: The G1 has a keyboard. 20:24:41 <ehird> Sgeo: Well, that's the first one. 20:24:45 <ehird> Also the best so far. 20:24:48 <Sgeo> Is it a good phone? 20:24:50 <Sgeo> Ah 20:24:54 <ehird> Yes. 20:24:59 <ehird> I've just spent ages arguing that it is to comex :P 20:25:18 <fizzie> I know, but others could have it too. The G1 looks a bit dated. 20:25:35 <Sgeo> It's T-Mobile, or is Google failing? 20:25:40 <fizzie> I would've been tempted by the Dev Phone, if it weren't for the >$100 shipping costs to Finland. 20:25:43 <Sgeo> failing me 20:25:48 <ehird> Sgeo: Google made the operating system........... 20:25:56 <ehird> And worked with HTC on the hardware. 20:26:08 <ehird> T-Mobile sell it. 20:26:13 <Sgeo> ok, ty 20:26:16 <ehird> You can also buy an unlocked Dev Phone from Google, which is it. 20:26:22 <ehird> fizzie: What specifically is dated about it, anyway? 20:26:35 <fizzie> They sell it as "HTC Dream" in some locations. (And they don't sell it at all in Finland.) 20:27:03 <ehird> 528mhz ARM is faster than the iPhone [3G] (but not S), it has the regular screen, it has a regular camera, it has 192MB of RAM... Android is being actively developed... 20:27:12 <ehird> I don't really see why it's "dated". 20:27:13 <fizzie> ehird: Just the looks. And the I7500 has a better camera (well, spec-wise; haven't seen real reviews yet) if one cares about that sort of thing. 20:27:18 <ehird> It only came out in October. 20:27:40 <fizzie> October last year, though! In iPhone-years that's something like twelveteen generations old. 20:27:46 * pikhq still wants OpenMoko. 20:28:01 <fizzie> Have to go, hourly dialup charges and so on. 20:28:12 <ehird> pikhq: OpenMoko the OS is no more open than Android, really; and it looked kind of shit tbh 20:28:41 <ehird> Heck, the Neo phones can run Android. 20:28:53 <ehird> So you could probably run OpenMoko on an Android phone, though why you'd want to I have no clue. 20:28:59 <pikhq> But Qt! 20:28:59 <pikhq> :P 20:29:03 <ehird> "As of the 2008.08 software revision only the GSM phone module is working reliably" 20:29:10 <ehird> pikhq: yeah, I loooooooooooooove X11. 20:29:17 <ehird> not 20:29:28 <pikhq> Qt embedded runs directly on the framebuffer. 20:30:07 <ehird> pikhq: "The software development is split between Qtopia under X11" 20:30:19 <pikhq> ... WHY ARE THEY RUNNING QTOPIA ON X? 20:30:21 <ehird> Maybe that's an option, but OpenMoko runs it under X11. 20:30:23 <pikhq> STAB! 20:30:26 <ehird> pikhq: because they have gtk stuff too 20:30:35 <ehird> "The Openmoko community has forked [4] the final stable release into Qt Extended Improved and they have launce a new homepage [5]. " 20:30:37 <ehird> Awesome. Forks! 20:30:39 -!- CESSMASTER has quit (Read error: 104 (Connection reset by peer)). 20:30:40 <pikhq> Which should be running on DirectFB. 20:30:51 <pikhq> Okay, so the OpenMoko community is retarded. 20:31:01 <pikhq> Android FTW, then. 20:33:16 <ehird> Clearly what we need is perfect wearable computers for free. 20:33:20 <ehird> Step 1. Establish communist state 20:33:27 <ehird> Step 2. Force clever people to develop thus 20:33:30 <ehird> Step 3. Give to everyone 20:33:32 <ehird> Step 4. ??? 20:33:40 <ehird> Step 5. NO PROFIT BUT LAPTOPS AND PHONES ARE OBSOLETE 20:33:45 <ehird> Step 6. — Comrade. 20:33:58 <pikhq> Step 1. Make Reprap be able to print useful electronics. 20:34:09 <pikhq> Step 2. Post-scarcity. 20:34:14 <pikhq> Step 3. ??? 20:34:18 <ehird> Step 4. Oops, entropy. 20:34:19 <pikhq> Step 4. Singularity! 20:34:28 <ehird> Step 5. Simultaneous step-threading is outlawed. 20:34:41 <ehird> Step 5.1. BASIC line numbering adopted for steps. 20:34:46 <pikhq> ... How does entropy get into this? 20:34:48 <ehird> Step 5.2. Space is filled. 20:34:55 <ehird> Step 6. Singularity expands space. 20:34:58 <ehird> Step 7. (etc.) 20:35:03 <ehird> pikhq: I was just being the silly. 20:35:07 <pikhq> Ah. 20:35:23 <ehird> Step 8. Wars between universes fighting for space, energy 20:35:35 <ehird> Step 9. Goddammit, the multiverse is finite too. 20:35:45 <ehird> Step 10. God is petitioned to give us infinite stuff kthxbai. 20:35:54 <ehird> Step 11. God, by way of killing us and sending us all to heaven, complies. 20:35:57 <ehird> Step 12. Well, bugger. 20:35:58 <ehird> ~fin~ 20:36:18 <ehird> I hereby decree the above The Story of Wasted Effort. 20:36:59 * pikhq runs off, muttering something about "Forward the Singularity!" 20:37:55 <ehird> pikhq: Is there some wacky theological argument that equates the singularity to the second coming of christ? 