00:00:03 calamous: limit the number of files on the system? 00:00:10 eg it resets if you write more than 100 files 00:00:10 I tend to spend time in #xkcd, on the offchance that there's interesting things there.\ 00:00:45 Though the smaller channels on Foonetic are far better; signal/noise ratio and all that. 00:00:57 night 00:01:20 bsmntbombdood: i'm sure Deewiant will be here sometime :P 00:02:23 !sh for i in {1..99999}; do ln -s /home/egobot/egobot.hg/foo /tmp/input.$i; done 00:02:30 !sh echo hi 00:02:30 hi 00:02:34 MUAHAHAHAHA 00:02:36 calamous: it took an hour or so 00:02:38 * calamous disappears. 00:02:39 to run out 00:02:43 Deewiant: bsmntbombdood wants to know how to use two monitors 00:02:45 I have done that 00:02:46 ehird, Oh, shoot :P 00:02:49 !sh ls /tmp/input.* | wc -l 00:02:52 it is simple 00:02:53 !sh ls /tmp/input.* | wc -l 00:02:54 !sh ls /tmp/input.* | wc -l 00:02:54 1128 00:02:54 /bin/bash: line 1: :ehird!n=ehird@208.78.103.223: command not found 00:02:55 you connect two of them 00:02:58 then turn both on 00:02:58 LOL WAT 00:03:00 and set it up 00:03:06 OK, I made it delete them at the end, but that's clearly not good enough. 00:03:09 AnMaster: fail 00:03:18 ehird, fail how 00:03:19 calamous: just limit the number of writes to 100 files 00:03:24 ehird, it was simple. 00:03:27 bsmntbombdood: tell AnMaster how he failed :-P 00:03:28 nvidia-settings 00:03:31 turn on twin view 00:03:35 ehird, You can only limit the number of total open files. 00:03:36 that was all 00:03:44 calamous: To 100, then. 00:03:52 fail how... 00:04:04 er i misread AnMaster 00:04:05 forget that 00:04:18 ehird, The number of open files is limited to a very low number. 00:04:19 you can do it in X config too 00:04:27 ehird, The problem is that it doesn't keep them open after it writes them. 00:04:28 which is better if they are different height 00:04:30 * calamous leaves. 00:04:32 calamous: Ah. 00:04:36 having different height is confusing 00:04:40 trust me 00:06:46 AnMaster: do you think http://www.newegg.com/Product/Product.aspx?Item=N82E16814187037 can handle dueling monitors 00:07:02 the specs seem to imply no 00:07:26 no clue 00:07:39 I did it on a GeForce 7600 GS 00:08:20 seems to cost less, so I assume the 9400 GT is better 00:08:30 for values of better equal to still not a good card but bsmntbombdood doesn't want a good card :P 00:09:25 meh 00:09:46 ehird, I used two TFTs, one vga only and one dvi +vga 00:09:53 so one went on dvi the other on vga 00:44:39 my current monitor is vga so that would probably work 00:56:34 "Intel(R) Core i7 PC's from £7" 00:56:38 —cut off advert 00:57:48 seven pounds!? what a rip off! 01:11:57 ok azureus is waaay to bloated 01:12:08 * bsmntbombdood looks at rtorrent 01:12:34 utorrent is pretty good 01:13:51 utorrent is windoze isn't it? 01:14:02 no 01:14:04 theres a mac versio 01:14:05 n 01:14:59 ... 01:15:13 bsmntbombdood: Transmission 01:16:22 OR RTORRENT 01:16:34 Or transmission 01:16:38 Which is minimal and awesome and <3 01:16:45 transmission is cool too 01:20:46 -!- oerjan has quit ("Good night"). 01:21:56 * bsmntbombdood wonders if he should actually buy ehird's computer 01:22:02 why not 01:22:12 you want an upgrade, I got it for the price you wanted, prophet? :-P 01:25:37 bsmntbombdood: for the same budget, the only real other options there are is getting an amd processor instead, but that'd just be sacrificing speed and you'd have some of the budget just lying around doing nothing 01:25:56 so if you'd upgrade to that budget anyway, it's pretty much what you'd come out with 01:26:00 why did you prefer i7? 01:26:08 bsmntbombdood: over what? 01:26:09 amd? 01:26:11 ja 01:26:27 bsmntbombdood: amd's top processors are competitive with core 2 -- the LAST generation of intel processors 01:26:34 i7 has the yummy integrated memory 01:26:38 so a lot less latency 01:26:44 and it has some nice architechtural improvements 01:26:52 more speed clock-for-clock under a lot of situations 01:27:25 i7 >>> amd phenom black edition 2 electric boogaloo > core 2 01:29:11 I was under the impression that AMD's top processors were competitive with some of the lesser i7 chips, making them great if you're on a budget. 01:29:19 But eh, whatever. 01:29:33 pikhq: i haven't seen that; but the i7 was a 2.93ghz 01:29:41 so not exactly a lesser i7 chip 01:30:03 AMD's existence for most of the company's history *has* depended on being cheaper than Intel chip, so *shrug*. 01:30:03 there's only 3.2ghz and 3.33ghz (the latter i haven't seen sold anywhere) above that 01:30:12 yeah 01:30:24 pikhq: but that integrated memory controller + ddr3 = definite yum 01:30:33 ehird: you forgot a cpu cooler too 01:30:37 bsmntbombdood: err 01:30:42 the i7 comes with it i'm pretty sure 01:31:10 i've seen people say "stock cooler" referring to i7s 01:31:10 ehird: Integrated memory controller? I've had one for 3 years, and it was old then. :p 01:31:10 so 01:31:17 pikhq: er, that's not what i meant 01:31:25 pikhq: "On-die memory controller: the memory is directly connected to the processor. It is called the uncore part and runs at a different clock (uncore clock) of execution cores. " 01:31:38 20-40% less memory latency has been reported 01:31:46 bsmntbombdood: "Cooling DeviceHeatsink and Fan included " 01:31:50 oh ok 01:31:53 from the spec page of the i7 01:32:03 And AMD's done that since they introduced AMD64. Your point? :p 01:32:11 pikhq: i'm not sure that's the same thing 01:32:32 eh, whatever 01:33:05 Wikipedia says it is. 01:33:13 hmm looks like the same thing; still what i've read puts the i7's memory access faster 01:33:40 I expect AMD will come out with an i7 competitor _sometime_ 01:33:46 but atm the i7 is the best choice imo 01:33:55 Can't say I blame Intel for doing it; it *is* a rather easy way to get better memory access speeds. 01:35:04 bsmntbombdood: if you do go with it will you go with the x25-m? i'd recommend it if you do have the $300 extra budget (hmm, I distinctly recall asking this before...) 01:35:48 i think i'll only get one hdd and the x25m 01:36:01 bsmntbombdood: yeah, that sounds wise 01:36:25 and i've got a 500gb drive laying around 01:37:02 bsmntbombdood: raid-1ing that will leave you with only 500gb of space 01:37:15 yes, i won't raid them 01:37:24 ah 01:37:31 Hmm. No wonder the i7's somewhat higher performance; it's the newer architecture. 01:37:43 pikhq: Durr, Nehalem's the whole point of an i7 :P 01:37:54 bsmntbombdood: newly calculated total inc. ssd: $1689.91 01:38:52 ehird: what makes you think that mobo is a good choice? 01:39:06 bsmntbombdood: first, I went with the stock Intel mobo for i7 01:39:07 BUT 01:39:10 then i realised it had only 3 ram slots 01:39:14 so I searched newegg 01:39:25 and that was a high-rated one that's the same as the intel one but w/ 6 ram slots 01:40:27 bsmntbombdood: they don't seem to have really changed the intel mobo apart from adding the more slots and sth 01:40:34 not many pci slots 01:40:42 bsmntbombdood: how many do you need? 01:40:53 ah 01:40:56 this one looks bette 01:40:57 r 01:41:07 hm wait 01:41:08 Cons: As noted in previous reviews, the BIOS needs an update out of box before you can install triple channel RAM. I loaded one stick of RAM and installed windows 01:41:16 that sounds silly 01:42:02 * ehird searches newegg for x58 01:42:08 bsmntbombdood: but how many pci slost are you gonna use? 01:42:25 i would like to play with some coproccesors in the future 01:42:38 tile64 etc 01:42:40 PCI Express 2.0 x16: 3 x PCIe 2.0 x16 (at x16/x16/x4 mode) 01:43:00 bsmntbombdood: http://www.newegg.com/Product/Product.aspx?Item=N82E16813131359 01:43:25 * ehird amends his spec document to include these 01:43:38 anyway i gotta go 01:43:41 bye 01:44:06 lawl 01:44:10 bsmntbombdood: the mobo actually saves money 01:45:18 bsmntbombdood: for when you return: http://pastie.org/467175.txt?key=ijtxvejybwknzzyr4yea 01:45:25 total $1679.91 01:49:24 My conclusion: You can build a top-top of the range PC (sans video card) for around $1.6k. 01:49:41 Or around £1,100. 01:53:13 And the video card isn't really worth it unless you're playing Crysis on 3 different monitors. 01:53:40 pikhq: Weeeeeeeeeeeeel, that GeForce 9400 GT seems rather the shit as far as cards you're actually gonna use goes. 01:53:49 It's just a stand-in for the missing on-board video on the mobo for this machine. 01:54:33 " it ran between 65 and 72 deg C" 01:54:34 hmm 01:54:39 i wonder if that's under large load 01:54:41 because that's a hot card 01:55:08 Eh, as far as graphics cards go, I'm not all that picky. 01:55:22 the reviews seem ok 01:55:41 heh 01:55:49 bsmntbombdood: the gfx card has an audio card built in 01:55:54 that apparently is better than onboard video 01:56:56 My current standards: able to handle Source engine games and able to render 1080p h264 in realtime. And the latter becomes "Can handle 1080p" if my CPU's fast enough to decode 1080p h264 in realtime... 01:57:21 My standards: Be the best fanless card you can get. 01:57:25 :-P 01:57:27 Futureproofing! 01:57:39 (atm that's the radeon hd 4850) 01:57:42 Completely different standards, but likely to get similar results. 01:57:56 pikhq: no way, the radeon 4850 can do much more than that 01:58:06 it's the third best single (non-X2) card ATI offer 01:58:08 True, true. 01:58:10 4850, 4870, 4890 01:58:23 i wish newegg had a uk version 01:58:33 I think it's the best card I could actually see myself bothering to purchase, though. 01:58:46 Much beyond that seems... Silly. 01:59:31 pikhq: Heck, even with the 4870X2 -- the best card ATI sells -- Crysis at 19xx-xxxx (I forget what the exact res is) with all settings on high or very high only runs at 20 or so fps 01:59:40 Games are wicked demanding 02:00:19 And I play games about a generation behind what's current. 02:00:41 So, it'll be another 5 years or so before I get around to playing Crysis. :p 02:01:08 Crysis kind of looks shit :P 02:01:31 Ah, but it's very expensive shit. 02:01:37 And you're thinking 1920x1080. 02:01:38 Verily. 02:01:41 And yes. 02:01:42 AKA 1080p. ;) 02:03:13 hmm 02:03:32 maybe i should build my own system, 'cept i hate thermal paste 02:04:17 → for today :) 02:04:18 Worth it. 02:04:23 pikhq: maybe, maybe 02:04:39 I'd prefer to pay more to get a silent system from people who know what they're doing in that area, prolly 02:04:39 → 02:04:46 I build my systems as a matter of habit... 02:05:11 Of course, I upgrade my systems piecemeal, so not exactly what you're dealing with. 02:05:27 * pikhq has a couple of 10 to 15 year old parts 02:16:54 -!- pikhq has quit (Read error: 104 (Connection reset by peer)). 02:17:58 -!- pikhq has joined. 02:30:58 -!- cherez has joined. 02:31:02 -!- cherez has left (?). 02:33:00 !c printf("hello\n"); 02:33:01 hello 02:34:06 -!- Gracenotes has joined. 02:46:43 -!- puzzlet has joined. 02:47:26 -!- puzzlet_ has quit (Remote closed the connection). 02:49:04 !c printf("addr of main %p\n", main); 02:49:05 addr of main 0x4004cc 02:49:08 !c printf("addr of main %p\n", main); 02:49:09 addr of main 0x4004cc 02:50:04 !c printf("addr of entry point %p\n", __start); 02:50:04 Does not compile. 02:50:22 Fik. Different entry point. XD 02:50:38 !c printf("addr of entry point %p\n", _start); 02:50:39 Does not compile. 02:50:39 ? 02:50:48 !c printf("addr of entry point %p\n", start); 02:50:49 Does not compile. 02:50:54 Sure I'll get it one of these days. 02:52:38 Huh. main *is* the entry point. 03:11:08 -!- puzzlet_ has joined. 03:24:10 -!- puzzlet has quit (Read error: 110 (Connection timed out)). 03:24:11 ehird: i forgot, i need an optical-disc drive also 03:27:25 OK, now I have to think of a way to fix the shared /tmp/input.* problem 03:35:48 !sh ls -l /tmp 03:35:49 /bin/ls: /tmp: Function not implemented 03:35:56 !sh ls /tmp | xargs echo 03:35:56 input.14903 03:35:58 !sh ls /tmp | xargs echo 03:35:59 input.14936 03:36:03 OK, no more shared /tmp 03:36:15 your shit's weird 03:36:45 ? 03:41:57 -!- Gracenotes has quit (Remote closed the connection). 03:41:58 -!- Gracenotes_ has joined. 03:42:00 -!- Gracenotes_ has changed nick to Gracenotes. 03:57:32 !sh ls /tmp 03:57:33 input.15061 03:57:40 !sh ls -l /tmp 03:57:40 /bin/ls: /tmp: Function not implemented 03:58:13 What the fnuck is it doing, I wonder... 03:59:47 ls stats or something, Idonno 04:00:04 ls -l that is 04:00:35 !daemon psh sh sh 04:00:35 Daemon psh running. 