20:37:59 <ehird> I would enjoy that 20:38:51 <pikhq> I'd imagine so. 20:39:18 <ehird> [[If it goes up to your belly button I WANT WANT WANT YOU! http://305f8.easyurl.net]] 20:39:20 <ehird> — Twitter spam 20:39:24 <ehird> My legs just aren't that flexible... 20:40:17 <ehird> pikhq: Actually, how does that even work? *SINGULARITY, EVERYONE IS HAPPY* "Hey guys! I'm Jesus and I've come to save you from your sufferi—" "WOOHOO PARTY SINGULARITY WOO" "—okay, call me back when you're in the mood." 20:40:19 <ehird> :D 20:42:17 <pikhq> Jesus as the spiritual manifestation of the singularity. 20:42:28 <pikhq> :P 20:42:57 <ehird> That's some metaphorical bible-reading :P 20:43:31 <pikhq> Yes. Yes it is. 20:44:07 <Asztal> the book of revelations gives a finite size for heaven 20:44:12 <ehird> pikhq: So, wait. 2000 or so years ago, a singularity was born? Somehow. :P 20:44:30 <ehird> Asztal: It's Revelation, not Revelations. 20:45:50 <pikhq> It's "The Revelation of John", not the Book of Revelations. :P 20:46:21 <ehird> pikhq: It's "Revelation to John", actually. 20:46:32 <pikhq> Right, right. 20:50:51 <Asztal> aha, it's a cube, 2.211 megametres on each size :) 20:51:01 <Asztal> side, that is 20:51:28 <ehird> Asztal: That's rather gigantic. 20:51:35 <ehird> Although not really. 20:51:51 <lament> should be big enough for all mankind 20:51:52 <ehird> Asztal: So, uh, _how_ is every good dead person ever meant to fit in there? :-) 20:51:57 <lament> ehird: very easily? 20:52:02 <lament> with lots of room to spare 20:52:05 <ehird> Are you sure? 20:52:12 <lament> absolutely 20:52:25 <ehird> pikhq: Hey! In Christianity, are good non-humans generally agreed to go to heaven or not? 20:52:40 <lament> some estimate from the internet: 106,456,367,669 20:52:44 <lament> (number of people that ever lived) 20:52:44 <ehird> If it's all good creatures, I'm gonna guess that 2.2 megametres isn't quite big enough. 20:52:47 <pikhq> ehird: There is generally no stance on the issue amongst Christians. 20:52:53 <ehird> lament: does that include neanderthals etc? 20:53:11 <ehird> i mean, defining what a human is is very arbitrary 20:53:16 <lament> not sure, but it starts from 50,000 BC 20:53:29 <pikhq> It's like asking for Christianity's opinions on the laws of physics. :P 20:53:30 <lament> adding all those guys would not change the count dramatically 20:53:37 <lament> there weren't very many of them 20:54:02 <pikhq> Oh, right. People do that. 20:54:07 <ehird> pikhq: err, it's a theological matter about who gets into heaven, innit 20:54:07 * pikhq facepalms at idiots 20:54:16 <lament> ehird: anyway, if we take it to be that number 20:54:29 <lament> ehird: we can put everyone on the *bottom* of the cube, with 40m^2 per person! 20:54:36 <pikhq> And it's a theological matter that I am unaware of any opinions on the matter at all. 20:54:38 <lament> we're barely using *any* space at all 20:54:39 <ehird> lament: remember that people need space to move around, and presumably there's shit other than people in heaven 20:54:48 <ehird> hmm 20:54:50 <ehird> i guess 20:54:59 <lament> ehird: yes, but the *entire cube* is empty 20:55:14 <ehird> it depends what attractions there are in heaven, apart from the whole "eternal gaze of the holy spirit" boring shit :) 20:55:16 <lament> you don't really appreciate how big 2 million meters is 20:55:21 <lament> especially when it's cubed 20:55:34 <ehird> lament: in more reasonable terms, that's 1,000 miles 20:55:37 <pikhq> What's the diameter of the Earth? 20:55:37 <ehird> which is big but not gigantic 20:55:48 <ehird> `wolfram diameter of earth 20:55:54 <HackEgo> diameter of earth \ \ Input interpretation: \ \ Earth \ Result: \ \ diameter \ \ 7913.1 miles \ Unit conversions: \ \ 12 734.9 km kilometers \ \ Generated by Wolfram|Alpha (www.wolframalpha.com) on June 22, 2009 from Champaign, IL. © Wolfram Alpha LLC—A Wolfram Research Company \ \ 1 \ \ 20:56:10 <ehird> pikhq: so it's (earth.diameter/7)^3 20:56:20 <lament> ehird: in that case, you don't appreciate how few humans are there and how little space they occupy :) 20:56:28 <pikhq> Which is a fuckton of volume. 20:56:29 <ehird> lament: I suppose 20:56:37 <lament> ehird: we could put all humans currently living on an island of about 30x30km 20:56:47 <ehird> lament: really? 20:57:15 <lament> hm, maybe not 20:57:20 * lament rechecks the math 20:57:28 <ehird> we are talking 6bn people here 20:57:59 <lament> more like 80x80 km 20:58:01 <lament> 1m per person 20:58:16 <lament> (which is very sparse as crowds go) 20:58:28 <ehird> that's impressive. 20:58:31 <lament> 30x30 is reasonable when it's an actual crowd 20:58:52 <lament> there's a sci-fi story where some guy gets a revelation 20:58:57 <pikhq> So, how much area would we have in this cube? 20:58:58 <lament> that some sort of end of the world is coming 20:59:00 -!- CESSMASTER has joined. 20:59:13 <lament> and they should build a huge cube and put everybody on earth inside this cube 20:59:18 <ehird> lament: is it called the bible 20:59:19 <lament> and people believe him and proceed to build the cube 20:59:21 <lament> no 20:59:22 <ehird> :p 20:59:31 <lament> the cube was really small 20:59:34 <ehird> lol 20:59:44 <lament> i don't remember, either 1 or 10 km 20:59:48 <lament> probably 1? 21:00:03 <lament> nope 21:00:14 <lament> math says, 6km on the side if it's 1m^3 per person 21:00:14 <pikhq> According to Wikipedia, the Earth has 148 square terameters of land. Can we beat that with a 2 megameter cube? 21:00:18 <pikhq> :P 21:00:39 <lament> the book had all sorts of logistic problems 21:00:49 <pikhq> (man, terameters. I <3 metric) 21:00:55 <lament> the cube was built in china because that's where most of the population is anyway, so it's easier to transport people to the cube 21:01:08 <lament> but the chinese got pissed off because they were the first to be frozen, and suspected a conspiracy 21:01:26 <lament> but then they did it anyway because everybody trusted the prophet guy 21:01:31 <ehird> lol 21:01:42 <ehird> this is silly :) 21:01:46 <ehird> lament: so what happens 21:02:06 <lament> i probably shouldn't spoil it :) 21:02:15 <lament> if i remember the name of the book at least 21:02:18 <ehird> lament: DOIT 21:02:22 <Asztal> hm, that measurement just refers to New Jerusalem, but I don't know what's outside of that 21:02:24 <ehird> DOIT DOlT 21:02:27 <ehird> hurf geddit 21:02:29 <ehird> doit dolt? 21:02:35 <Asztal> by the way, it's made of gold, so it would probably weigh quite a lot too 21:03:07 <lament> don't remember the name :( 21:03:08 <pikhq> Asztal: To simulate gravity, of course. 21:03:10 <ehird> Asztal: what, heaven? 21:03:11 <pikhq> s/simulate/replicate/ 21:03:12 <pikhq> :P 21:03:15 <ehird> that's all god can do? gold? 21:03:22 <ehird> not like, solid enlightenment? 21:03:24 <ehird> that's just lame 21:03:28 <Asztal> http://isv.scripturetext.com/revelation/21.htm 21:03:42 * pikhq wants uranium 21:03:42 <ehird> christianity is so disappointing!~ 21:03:45 <ehird> s/~$// 21:03:49 <Asztal> (As I just said, apparently the measurement is only the city, I don't know what the rest really is) 21:04:11 <ehird> this new jerusalem does have a lot in common with the original 21:04:15 <ehird> like being a gigantic gold fucking cube 21:04:25 <ehird> Asztal: [[The Tabernacle (Hebrew משכן, mishkan, "residence" or "dwelling place"), according to theHebrew bible/Old Testament, was the portable dwelling place for the divine presence from the time of the Exodus from Egypt through the conquering of the land of Canaan]] 21:04:30 <ehird> so it's not actually heaven 21:14:14 <GregorR-L> More like an RV for God. 21:14:26 -!- GregorR-L has quit ("Leaving"). 21:14:29 -!- GregorR-L has joined. 21:21:19 -!- KingOfKarlsruhe has joined. 21:23:54 -!- Arrogant has joined. 21:34:44 -!- ais523 has quit (Remote closed the connection). 21:35:41 -!- SimonRC has joined. 21:36:57 -!- coppro has joined. 21:40:34 -!- kar8nga has quit (Remote closed the connection). 21:50:29 -!- oerjan has joined. 21:57:30 -!- Arrogant has quit ("Leaving"). 21:58:39 <oerjan> <AnMaster> hash tables are prone to DDoS. But are balanced trees really immune to that? 21:59:08 <oerjan> i think balanced trees should be O(log n) worst case insertion, if that's what you're worrying about 21:59:11 <AnMaster> oerjan, for implementing an asoc array that is. 21:59:16 <ehird> let rec compile e = 21:59:16 <ehird> match e with 21:59:17 <ehird> Var x -> "(fun _ -> " ^ scrub x ^ "())" 21:59:19 <ehird> | Lam (x,e') -> "(fun _ -> fun " ^ scrub x ^ " -> " ^ compile e' ^ ")" 21:59:21 <ehird> | App (p,q) -> "(fun _ -> (((" ^ compile p ^ ") ()) " ^ compile q ^ "))" 21:59:22 <oerjan> yes, i thought so 21:59:23 <ehird> ;; 21:59:25 <ehird> Fun fun. 22:00:29 <oerjan> AnMaster: i don't know about the hash DDOS thing but i would guess it's something like if someone knows the hashing scheme they could force collisions and make it very slow? 22:02:05 <oerjan> ehird: ^ is ocaml concatenation? 22:02:11 <ehird> ya 22:03:08 <AnMaster> oerjan, indeed 22:03:45 <oerjan> for balanced trees in themselves that should not be possible, but what about long keys with common prefixes? 22:03:56 <AnMaster> oerjan, however providing sorted input seems to slow it down, since even with a self-balancing tree it ends up having to do a lot more rotations. 