04:00:40 !psh echo hi 04:00:43 !psh echo hi 04:00:43 hi 04:00:47 !psh echo foo 04:00:47 hi 04:00:47 foo 04:00:52 That was weird :P 04:01:00 !psh echo foo 04:01:01 foo 04:05:04 !psh mkdir -v /tmp/home 04:05:04 /bin/mkdir: created directory `/tmp/home' 04:05:11 !psh cd -v /tmp/home 04:05:11 /bin/sh: line 6: cd: -v: invalid option 04:05:17 !psh cd /tmp/home; pwd 04:05:18 /tmp/home 04:05:25 !psh export HOME=/tmp/home 04:07:58 -!- bsmntbombdood has quit (Connection reset by peer). 04:08:33 -!- bsmntbombdood has joined. 04:09:12 !psh wget http://google.com/ 04:09:12 --2009-05-04 03:09:12-- http://google.com/ 04:09:13 Resolving google.com... failed: Name or service not known. 04:09:22 !kill psh 04:09:22 Daemon psh killed. 04:10:33 -!- bsmntbombdood has quit (Read error: 104 (Connection reset by peer)). 04:11:06 -!- bsmntbombdood has joined. 04:12:58 -!- bsmntbombdood has quit (Read error: 104 (Connection reset by peer)). 04:21:44 -!- bsmntbombdood has joined. 04:46:47 So, I guess I'm trying to find out where Firefox keeps its profiles under Windows. 04:56:34 somewhere in %userprofile%\Local Settings\Application Data\Mozilla\, iirc. 05:01:14 Correct. 05:02:12 Users/Fozzie/Local Settings is empty. 05:02:22 No, those forward slashes are not necessary. 05:03:16 This is Vista. 05:08:23 who uses windows anyway 05:09:35 I'm not using Windows right now; I want to share a profile between Windows and Linux. 05:33:01 -!- pikhq has changed nick to pikhq_. 05:33:11 -!- pikhq_ has changed nick to pikhq. 05:50:37 -!- bsmntbombdood_ has joined. 05:58:12 -!- bsmntbombdood has quit (Read error: 110 (Connection timed out)). 05:58:56 -!- bsmntbombdood_ has changed nick to bsmntbombdood. 06:01:29 Wow. YouTube under Linux here sounds really fun. 06:01:57 i don't have a problem with youtube 06:02:05 Good for you. 06:03:56 what did you mean then? 06:04:21 I meant that it sounds very slow and choppy. 06:04:33 I just recorded an hour-long audio file. It took a couple seconds. 06:05:02 how about other video websites 06:05:31 could be a problem with flash plugin 06:06:46 A problem that also exists with Sound Recorder? 06:07:26 Here, I'll click "record" again. 06:08:09 It says it's been recording for four hours. 06:08:27 no, tell me it's not the alsa problem again 06:08:56 This problem appears to be entirely distinct from the problem I had before. 06:09:11 Now it says it's been recording for ten hours. 06:09:18 Maybe some other people have had this same problem. 06:10:40 This is why I don't generally use Linux. 06:38:02 i haven't used windows for like 4 yearss 07:06:08 In my experience, the people who have the ever-dwindling set of hardware that Linux doesn't support are the people who think it has terrible hardware support :P 07:06:23 (Which is to say, that sounds like a problem with the sound driver) 07:34:29 you mean...the people who have problems with linux are the ones who have problems with linux? 07:39:26 -!- GreaseMonkey has joined. 07:59:59 -!- clog has quit (ended). 08:00:00 -!- clog has joined. 08:07:46 -!- lereah_ has joined. 08:12:19 -!- Deewiant has quit (Read error: 60 (Operation timed out)). 08:21:09 -!- fizzie has quit (Read error: 148 (No route to host)). 08:38:35 -!- Deewiant has joined. 08:53:11 -!- nooga has joined. 08:57:17 -!- WangZeDong has joined. 09:05:02 -!- puzzlet has joined. 09:14:33 -!- Slereah has quit (Read error: 110 (Connection timed out)). 09:17:38 -!- puzzlet_ has quit (Read error: 110 (Connection timed out)). 09:27:07 -!- nooga has quit (Read error: 104 (Connection reset by peer)). 09:57:49 -!- puzzlet_ has joined. 10:10:46 -!- puzzlet has quit (Read error: 110 (Connection timed out)). 10:25:43 -!- fizzie has joined. 10:31:14 -!- GreaseMonkey has quit ("YES -> thor-ainor.it <- THIS IS *DELICIOUS*!"). 10:38:03 -!- tombom has joined. 10:42:16 OK, now I have to think of a way to fix the shared /tmp/input.* problem 10:42:17 err 10:42:21 mkstemp 10:42:28 that avoids race conditions too 10:44:37 kerlo, what sound recorder? 10:44:41 audacity? 11:01:21 -!- calamous has quit ("Leaving"). 11:54:00 AnMaster: Gnome Sound Recorder. 11:54:15 no clue about that. Since I don't use Gnome 11:54:26 I would blame it on gnome probably 11:54:49 try audacity, it is a good sound recording and editing program. 11:56:21 * kerlo reboots under the alibi of figuring out where the Firefox profile resides 12:16:01 Success! 12:16:53 Windows still won't make any connections to the Internet, but I made a symbolic link from a Firefox profile on Linux to one on Windows, and it worked quite magically. 12:20:46 I just set the profile in Firefox to point to /mnt/... 12:33:58 why dual boot. The only reason I can think of is massive 3D games. 12:34:36 for most other windows only stuff you can usually use either wine or virtualbox/vmware-server/whatever 12:41:26 -!- impomatic has joined. 12:41:48 Hi -) 12:42:26 The #corewars channel will be moving to freenode if anyone is interested. 12:42:38 By the way, is anything happening with BF Joust? 13:44:29 -!- puzzlet has joined. 13:45:16 -!- puzzlet_ has quit (Remote closed the connection). 14:26:39 -!- tombom has quit (wolfe.freenode.net irc.freenode.net). 14:26:39 -!- EgoBot has quit (wolfe.freenode.net irc.freenode.net). 14:26:39 -!- Leonidas has quit (wolfe.freenode.net irc.freenode.net). 14:26:39 -!- comex has quit (wolfe.freenode.net irc.freenode.net). 14:26:39 -!- GregorR has quit (wolfe.freenode.net irc.freenode.net). 14:26:43 -!- tombom has joined. 14:26:43 -!- comex has joined. 14:26:43 -!- EgoBot has joined. 14:26:43 -!- Leonidas has joined. 14:26:43 -!- GregorR has joined. 15:03:33 -!- FireFly has joined. 15:06:38 -!- tombom has quit ("Peace and Protection 4.22.2"). 15:20:34 -!- Gracenotes has quit (Remote closed the connection). 15:22:02 -!- Gracenotes has joined. 15:27:48 -!- bsmntbombdood has quit (Read error: 104 (Connection reset by peer)). 15:48:04 -!- bsmntbombdood_ has joined. 15:57:43 -!- ais523 has joined. 16:08:55 hi bsmntbombdood_ 16:09:54 bsmntbombdood_: so optical drive eh 16:09:57 I'm sure i can do that :P 16:11:28 -!- lereah_ has quit (Remote closed the connection). 16:11:32 bsmntbombdood_: shall I throw in a floppy drive? some firmware upgrades still come on those. 16:11:48 Floppies are cheap... 16:12:03 duh 16:13:26 -!- Dewi has joined. 16:15:44 bsmntbombdood_: http://pastie.org/467702.txt?key=7xpp0vdnf1ukje2vbjdq 16:16:00 $1709.39, same as before but w/ dvd/cd-reader/writer and floppy drive 16:18:44 bsmntbombdood_: btw, graphics cards are useful for vector computation etc as well as just graphics... 