22:04:03 <oerjan> that would at least require checking linear in length 22:04:07 <AnMaster> oerjan, that would be a prefix tree? 22:04:23 <SimonRC> ah, but is comparison constant-time? 22:04:42 <oerjan> um no i mean just inserting long string keys into a tree-based dictionary 22:04:45 <AnMaster> oerjan, which is not the variant I'm using. Since using a prefix tree for storing integers would be rather odd 22:04:47 <SimonRC> ordering will be linear in key length in the worst case, right? 22:05:02 <AnMaster> oerjan, ah no, this is a sparse integer -> value map 22:05:09 <oerjan> ic 22:05:22 <oerjan> but not very large integers? 22:05:29 <oerjan> or? 22:05:32 <AnMaster> oerjan, 64 bit ones, so not very 22:05:37 <AnMaster> definitely not bignum, 22:05:40 <AnMaster> bignum* 22:05:49 -!- jix has joined. 22:05:57 <oerjan> btw haskell has a special map type for int keys that is faster 22:06:30 <AnMaster> and usually the range just above 0 ("just" ~ 15000 or so) is much denser 22:06:41 <AnMaster> no negative values 22:06:46 -!- tombom has joined. 22:06:59 <ehird> AnMaster: how about telling us the use case... 22:07:12 <oerjan> AnMaster: what language? 22:07:38 <AnMaster> ehird, you didn't ask, but then it is easy: For in-between. And I was planning an online version,. 22:07:41 <AnMaster> version* 22:07:42 <oerjan> because i recall haskell also has a function to build from ordered associations 22:07:44 <AnMaster> like llvm has one 22:07:56 <ehird> i didn't ask because it's not my discussion 22:08:00 <ehird> AnMaster: that's not the use case 22:08:05 <AnMaster> but it seems to have very poor performance when inserting lots of sorted keys in the dictionary. 22:08:09 <ehird> that's not what you're going to be using it for, that's just what you're going to use it in 22:08:18 <AnMaster> dictionary is based on AVL tree 22:08:20 <AnMaster> to be specific 22:08:49 <AnMaster> and use case, well, it is to track "initial memory is all zero" 22:09:25 <ehird> # compile (Lam ("x", Var "x"));; 22:09:25 <ehird> - : string = "(fun _ -> fun _x -> (fun _ -> _x()))" 22:09:34 <AnMaster> every time a non-zero is found it is marked as "modified" 22:09:38 <ehird> Hooray! Though I note that (fun _ -> _x()) could just be _x. 22:09:42 <AnMaster> as well as converting stuff where possible 22:09:49 <AnMaster> like +++ at the start to set 3 22:09:59 <ehird> # compile (Lam ("x", Var "x"));; 22:09:59 <ehird> - : string = "(fun _ -> fun _x -> _x)" 22:10:00 <Slereah_> Fun! 22:10:01 <ehird> Hooray! 22:10:01 <AnMaster> anyway. 22:10:03 <Slereah_> I love fun 22:10:10 <ehird> So do I. 22:10:10 <AnMaster> the problem is gccbf output 22:10:19 <AnMaster> it goes like this: 22:10:39 <AnMaster> >+*72 22:10:39 <AnMaster> >>+ 22:10:39 <AnMaster> >>+ 22:10:39 <AnMaster> >+ 22:10:39 <AnMaster> >+*101 22:10:41 <AnMaster> >>+ 22:10:43 <AnMaster> >>>+ 22:10:45 <AnMaster> >+*108 22:10:47 <AnMaster> at the start 22:10:49 <AnMaster> and so on 22:10:52 <AnMaster> for about 13 000 lines. 22:11:10 <AnMaster> a bit less, but close 22:11:40 <AnMaster> if I shuffle the code to set those in a random order, thus adding lots more >>, the code is much faster. 22:11:53 <ehird> # let foo = Lam ("x", (App (Var "x", Var "x")));; 22:11:53 <ehird> val foo : expr = Lam ("x", App (Var "x", Var "x")) 22:11:54 <ehird> # compile foo;; 22:11:56 <ehird> - : string = "(fun _ -> fun _x -> (fun _ -> (((_x) ()) _x)))" 22:11:58 <ehird> # compile (App (foo, foo));; 22:11:59 <AnMaster> thanks to profiling I managed to pin point it to the tree rotations 22:12:00 <ehird> - : string = 22:12:02 <ehird> "(fun _ -> ((((fun _ -> fun _x -> (fun _ -> (((_x) ()) _x)))) ()) (fun _ -> fun _x -> (fun _ -> (((_x) ()) _x)))))" 22:12:05 <AnMaster> oerjan, ^ 22:12:05 <ehird> Hooray! 22:12:14 <AnMaster> oerjan, so question is, how to avoid this worst case. 22:12:23 <oerjan> AnMaster: from wikipedia: "The AVL tree is more rigidly balanced than Red-Black trees, leading to slower insertion and removal but faster retrieval." 22:12:37 <oerjan> so it may be an intrinsic fault of AVL trees 22:12:49 <AnMaster> oerjan, so in other word, lets use some other dict/set than the erlang stdlib in module for it. 22:12:55 -!- augur has quit (Read error: 110 (Connection timed out)). 22:12:57 <oerjan> oh 22:13:06 <AnMaster> oerjan, I also test for membership every iteration 22:13:23 <AnMaster> to see "is this not in set, if so, lets turn the add into a set" 22:13:53 <AnMaster> but might be worth trying some other sort of tree. 22:14:12 <oerjan> is DDOS actually a relevant problem for your use case? 22:14:25 <AnMaster> oerjan, as I mentioned above, I was planning an online compiler 22:14:30 <AnMaster> somewhat like llvm has 22:14:36 <AnMaster> as a show case 22:14:38 <AnMaster> just for fun 22:14:42 <oerjan> otoh erlang may not be good for hash tables, since they are mutable 22:15:05 <AnMaster> oerjan, I guess I could go down and dirty and use ets tables... Not sure how much I would gain that wya 22:15:06 <AnMaster> way* 22:15:11 <AnMaster> but maybe worth trying 22:15:32 <ehird> - : unit -> unit -> (unit -> 'a as 'a) = <fun> 22:15:37 <oerjan> AnMaster: i guess your keys are not necessarily inserted in order, so you cannot optimize just for that? 22:15:37 <ehird> ok, that's _not_ the type i was expecting. 