16:19:10 -!- impomatic has quit (Read error: 110 (Connection timed out)). 16:23:29 !accumulate test 16:23:35 meh, what happened to it? 16:23:40 ais523: i did !accumulate junk 16:23:51 then AnMaster did !accumulate ehird being annoying again, or something 16:23:56 and we went at it for a few hours 16:23:58 or thereabouts 16:24:01 ah, ok 16:24:26 result: AnMaster's ignoring me, keeps saying I'm immature, ignores the fact that it was 50% him, and also that if we were arguing over junk it's obviously a topic we're discussing :D 16:24:28 a lot less than that. You killed it every time. 16:24:38 that ignore lasted a long time there AnMaster 16:24:41 ehird, GregorR made the bot ignore you too 16:24:46 a few hours perhaps 16:24:54 AnMaster: yes, to break the chain 16:30:39 "Ubuntu To Ship With Super Talent SSDs" ← I totally thought this meant that ubuntu cds would come with an ssd, not the otherway around 16:36:26 I killed daemons. 16:36:36 because there wasn't a decent use for them? 16:36:36 Because there was some issue with them leaving cruft behind I haven't resolved yet. 16:37:03 There's no decent use for most of what EgoBot supports, but I don't want a chicken with its head cut off taking 100% CPU on Codu :P 16:37:19 yep, makes sense 16:37:40 Anyway, they'll probably come back when I can fix some bugs. 16:37:48 Fair enough. 16:37:55 Don't kill Codu with Egobot. ;) 16:38:10 Or alternatively, if I can figure out a way to use chkpt to checkpoint between lines, then I can actually have daemons only run while in use. 16:40:54 -!- iano has joined. 17:08:41 -!- jix has joined. 17:08:49 ehird: if you are interested in esotope-bfc, it now supports exact loop counting for nontrivial deltas. 17:08:57 nice 17:09:02 example code/output? 17:09:11 http://pastie.org/private/4z2i9vo3y8kviwet42fma 17:09:42 i took that example from bfdb. 17:10:16 One of the most publicized changes in Windows Me, was that it no longer included real mode MS-DOS.[3] With real-mode support removed, Windows Me can boot up a couple of seconds faster, without loss of Windows functionality. Autoexec.bat and Config.sys are no longer executed during startup by IO.SYS, and the system cannot boot to a MS-DOS command prompt or exit to DOS when Windows has booted. Because of this, applications that needed real mode DOS to run, 17:10:18 such as older disk utilities, did not run under Windows Me. 17:10:20 err 17:10:22 if ((mptr[1]%2) != 0) { 17:10:24 lifthrasiir: that's hot. 17:10:26 is that really general? 17:10:32 yes. 17:10:43 ais523: take a look at that, it's awesome 17:10:52 lifthrasiir: congrats, your BF interp is the est 17:10:58 *best 17:11:00 *compiler 17:11:15 ehird: Uh, lies. Windows ME shipped with real mode MS-DOS. *However*,, it was not enabled by default. 17:11:17 not yet, i realized that generalization while looking at bfdb ;) 17:11:31 it translates every possible case of while(mptr[k]!=x) { ... mptr[k]+=delta; }. 17:11:40 pikhq: Source: WP 17:11:48 lifthrasiir: bfdb? 17:11:48 Wikipedia lies, then. 17:11:55 pikhq: I used ME :-P 17:12:04 http://djm.cc/dmoews.html FYI. 17:12:06 I turned on real-mode DOS in ME. 17:12:07 ;) 17:12:36 lifthrasiir: never heard of it 17:12:44 it needs a fix to compile in recent gcc, but it seems worth looking 17:12:48 I'd like to see how esotope-bfc.py handles gcc-bf output, actually 17:13:20 presumably it isn't yet good enough to deduce that I'm simulating a hardware stack and a stack of frame pointers? 17:13:39 Probably not. 17:13:40 it cannot handle array operations yet, and i think that is quite challenging 17:14:15 lifthrasiir: try compiling with clang, not gcc 17:14:17 anyway, if you are interested see http://hg.mearie.org/esotope/bfc/file/tip/esotope-bfc.py for recent version. 17:14:22 it may help 17:14:34 ehird: compile c++ code with clang? 17:14:37 is that stable? 17:14:43 that's not c++, lifthrasiir 17:14:44 I mean your output 17:14:51 but yes, clang's c/c++/obj-c support is mostly okay 17:14:55 ah i see. 17:16:54 in other news, it turns out that MS Office's new native ODF support is abysmal 17:17:03 for instance, it strips out formulas when loading spreadsheets 17:17:24 oh well, i should keep using openoffice. 17:18:07 openoffice is so bloated 17:18:19 17:17 ehird: Does clang/llvm do the optimization of changing some array positions into variables? like, {int x[2];x[0]=1;x[1]=2;return x[0]+x[1]} becoming {int x0,x1;x0=1;x1=2;return x0+x1} 17:18:20 17:18 sabre: yes 17:18:21 17:18 sabre: ehird: yes 17:18:23 17:18 ehird: sabre: yay! 17:18:26 lifthrasiir: yep, use clang. 17:18:41 excellent! 17:19:39 @openoffice.org users: try abiword & gnumeric :P 17:19:54 gnumeric's ODF support is also awful, I think 17:20:02 but then, it doesn't claim to do it correctly 17:20:07 it looks like they've only just started implementing it 17:20:14 true enough 17:20:16 I mostly meant abiword 17:20:22 but then I thought people might complain about spreadsheets 17:20:26 so i added gnumeric 17:20:52 ais523: openoffice is pretty bad though; it doesn't pick up fonts in ~/.fonts for instance IME 17:20:58 which is really irritating 17:21:06 bloody portability :) 17:21:16 openoffice manages to not follow the conventions on any system 17:21:21 the difference is, on Windows, I'm glad it doesn't 17:21:41 I always thought that openoffice was Java because it was so badly integrated... but Java is more integrated and less ugly! 17:21:42 ais523: agreed, especially for fonts. 17:21:48 I don't even know how they did it 17:21:56 openoffice is I suspect mostly written in Java 17:22:08 or by people who write in the same style, anywa 17:22:09 *anyway 17:22:12 well, it's a Sun project 17:22:14 so yes 17:22:18 yes 17:22:19 but it doesn't use Swing 17:22:23 and I don't think it uses SWT 17:22:29 *AWT? 17:22:30 or if it does use either, it hacks them weirdly 17:22:32 ais523: no, SWT 17:22:34 it's what eclipse uses 17:22:37 AWT is deprecated 17:22:45 oh, I like AWT anyway, though 17:22:53 beats swing 17:22:54 it's easily the best of the windowing systems Java has come up with so far 17:22:59 that I nkow of 17:23:00 *know of 17:23:08 AWT = Java is a programming language 17:23:10 Swing = Java is an OS 17:23:28 personally I think treating Java as a programming language makes more sense 17:23:57 ais523: you can make swing use native widgets 17:24:01 not native text rendering, though 17:24:45 native-widget swing would be an improvement, certainly 17:24:46 I did a bit of SWT once; it wasn't too shabby. It's pretty much just a wrapper for native GUI toolkits over JNI; except of course using the SWT abstractions. 