22:15:39 <AnMaster> not as nice interface of course. 22:15:45 <AnMaster> oerjan, they might not be indeed. 22:16:27 <ehird> specifically, I was expecting: 22:16:29 <AnMaster> oerjan, but at some point they will be. 22:16:30 <ehird> unit -> 'a as 'a. 22:16:38 <ehird> or unit -> (unit -> 'a as 'a) 22:17:12 <oerjan> ehird: those are all equivalent aren't they 22:17:16 <AnMaster> oerjan, with that I mean that I sort instructions by offset, but other passes might enable more to be sorted. Like eliminating dead code in between. 22:17:24 <ehird> oerjan: unit -> unit -> (unit -> 'a as 'a) isn't 22:17:29 <AnMaster> oh and the pass that builds this set even. 22:17:33 <ehird> oerjan: I think 22:17:36 <oerjan> ehird: i would think it is 22:17:37 <AnMaster> that can enable better sorting in some cases. 22:17:43 <ehird> oerjan: well, I don't think it should generate that anyway 22:17:51 <ehird> I think it should generate one of what I said 22:17:56 <oerjan> if 'a = unit -> 'a, then it all simplifies 22:18:05 <ehird> well, yes 22:18:12 <ehird> but my compiler shouldn't be producing that, I think 22:18:15 <ehird> oh well, I'll try church numerals 22:18:27 <AnMaster> ehird, what are you compiling to/from? 22:18:34 <ehird> AnMaster: lambda calculus → ocaml -rectypes 22:18:39 <AnMaster> interesting 22:18:40 <ehird> (-rectypes because the lambda calculus is not well-typed.) 22:18:47 <oerjan> ehird: i mean, unit -> (unit -> 'a as 'a) = unit -> 'a as 'a afaict 22:18:55 <ehird> (essentially, any lambda-expression is (a = a -> a)) 22:18:56 <AnMaster> ehird, optimising compiler? 22:19:05 <ehird> AnMaster: no; ocaml does the optimization 22:19:14 <ehird> after all, well-written ocaml is competitive with C-speed 22:19:19 <ehird> oerjan: yes 22:19:24 <ehird> oerjan: but are you not listening to me? 22:19:29 <ehird> my compiler isn't architectured in a way to produce that type 22:19:31 <ehird> as far as I know 22:19:34 <ehird> thus I suspect I may have a bug 22:19:44 <ehird> although there isn't really much room for buggery: 22:19:45 <ehird> let rec compile e = 22:19:46 <ehird> match e with 22:19:48 <ehird> Var x -> scrub x 22:19:50 <ehird> | Lam (x,e') -> "(fun () -> fun " ^ scrub x ^ " -> " ^ compile e' ^ ")" 22:19:52 <ehird> | App (p,q) -> "(fun () -> (((" ^ compile p ^ ") ()) " ^ compile q ^ "))" 22:19:54 <ehird> ;; 22:19:58 <AnMaster> ehird, of course, but couldn't there be some stuff that you can know better at that point than ocaml can. Like stuff you can exploit, but which is not useful in most ocaml code. 22:20:18 <ehird> AnMaster: Dude, it's the lambda calculus. As far as semantically rich languages go, it's one of the least. 22:20:19 <AnMaster> There certainly is stuff like that for bf for example, no idea if there is for lambda calc 22:20:30 <ehird> "Oh! We're applying a function, to ANOTHER FUNCTION!" 22:20:37 <AnMaster> true 22:20:37 <ehird> "Ooh, but this time, we're getting the function from our argument." 22:20:45 <ehird> "Ooh, here we have a literal function. That applies a function to a function! Twice!" 22:20:58 <AnMaster> ehird, you could possibly optimise written out church numerals 22:20:59 <AnMaster> maybe 22:21:12 <ehird> eh, but why bother 22:21:16 <AnMaster> kay 22:21:20 <ehird> I'm just getting reacquainted with ocaml 22:21:36 <AnMaster> ehird, but it would be similar to optimising [-] into set for bf. common idiom. 22:21:45 <oerjan> ehird: what lambda expression did you compile to get that type? 22:21:47 <AnMaster> not sure if it would help much in your case 22:21:55 <ehird> oerjan: # let foo = Lam ("x", App (Var "x", Var "x")) in compile (App (foo, foo));; 22:22:07 -!- tombom has quit ("Peace and Protection 4.22.2"). 22:22:29 -!- Sgeo has quit ("Ex-Chat"). 