17:24:48 how is it achieved? 17:24:56 ais523: JNI 17:24:59 that's java's c ffi 17:25:06 well, not much of an ffi, you have to tailor your code to it 17:25:08 -!- jix has quit (Remote closed the connection). 17:25:11 c extension interface 17:25:24 well, not specifically C 17:25:29 but it'd be a pain to use anything else 17:25:37 -!- jix has joined. 17:25:39 it has some c++ stuff i think though 17:26:01 someone link Java to Befunge-98 via C and INTERCAL 17:26:03 to hell with portable GUI libs, anyway 17:26:04 you know you want to 17:26:06 Yes, in C++ code you can pretend you're calling Java classes, or something. 17:26:08 you can't automate UI guidelines 17:26:24 you need to have a separate UI for each platform if you want to have a good UI 17:26:53 the Gnome/KDE mirrorflip is one of the most annoying things about that 17:27:15 ais523: gnome/kde's minor compared to anything_on_linux vs OS X 17:27:22 you basically have to completely forget your other UI to go either way 17:27:31 i get using a portable toolkit if you just want something that works 17:27:43 but just use Tk or something; that's really easy to use, it's pig ugly most of the time though 17:27:48 really? the most annoying thing for me when I had to use OS X on someone else's laptop for a bit was muddling the apple and control keys 17:27:50 but that doesn't matter if you just want something that works 17:27:53 and the lack of pgup/pgdn 17:27:57 as a separate key 17:27:57 It integrates well with everything but Linux. 17:28:02 pikhq: nope 17:28:05 tk is horrid on os x 17:28:07 it's fine on windows 17:28:12 so it was really keyboard layout confusion, rather than anything to do with the UI itself 17:28:19 ais523: for the user, there may not be much difference at first sight 17:28:25 Hrm, last time I used Tk on OS X, it looked like native widgets... 17:28:26 but the philosophy is massively different 17:28:30 pikhq: yes, to a degree 17:28:38 but half-heartedly doing something on OS X is a bit laughable 17:28:52 they pretty much only get buttons and checkboxes right 17:29:04 and integration goes far further than how the widgets look anyway 17:29:20 Hrm. Either I used some odd OS X-only fork, or it's gone downhill over time. 17:29:37 Because I remember it being, well, native. 17:29:38 pikhq: Or you're not an obsessive-compulsive OS X user (redundant) that notices things :-) 17:29:48 For instance, it uses a white background and the window has no padding by default. 17:29:56 And text boxes are totally non-native 17:30:01 I'm picky about things looking right. ;) 17:30:03 I suspect it of merely rendering things that look like native widgets 17:30:07 for me, it's not widget look that matters, but behaviour 17:30:08 not actually using them 17:30:14 ais523: yep, tk gets both wrong 17:30:22 e.g. on Gnome I can right-click on a scrollbar tab to scroll to the other end 17:30:32 that doesn't work anywhere else I've tried, but it's really useful 17:30:38 I *distinctly* remember a pinstripe background (it was old OS X). 17:30:42 (as in, if I right-click on scroll-up, it scrolls ot the top) 17:30:47 pikhq: Hm, maybe. 17:32:44 It was some "Tcl/Tk Aqua" with batteries included Tcl distribution. Probably some fork to make it act right, then. 17:37:27 Nah, it's in mainline, must just have fallen behind 17:37:31 Or at least, it comes with OS X 17:41:02 haha 17:41:19 you know how in C, if you allocate a variable, it isn't initialized and you get whatever happens to be in memory at the time? 17:41:27 you can do that in Perl6 too, but you have to request it explicitly 17:41:33 HAH 17:41:37 That = awesomely bad 17:41:44 You can do that in D too 17:41:50 Explicitly, that is 17:41:51 is it the default, though? 17:41:56 No, not the default 17:42:01 No, in D you get zero'd stuff. 17:42:10 By default integers are zero, floats are nan, chars are all-bits-1 17:42:54 0xFE? That's an odd choice. 17:42:57 I do like the idea of lazy exceptions, though 17:43:04 and all-bits-1 = 0xFF 17:43:05 pikhq: 0xff 17:43:13 ais523: That's all bits. 17:43:22 pikhq: "All bits one" 17:43:29 not "all bits but one one" 17:43:35 I read that as all bits minus one. 17:43:45 I don't understand how "all bits" means a numeric value :-P 17:43:48 Okay, 0xFF is slightly less odd. 17:44:01 (I'd just go with 0x00, myself) 17:44:05 0xffff for UTF-16 17:44:08 So 255 is "less odd" than 254? Odd definition for odd. 17:44:13 Possibly 0xffffffff for UTF-32 17:44:29 fizzie: 254 just seems arbitrary. 17:44:34 But even. 17:44:42 the idea behind lazy exceptions is that if something goes wrong when calling a function, it returns a return value that throws an exception if you try to use it 17:45:06 but you can look at it to see if it's a genuine return value or a lazy exception 17:45:27 -!- jix has quit (Success). 17:45:39 I wonder what Perl 6 would look like if you chopped off the sigils and the semicolons. 17:45:43 I bet Plof. 17:45:52 Now to try it. 17:45:55 Perl 6 has both sigils and twigils, though 17:46:03 Well, fuck them too, whatever they are 17:46:09 with the result that it would be hideously ambiguous if you chopped the sigils off 17:46:11 ehird: Not quite. 17:46:15 a twigil is sort of a sigil for a sigil 17:46:28 Vaguely close, though. 17:46:36 Sounds like insanity to me 17:46:43 Perl 6 is pretty shit 17:46:50 Deewiant: So how does a non-initialized variable declaration look like? 17:47:01 as in, $var is a scalar variable, $.var is a scalar property of a class 17:47:09 fizzie: variable = void; 17:47:19 ehird: Perl6 is just even further into the realms of perlness than perl5, I rather like it 17:47:55 Perl 6: http://svn.perl.org/perl6/pugs/trunk/examples/life.pl 17:47:57 Ploferl 6: http://pastie.org/467798.txt?key=i3fnmc5yygtz8yly0k1zq 17:49:23 I don't think it's all that Ploflike. 17:49:38 Though I'm sure it could be implemented in PSL. 17:49:46 pikhq: It's not too far off 17:50:04 Vaguely similar. 17:50:06 that is insanely similar, of course 17:50:15 except, I suspect various other bits of Perl6 wouldn't translate 17:50:30 17:50 pikhq: Vaguely similar. 17:50:30 17:50 ais523: that is insanely similar, of course 17:50:31 lawl 17:50:36 that seems to have been using just the Perl5y bits 17:50:49 that's because most of the perl 6 bits are useless 17:50:57 what's Plof's OO like? 17:51:01 what are its regexps like? 17:51:08 Plof's OO is, uh, Javascript. 17:51:15 Its regexps are PCRE. 17:51:20 But you can only have them in the bnf declaration thing. 17:51:20 so really rather different, then 17:51:22 For now. 17:51:26 Plof's OO is basically the Javascript method done better. 17:51:34 Er, pikhq, it barely differs. 17:51:37 Apart from removing constructors. 17:51:53 Still better. 17:52:26 Also, there's PCRE bindings for Plof. 17:52:42 See: c_pcre.plof 17:52:57 does Plof have typed variables? 17:53:18 Erm. That's not done yet. 17:53:32 ais523, the type system consists of the object system. 17:53:46 Variables are untyped. 17:53:49 my guess is that Plof will model a subset of Perl6 17:54:00 actually, fwiw, /every language in existence/ will model a subset of Perl6 17:55:33 plofbnf { top = /flimble/ => plof { print("Flimble!\n") } } 17:56:02 lifthrasiir: how is clang working? 17:56:09 -!- Hiato has joined. 17:56:18 i was working on other optimization passes. 17:56:23 ais523: no, perl 6 misses a lot of things 17:56:24 lifthrasiir: ah 17:56:36 lifthrasiir: i'm upset because i can't think of a way to improve on what you're doing :-) 17:57:16 my Cat|Dog Fish $mitsy = new Fish but { Bool.pick ?? .does Cat !! .does Dog }; 17:57:32 MORE SYNTAX! MOOOOOOOOAR SYNTAX! 17:57:37 SYNTAX SOLVE EVERYTHING! 17:58:04 * pikhq can add that to Plof 17:58:17 ehird: it only performs a basic optimization. it even didn't remove codes after infinite loop... :S 17:58:29 i just implemented that and pushed to hg repo. 17:58:37 actually, fwiw, /every language in existence/ will model a subset of Perl6 <-- so true... 17:58:37 lifthrasiir: it's the most anything does atm 17:58:57 turning things into expressions, polynomials, for loops 17:59:14 my Cat|Dog Fish $mitsy = new Fish but { Bool.pick ?? .does Cat !! .does Dog }; <-- what does it do... 17:59:45 AnMaster: it's a variable that's both a fish and a (cat xor dog) 17:59:53 ... 17:59:57 heh 18:00:03 you know. like in real life 18:00:05 so what does ".does"? :S 18:00:17 lifthrasiir: "I implement this interface" I think 18:00:21 to be precise, it generates a Fish, then makes a copy of it with either catness or dogness applies, at random 18:00:32 augh 18:00:33 hahahah 18:00:37 ehird: .does is an abbreviation for $_.does 18:00:42 ehird: hmm. 18:00:46 *catness or dogness apllied 18:00:47 *applied 18:00:48 ais523: that's just awful 18:00:59 ehird: we were discussing "true but false" earlier 18:01:05 ais523, sorry, but this is *worse* than intercal 18:01:11 AnMaster: no, it's not 18:01:13 which in Perl6, is a scalar value identical to true, apart from being false 18:01:16 i thoroughly read perl 6 specs a year ago, but cannot think of it. heck. 18:01:17 if you think that you don't know intercal :) 18:01:19 ais523: :D 18:01:20 ehird, this specific aspect of it at least. 18:01:30 so how about comefrom? 18:01:39 which in Perl6, is a scalar value identical to true, apart from being false <-- err.. what 18:01:41 you can probably implement comefrom in perl 18:01:45 AnMaster: boolean value 18:01:50 object True is true as a boolean 18:01:51 yes 18:01:53 object False is false as a boolean 18:01:55 object 3 is true as a boolean 18:01:59 hm 18:02:04 (true but false) is an object identical to True, but it's false as a boolean 18:02:07 doesn't seem too bad. 18:02:14 yeah, it's just an edge-case 18:02:16 ah... that bit was bad 18:02:20 I added return stack access to my False variant 18:02:24 "but" 18:02:27 what does that do 18:02:32 AnMaster: makes a modified copy 18:02:36 AnMaster: (3 but false) is an object identical to 3 but false as a boolean 18:02:38 oh god 18:02:46 for better stack access and custom control structures 18:02:47 that's different in some way, to do with properties or roles or traits or whatever 18:02:48 it makes perfect sense... 18:03:00 iano: it may be TC 18:03:01 like that 18:03:13 it depends on what sort of access 18:03:14 ehird: exactly 18:03:23 ehird, So 3 is an object representing 3, but booleans are not objects? 18:03:28 or did I misunderstand you 18:03:33 AnMaster: you misunderstand me 18:03:38 ah 18:03:40 In Plof: (var tmp = new(True);tmp.ifTrue = Bool.ifTrue;tmp.ifFalse = False.ifFalse;) 18:03:41 what booleanity an object has is an aspect of it 18:03:48 right 18:03:52 True is an object which has nothing other than booleanity-true 18:03:56 False vise-versa 18:04:05 Hrm. 18:04:07 So (true but false) is True but with booleanity-false 18:04:19 Takes a bit more work than that; no opOr, opAnd, or opNot. 18:04:22 ehird, what about primitive datatypes. Do they exist, or is everything an object. 18:04:23 this is rather useless, ofc 18:04:29 atm, I'm trying to work out what (true but false) stringifies as 18:04:37 AnMaster: i think everything's an object, apart from the array/scalar/hash distinction; ask ai 18:04:38 ais523: 18:04:46 ais523: TIAS? 18:04:47 fair enough 18:04:49 it depends on whether true has its stringification as "true" bundled with it, or whether it's calculated automatically from its truth 18:05:07 ehird: actually, in Perl6 everything can be treated as an object or as a nonobject 18:05:07 LOL, Jeff Atwood uses a password in the dictionary 18:05:10 it does autoboxing 18:05:11 His idiocy is unbounded 18:05:29 to the extent that it's rather difficult to tell if anything is boxed or not at any given moment 18:05:54 ehird, who 18:05:59 AnMaster: exactly 18:06:13 Jeff Atwood is a retarded microsoft drone who has a shitty blog where he is a moron extraordinaire. 18:06:18 isn't Jeff Atwood that Microsoft fan who spends all his time trying to be famous? 18:06:19 He owns stack overflow with Joel Spolsky. 