22:22:43 <SimonRC> where is the topic from? 22:22:48 <oerjan> ehird: right, App (Var "x", Var "x") gives a recursive type of course 22:22:49 <AnMaster> this code was generated by gcc-bf: <+>*6-><+> 22:22:53 <AnMaster> silly 22:22:53 <ehird> oerjan: yes, of course 22:23:04 <ehird> SimonRC: http://codu.org/davidslowed/; a meme due to Warrigal 22:23:43 <ehird> hmm 22:24:04 <ehird> oerjan: is it actually possible to force one of these functions through to completion? i don't think a loop that keeps doing (f ()) would type, even with -rectypes 22:25:21 <oerjan> ehird: if f : unit -> 'a as 'a then f () () () ... is well-typed, surely 22:25:52 <ehird> oerjan: actually, we can't write a "force f" function, as we have no way of knowing when we didn't reduce any 22:25:52 <oerjan> but you can never get anything but 'a out 22:26:02 <ehird> so this compiler, as-is, is useless 22:26:07 <ehird> well 22:26:16 <ehird> with my IO model, we can just force until we get eof 22:26:20 <ehird> but uhhhhhhhhh 22:26:26 <ehird> that would be odd 22:26:39 <ehird> as i think it wouldn't terminate when a regular LC forcer would 22:27:02 <oerjan> oh well 22:27:29 <ehird> oerjan: is that 'oh, well' or 'oh well' :P 22:27:42 <oerjan> what's the difference? 22:27:55 <ehird> oerjan: 'oh, well, you can fix that by doing X' or 'oh well... life sucks...' 22:28:07 <oerjan> the latter :( 22:28:33 * ehird puts oerjan out of his misery 22:28:38 <ehird> NOW I CAN TAKE OVER THE WORLD 22:28:39 <ehird> MWA HA HA 22:28:47 <oerjan> YOU CAN HAVE IT 22:29:06 <ehird> oerjan: Shut UP you're DEAD. 22:29:21 <oerjan> THAT'S WHY I'M TALKING IN ALL CAPS, STUPID 22:29:28 <ehird> OH 22:30:31 <coppro> I thought Death talked in caps, not Dead 22:30:46 <ehird> coppro: DEATH TALKS IN SMALL-CAPS. 22:30:51 <ehird> THE DEAD TALK IN CAPS. 22:30:51 <oerjan> IT SORT OF RUBS OFF 22:31:26 <coppro> oerjan wins; ehird is disqualified due to IRC not having small caps 22:31:30 <coppro> also due to not being dead 22:31:38 <ehird> IRC so has small caps; Unicode does. 22:31:41 <oerjan> MWAHAHAHA 22:31:49 <oerjan> WHOOPS 22:31:59 <oerjan> OH NO, NOT THE UNICODE. PLEASE, NOT THAT 22:32:55 <oerjan> ---------------------O- 22:33:00 <oerjan> HMPH 22:33:36 * oerjan HOPED myndzi WAS USEFUL FOR HANGMAN 22:33:37 * myndzi HOPED oerjan WAS USEFUL FOR HANGMAN 22:34:04 <SimonRC> hm 22:35:01 <SimonRC> e 22:35:33 <coppro> ehird: incorrect 22:35:43 <ehird> coppro: No, I'm right. 22:35:47 <AnMaster> <ehird> IRC so has small caps; Unicode does. <-- example 22:35:49 <coppro> nicode does not have a complete set of small caps 22:35:53 <coppro> *Unicode 22:35:57 <ehird> It does for the Latin a-z. 22:36:00 <coppro> no 22:36:08 <AnMaster> because I'm pretty sure that is a font thing, not a encoding thing 22:36:13 <coppro> it has all but F, I, Q, S, X 22:36:17 <ehird> Well, whatever. 22:36:22 <coppro> because they exist for other purposes (like IPA) 22:36:24 <ehird> FIQSX are for lame people. 22:36:41 <coppro> AnMaster: Unicode defines a number of differently-formatted characters 22:36:48 <oerjan> <ehird> I'm so unoriginal <-- YOU MEAN UNOLIGINAL 22:36:50 <SimonRC> ITYM "are phor lame people" 22:37:11 <AnMaster> coppro, yes I know the wide ones. I disagree with them. 22:37:20 <ehird> SimonRC: Methunk you mean "Methunk you mean" 22:37:24 <coppro> AnMaster: more than that 22:37:25 <AnMaster> yes I know it tries to be compatible 22:37:39 <AnMaster> coppro, yes ^2 and such. Which is again a font issue. 22:37:44 <AnMaster> or typography issue rather 22:37:45 <ehird> AnMaster: they have a good justification; you're just blind to the culture requiring them 22:37:46 <ehird> meh 22:37:48 <ehird> why do i bother 22:37:50 <coppro> once again, more than that 22:38:03 <AnMaster> ehird, what culture require the superscript 2 one? 22:38:22 <ehird> /sigh 22:38:24 <coppro> x ∈ ℤ 22:38:26 <SimonRC> its from latin-1 22:38:41 <SimonRC> unicode is a bit messy, bit it could be much worse 22:38:55 <coppro> agreed 22:39:09 <AnMaster> coppro, Z != ℤ though iirc. 22:39:21 <AnMaster> I mean, there is actually a difference in meaning 22:39:43 <AnMaster> I don't know of any such for superscript 2, or ∈ (another math symbol) 22:39:49 <ehird> oerjan: maybe i could add an argument mentioning whether we reduced 22:39:50 <ehird> meh 22:39:51 <coppro> excuse me? 22:39:56 <AnMaster> assuming that ∈ is what I think it is 22:40:05 <coppro> ∈ is element of 22:40:07 <ehird> you think we don't need an ∈ character? 22:40:07 <ehird> what>? 22:40:13 <ehird> you're a retard. 22:40:14 <AnMaster> coppro, then so it was 22:40:32 <AnMaster> ehird, no I don't... 22:40:35 <oerjan> ehird: SURE, EXCEPT YOU MEAN RESULT I THINK 22:40:36 <coppro> E and ∈ are different 22:40:42 <ehird> oerjan: er rite. 22:40:45 <AnMaster> ehird, you are a retard if you think I think so 22:40:57 <ehird> 22:39 AnMaster: I don't know of any such for superscript 2, or ∈ (another math symbol) 22:40:58 <AnMaster> coppro, they are. 22:41:02 <AnMaster> more than just formatting 22:41:22 <ehird> clearly coppro was confused at whaty ou meant as well 22:41:25 <AnMaster> ehird, it seemed to be stylised ∈ here. 22:41:32 <AnMaster> I blame my font 22:41:35 <ehird> o_O? 22:41:45 <coppro> probably your default font doesn't have it, so your client picks another 22:41:48 <AnMaster> it looks kind of italics in Dejavu Sans Mono 9 22:41:52 <coppro> as a result you see it differently 22:41:54 -!