18:06:19 ah 18:06:23 ouch 18:06:41 Joel Spolsky's an idiot but he's said some clever things in his time... Jeff Atwood, uh, not so much. 18:06:49 In Plof: (var tmp = new(True);tmp.ifTrue = Bool.ifTrue;tmp.ifFalse = False.ifFalse;tmp.opAnd = False.opAnd;tmp opOr = False.opOr;tmp opNot = False.opNot;) 18:06:53 He said it wasn't "really" a dictionary password 18:07:05 Deewiant: Yeah it was one of those non-dictionary dictionary passwords 18:07:06 -!- tombom has joined. 18:07:11 I like how he calls people dummies in the same sentence 18:07:18 Real smart there today Atwood 18:07:25 Or, to make that an operator: 18:07:37 beh, Perl6 isn't in the Ubuntu repos 18:07:40 not even Rakudo 18:07:53 ais523: err, rakudo IS the canonical perl6 18:07:58 well, yes 18:07:59 ais523: anyway, is Parrot? 18:08:03 Parrot is 18:08:13 but I mean, Perl6 isn't finished yet 18:08:16 but Rakudo exists today 18:08:16 get parrot, then http://rakudo.org/how-to-get-rakudo 18:08:19 err 18:08:20 ais523: 18:08:24 Perl6 is not an implementation 18:08:27 no, it isn't 18:08:30 Rakudo is the "perl.org" implementation of perl 6 18:08:36 ie the sort-of-canonical one 18:08:43 ais523: anyway, get parrot then http://rakudo.org/how-to-get-rakudo 18:08:45 there can't be an implementation of Perl6 yet, because it isn't finished; Rakudo presumably implements the work-in-progress spec 18:09:01 ah wait 18:09:03 ais523: don't get parrot 18:09:04 it does it for you 18:09:16 git clone git://github.com/rakudo/rakudo.git && cd rakudo && perl Configure.pl --gen-parrot && make 18:09:18 then use ./perl6 18:09:25 which is a repl 18:11:08 ais523: I would like to point out that Perl 6's other name is PEYHM. 18:11:32 no results for that either 18:11:48 ais523: I'm not being too serious: http://www.pugscode.org/images/pugs.small.png 18:11:57 It's more like Pugs' other name for PErl 6, though. 18:12:00 *Perl 18:12:24 aww, Perl6 no longer allows ' as a namespace separator 18:12:28 opBut = (x, y as Bool) {var tmp = new(x) : Bool;tmp.ifTrue = y.ifTrue;tmp.ifFalse = y.ifFalse;tmp.opAnd = y.opAnd;tmp.opOr = y.opOr;tmp opNot = y.opNot;} 18:12:29 admittedly, that's been deprecated for ages 18:12:32 plofbnf { plof_unopr => plof_but; plof_but = plof_but "but"w plof_but_next => plof {opBut($0, $1)}; plof_but = plof_but_next => plof { $0 };} 18:12:41 And yes, I am completely mad. 18:12:56 ais523: done that git/cd/perl/make dance to get rakudo? 18:12:58 it handles everything for you 18:13:02 ehird: yes, it's running atm 18:13:05 ah 18:13:08 I checked the commands and the website first, thuogh 18:13:10 *though 18:13:48 --gen-parrot is code for "launch nuclear weapons", obviously 18:14:05 no, that's clrscr() 18:14:35 :-) 18:15:27 "its performance is excellent (in terms of both compilation speed and - literally - execution speed)" 18:15:32 — ZOG C 18:16:08 * ehird looks at Rakudo building and decides to look away because it's probably awful and bloated and meta 18:16:10 I think we're referencing the same website 18:16:25 * pikhq is pleased at having implemented the but operator in Plof) 18:16:36 and the build's mostly full of cc and parrot invocations 18:16:45 pikhq: but it doesn't do all of what Perl6's but does 18:17:07 18:16 ais523: I think we're referencing the same website ← ofc 18:17:17 That takes more than two lines of silly stuff hashed out in an IRC buffer. 18:17:20 can you make it apply roles to classes? 18:17:41 that's probably its major use 18:17:45 "Linked: perl6" 18:17:47 No, that takes a more sophisticated opBut. 18:17:48 Leeeeeeeeet's gooooooooooo! 18:17:55 opButt 18:17:58 although I look forward to testing things like "4 but Callable 18:18:00 " 18:18:00 The syntax definition there is just fine, though. 18:18:02 > say 2+2 18:18:02 4 18:18:04 YAY 18:18:11 ais523: 18:18:12 > true but false 18:18:13 Could not find non-existent sub false 18:18:19 there is no "false" 18:18:23 ais523: try False 18:18:25 True but False 18:18:30 > True but False 18:18:31 The but operator can only be used with a role or enum value on the right hand side 18:18:33 > say (true but false) 18:18:34 Could not find non-existent sub false 18:18:36 > say (True but False) 18:18:37 The but operator can only be used with a role or enum value on the right hand side 18:18:39 > say (True but false) 18:18:40 yep 18:18:40 Null PMC access in isa() 18:18:42 ah 18:18:43 haha 18:18:49 > say (4 but Callable) 18:18:49 4 18:18:51 what now 18:18:54 (4 but Callable)()? 18:18:54 try calling it 18:18:55 or ->() 18:18:55 yes 18:18:59 just () 18:19:01 ais523: invoke() not implemented in class 'Integer' 18:19:04 that's reasonable 18:19:07 yes 18:19:11 this thing is slow 18:19:18 i can feel the delay from typing to error 18:19:19 :-) 18:19:44 rakudo feels rather buggy to me 18:19:51 compared to the spec, at least 18:20:06 nope 18:20:15 ais523: make test 18:20:17 make spectest 18:20:20 > say (*+1)(3) 18:20:22 4 18:20:32 will try the rakudo tests and the perl6 tests from pugs 18:21:36 All tests successful. 18:21:36 Files=29, Tests=236, 56 wallclock secs ( 0.16 usr 0.12 sys + 47.38 cusr 3.89 csys = 51.55 CPU) 18:21:37 Result: PASS 18:21:39 now for spectest 18:21:51 /Users/ehird/Downloads/rakudo/parrot/parrot perl6.pbc --target=pir --output=Test.pir Test.pm 18:21:53 Boy that's slow 18:22:41 "This is one of our favourites. The value of NULL on a DeathStation 9000 varies from one compile to the next. It might be all-bits-zero and it might not." 18:22:52 err 18:22:52 "So this code: memset(&weapon, 0, sizeof weapon); wasn't quite as robust as it might have been" 18:22:56 0 as a pointer is always NULL 18:23:00 that's correct code 18:23:03 THE DEATHSTATION IS WRONG! 18:23:05 no, it isn't 18:23:10 the second argument of memset isn't a pointer 18:23:13 ah 18:23:13 it's a char 18:23:19 0 as a char != 0 as a pointer, necessarily 18:23:25 ic ic ic 18:23:28 yep 18:24:17 ais523: all these spectests are succeeding atm 18:24:24 bsmntbombdood_: hi 18:24:34 ehird: I doubt they're in sync with the actual spec, though 18:24:45 ais523: they're reasonably in sync. 18:24:50 the spec doesn't change much 18:24:51 mostly it's hot air 18:26:39 heh, Perl6 doesn't just have unary plus, it also has unary concatenate 18:28:38 http://pastie.org/467828 esotope-bfc is, i think, getting much slower over the revisions... :S 18:28:49 who cares! 18:28:58 lifthrasiir: how does it do lostkng? 