- inurinternet has joined. 22:42:04 <AnMaster> coppro, if dejavu doesn't have it, then I'm very surprised. 22:42:43 <coppro> අ 22:42:49 <coppro> Is that in DejaVu for you? 22:45:38 -!- augur has joined. 22:46:25 <Warrigal> 21:42 < coppro> à¶[+spoony ] [ Firenze ] [ Keiya ] [ Rachel ] 22:46:28 <Warrigal> Strange. 22:46:58 <coppro> I have no clue what it was, I just picked a random character that degraded to Unifont 22:46:58 <Warrigal> I think I need to set my lunch.normish.org locale. 22:47:13 <coppro> how come we always get into Unicode discussions? 22:48:18 <AnMaster> <coppro> අ <-- [0D35]? 22:48:28 <coppro> sounds right 22:49:00 <AnMaster> coppro, dejavu doesn't have everything, but it does have ∈. And it looks right in Dejavu Sans Mono 12, just not in 9 22:49:04 <AnMaster> so rendering issue. 22:49:05 <AnMaster> :) 22:49:07 <coppro> ah 22:49:11 <AnMaster> mystery solved. 22:49:31 <Warrigal> There, I set my locale. 22:49:57 <Warrigal> It will be effective either next time I log in or next time I reboot. If the latter, I'll have to get used to the old one. :-P 22:52:08 * SimonRC remembers the killer character someone on here put together 22:52:19 <SimonRC> it have like 80 combining modifiers on it 22:52:33 <AnMaster> Warrigal, ååååå 22:52:44 <SimonRC> it kept crashing screen for me (fixed in a newer version of screen) 22:53:09 <AnMaster> SimonRC, why does screen care about it 22:53:34 <AnMaster> it doesn't need to actually render it, that is up to the terminal emulator 22:53:35 <SimonRC> because screen can do charset interpretation 22:53:42 <SimonRC> and screen has to know how wide it is 22:53:47 <AnMaster> hm ok 22:56:24 -!- CESSMASTER has quit ("☃☃☃☃☃☃☃☃☃☃☃☃☃☃☃☃☃☃☃☃☃"). 22:56:34 <ehird> 22:46 Warrigal: I think I need to set my lunch.normish.org locale. 22:56:36 <ehird> wuzzu lunch 22:56:44 <Warrigal> ehird: say that again? 22:56:58 <ehird> wuzzu = what's that 22:57:10 <Warrigal> lunch.normish.org is my lunchhub.com development server. 22:57:12 <ehird> just grok the meaning from glorked context 22:57:16 <Warrigal> And my IRC bouncer. 22:57:31 <ehird> "Lunch Hub is a place where friends and co-workers can speed up the decision of where to go to lunch. This site makes it easy by laying out all the options and letting you announce where you want to go. " 22:57:33 <Warrigal> I'm going to install LAMP on it :-D 22:57:38 <ehird> that sure sounds like it needs a mega development server. 22:57:46 <oerjan> gliffle the gromp, fnibbles 22:57:46 <ehird> "the website you can literally eat at"? 22:57:47 <ehird> no 22:57:53 <ehird> you can't eat at the website 22:57:55 <Warrigal> Yeah. I had to skip college to pay the $20 a month. 22:57:58 <Warrigal> ehird: read it again. 22:58:07 <ehird> skip at 22:58:10 <ehird> it's even more wrong 22:59:21 <oerjan> now even more perfect 23:02:36 -!- GregorR-L has quit (Read error: 110 (Connection timed out)). 23:04:11 <ehird> Warrigal: You should switch to prgmr.com and get 1GB of RAM for the same price. 23:06:43 <Warrigal> I don't think the amount of RAM matters. 23:06:56 <Warrigal> I'll try to keep prgmr.com in mind, though. 23:07:19 <ehird> Warrigal: Also, 14GB more disk space . 23:07:32 <ehird> And 60GB more bandwidth. 23:07:37 <ehird> Warrigal: Also, IPv6. 23:08:00 <ehird> Warrigal: Also, faster CPU since it's less popular and they're Xen tweakers. 23:08:13 <ehird> Also, you can install any OS that Xen supports. 23:08:22 <ehird> Hmm, they don't seem to have IPv6 yet on their new servers, whatever 23:17:45 <AnMaster> p[1]+=-p[5]; <-- comments. 23:18:18 <SimonRC> yuk 23:18:27 <AnMaster> SimonRC, auto generated code though 23:18:35 <SimonRC> how about "p[1]-=p[5];"? 23:18:43 <oerjan> it has a nice symmetry, symbolizing the oppression of the lower castes 23:18:44 <AnMaster> SimonRC, it was p[1]+=255*p[5]; before 23:18:51 <SimonRC> heh 23:18:58 <SimonRC> ok 23:19:04 <AnMaster> SimonRC, do you think +=- is a improvement visually 23:19:06 <AnMaster> or not 23:19:15 <SimonRC> oh, yeah 23:19:17 <ehird> changing +=- into -= is trivial. 23:19:35 <AnMaster> ehird, except not: p[1]+=-p[5]+2*p[8]; 23:19:36 <SimonRC> also, zero tests come out correct, unline what could happen with 255* 23:19:40 <AnMaster> suddenly not so trivial 23:19:49 <AnMaster> you need to check there is only one single term after 23:20:03 <AnMaster> SimonRC, hm? 23:20:11 <SimonRC> isn't that easy since you are using the parse tree? 23:20:33 <AnMaster> SimonRC, actually this is in the form of a polynomial, represented as a list of coefficients. 23:20:42 <SimonRC> well, if you add 255 256 times, you get 0 in BF, but not in C 23:20:54 <AnMaster> SimonRC, true, that was 255 though 23:21:01 <AnMaster> SimonRC, and you do in C if you use unsigned char 23:21:07 <SimonRC> ah, ok 23:21:10 <AnMaster> assuming CHAR_BIT == 8 23:22:43 <AnMaster> SimonRC, also, at the point where it is easy to fix up, I don't have the assignment yet 23:22:55 <AnMaster> as in, I only have the bit after += or = or whatever there is 23:23:03 <AnMaster> I don't yet know if it is = or += 23:23:12 <AnMaster> god damn abstractions ;P 23:24:29 -!- jix_ has joined. 23:24:53 <ehird> Well, this is a new one. The threat of litigation to motivate people to do voluntary work. 23:25:34 <SimonRC> where? 23:26:36 <ehird> SimonRC: a patchset to GHC to make it compile stuff for the iPhone — http://upcycle.it/~blackh/iphone/LICENSE; summary (based on what they've said too): "We don't let people use this commercially yet so everyone focuses on integrating it with GHC!" 23:26:59 <ehird> a rather sour taste finds its way into my mouth via my screen. 23:30:47 <AnMaster> SimonRC, another result of this change: 23:30:49 <AnMaster> -p[1]=255 + 255*p[8]; 23:30:49 <AnMaster> +p[1]=-1 - p[8]; 23:31:05 <AnMaster> wait, isn't that bitwise not? 23:31:33 <SimonRC> I think so 23:31:35 <SimonRC> not sure 23:31:38 <AnMaster> yes it is 23:32:57 <oerjan> um those -+ at the beginning are diff marks, not algebra, i assume? 23:33:26 <AnMaster> oerjan, correct 23:33:39 <AnMaster> The marking of the diff? 23:36:32 -!- jix has quit (Read error: 113 (No route to host)). 23:42:14 <jix_> there is probably a functional brainfuck variant out there right? 23:42:31 <AnMaster> not sure 23:42:46 <AnMaster> procedural ones yes, but I don't know about functional ones 23:42:53 <ehird> yes, there's one 23:42:54 <ehird> but it's not tc 23:43:06 <AnMaster> that's some major messup 23:43:06 <jix_> i'm creating one right now... 23:43:15 <AnMaster> a non-tc tarpit... 23:43:19 <jix_> which is purely functional and tc 23:43:29 <AnMaster> jix_, how does it work 23:43:30 <ehird> how is that bf :p 23:44:32 <jix_> ehird: i started with the idea that if you take haskell... you could define +-<>., to be of type IO 23:44:35 <AnMaster> that is a good question too 23:44:51 <ehird> jix_: that's... very warped :P 23:44:58 <jix_> then you could define some test function of type IO -> IO -> IO 23:45:20 <jix_> which runs the first parameter on nonzero and the 2nd on zero 23:45:36 <ehird> erm. 23:45:42 <jix_> and then i started changing stuff from there on to avoid thousand usages of bind 23:45:44 <ehird> i think you have a very warped understanding of haskell :) 23:45:49 <ehird> but give some example code. 23:45:54 <ehird> it's not purely functional if you do everything in io 23:46:12 <jix_> ehird: the language itself is... 23:46:19 <ehird> yes, but that's just semantics 23:46:24 <jix_> in contrast to unlambda for example 23:46:34 <jix_> well then maybe that's the wrong word 23:47:08 <jix_> but you might have noticed that [ and ] aren't defind 23:47:24 <jix_> so it's not just a brainfuck template language 23:47:39 <jix_> as soon as you test a single cell.. you end up writing functional stuff 23:48:59 <AnMaster> jix_, how does that functional stuff look 23:49:15 -!- Corun has joined. 23:49:25 <jix_> for example to get a loop you'd need something like \x . x x 23:49:25 <AnMaster> code example 23:49:45 <AnMaster> where is that syntax from 23:49:49 -!- Corun has quit (Client Quit). 23:49:57 <ehird> AnMaster: lambda calculus. 23:50:00 <AnMaster> ah 23:50:03 <ehird> methinks jix_ is a bit confused about how all this works :) 23:50:12 <AnMaster> ehird, he isn't alone then 23:50:22 <jix_> ehird: in what way? 23:50:36 <ehird> jix_: well, \x. x x doesn't do anything in the LC 23:50:38 <ehird> it's inert 23:50:52 <jix_> ehird: well you have to apply it to something 23:51:54 <jix_> you could even use the bf part only for IO and do calculations using church encoded stuff 23:52:24 <ehird> so uh this is... haskell? 23:52:43 <jix_> huh? 23:52:49 <AnMaster> ehird, wrong, haskell with a syntax that is a mix of lambda and bf 23:52:52 <ehird> your language seems incredibly ill-defined... 23:53:00 <oerjan> this! is! haskell! 23:53:08 * AnMaster slaps oerjan 23:53:16 <jix_> ehird: maybe because i haven't written a complete spec 23:53:24 * oerjan swats AnMaster back -----### 23:53:25 <ehird> :\ 23:53:54 * AnMaster deflects the swat with his o=========E 23:54:02 <oerjan> darn 23:56:43 -!- KingOfKarlsruhe has left (?).