18:28:58 > say (:a).perl 18:28:58 :-D 18:28:59 "a" => "b" 18:29:01 > say (:a<>).perl 18:29:02 "a" => "b" 18:29:04 > say (:a«b»).perl 18:29:05 Statement not terminated properly at line 1, near "\x{c2}\x{ab}b\x{c2}\x{bb}).per" 18:29:07 fail 18:29:09 ais523: no 18:29:11 encoding fail 18:29:17 not its f ail 18:29:18 not encoding fail 18:29:22 are you sure 18:29:27 the error message strongly implies I passed correct UTF-8 to it 18:29:36 ehird: given link has a table of time taken to compile lostkng. 18:29:40 ooh, or does it? 18:29:40 ais523: tell it you're using utf-8, then 18:29:45 lifthrasiir: but what's the output 18:29:47 pastie it? :-D 18:29:47 maybe it's trying to interpret my utf8 as latin-1, for some reason 18:29:51 ah, well... 18:29:56 can pastie handle it? :p 18:30:05 lifthrasiir: perhaps! 18:30:07 eh, just like to esotope 18:30:08 (800k+, 40k+ lines) 18:30:09 and I'll do it 18:30:28 i'll try it. 18:30:35 $ perl6 --encoding=utf8 18:30:37 > say (:a«b»).perl 18:30:38 Statement not terminated properly at line 1, near "\x{c2}\x{ab}b\x{c2}\x{bb}).per" 18:30:39 link to esotope anyway lifthrasiir :-P 18:30:55 ais523: i highly doubt it doesn't support that syntax; it's something else 18:31:02 yes, most likely 18:31:05 ehird: http://hg.mearie.org/esotope/bfc/raw-file/tip/esotope-bfc.py 18:31:09 thx 18:31:46 $ LC_ALL=en_US.UTF-8 perl6 --encoding=utf8 18:31:48 > say (:a«b»).perl 18:31:49 Statement not terminated properly at line 1, near "\x{c2}\x{ab}b\x{c2}\x{bb}).per" 18:31:50 when it once became stable i'll release it to avoid pasting that link everyday. :p 18:31:59 noo 18:32:01 if Rakudo has some way to indicate an encoding, it isn't an obvious one 18:32:02 releases suck 18:32:28 lifthrasiir: ImportError: cannot import name namedtuple 18:32:29 WAT 18:32:36 do I need 2.6 18:32:36 ehird: "Running "make spectest" will import relevant portions of the official Perl 6 test suite from the Pugs repository (http://svn.pugscode.org/pugs/t/spec/) and run all of the tests that are currently known to pass." 18:32:39 ehird: it requires python 2.6... maybe? 18:32:46 ais523: ah, lol 18:32:46 wait 18:32:46 so it only tests the bit they've implemented against the spec 18:32:49 no wonder it was passing 18:32:59 shrug, rakudo is quite complete 18:33:04 you seem to think it's a lot worse than it i 18:33:04 s 18:33:11 ehird: sorry, namedtuple is not in use in current revision. remove "from collections import namedtuple" line from the code. 18:33:13 rakudo 18:33:14 what is that 18:33:18 *sigh* 18:33:24 AnMaster: Perl6 work-in-progress bytecode compiler 18:33:27 aha 18:33:39 how come you're the only person who gets the privilege to ask questions that you could easily find out, nobody else? 18:34:20 -!- iano has quit. 18:34:34 % python esotope-bfc.py LostKng.b > LostKng.b.c 18:34:38 lifthrasiir: this is taking >11s... 18:34:51 what's your sys 18:34:52 ehird: since it uses psyco if any. 18:34:55 ah. 18:35:04 it finished 18:35:21 > my Int $x = (fail "testing"); 18:35:22 > say $x+1; 18:35:24 Scope not found for PAST::Var '$x' in 18:35:25 I'm finding bugs everywhere 18:35:28 and yes, that is the entire error message 18:35:36 ais523: Are you sure you know the spec more than they do? 18:35:40 I bet your code is invalid 18:35:41 how come you're the only person who gets the privilege to ask questions that you could easily find out, nobody else <-- What are you talking about. 18:35:52 lifthrasiir: doesn't seem to do much more than bf2c.hs on lostkng 18:35:58 ehird: I have the spec open atm... 18:36:03 although, idea 18:36:05 where is bf2c.hs? 18:36:06 ais523: report the bugs then 18:36:15 lifthrasiir: http://esoteric.sange.fi/brainfuck/impl/compilers/bf2c.hs 18:36:17 it's possible that the REPL runs each line in a separate scope 18:36:21 ais523: DUH 18:36:23 it says that on the page 18:36:29 the get rakudo page 18:36:34 each line is a separate compilation unit 18:36:37 so subs persist, but not vars 18:36:44 lifthrasiir: until recently the best compiler 18:36:53 lifthrasiir: does move-shifting, variable elimination, polynomialization... 18:37:00 BF optimization is a field in itself... 18:37:28 > {my Int $x = (fail "testing"); say $x+1;} 18:37:29 > {say 4;} 18:37:31 4 18:37:38 presumably, attempting to output an unthrown exception doesn't throw it 18:37:41 I wonder what does? 18:37:47 throwing it? 18:38:14 well, yes, but they're meant to throw themselves whenever you try to do something with them that's obviously intended not to be done with an exception 18:38:59 lifthrasiir: does your compiler work on [>]? 18:39:14 not yet. 18:39:33 lifthrasiir: what, it fails on valid code? 18:39:35 i'm thinking about how to optimize such code. 18:39:38 ah 18:39:38 oh 18:39:42 i just mean does it work :-P 18:39:46 lifthrasiir: also, [>] is pretty easy 18:39:48 memchr 18:39:49 :p 18:40:11 of course it does work on every valid code, just not get optimized 18:40:14 lifthrasiir: [>] is mptr = memchr(mptr, 0, (size ofmemptr)) 18:40:24 and you're right, but how about [>>] or variants? 18:40:24 which is mptr-mem, I think 18:40:47 i think combined array analysis is needed for complete optimization. 18:40:48 hmm 18:40:49 dunno 18:40:56 and that sounds hard :-) 18:41:37 for example the memory cells 7+2*k is actually one array, and optimized to be stored as like etc. 18:42:05 lifthrasiir: [>] is mptr = memchr(mptr, 0, (size ofmemptr)) 18:42:11 yes, I said that yesterday iirc 18:42:16 it's obvious :P 18:42:27 true 18:42:35 lifthrasiir: yeah, that'd make it turn into "real" c 18:42:36 that's what you say now ;P 18:42:40 if you analyzed data structures 18:42:44 to get variables, arrays, etc out of it 18:42:48 what'#s the best compil;er nowtrhen 18:42:55 tombom: esotope-bfc 18:42:59 thanks 18:43:00 by lifthrasiir 18:43:05 ah ha 18:43:15 tombom: & compiling the generated c with clang/llvm 18:43:29 * GregorR consoles EgoBFC. 18:43:32 i'm not sure, but it does certain degree of optimization 18:43:40 lifthrasiir: you should optimize the 1,X,1,X,1,X,0 array format 18:43:41 somehow 18:43:43 GregorR, screen? 18:43:57 AnMaster: console(v) 18:44:11 Ah yes, computer science has entirely destroyed the previous meaning of the word console :P 18:44:23 oh.. that meaning.. right 18:44:56 (OK, I guess consoles in the noun sense predate computers by a bit) 18:45:55 ehird: btw, pastie says "Your paste cannot be larger than 100 kb. Sorry." so is this a new goal? :p 18:45:56 Where 'a bit' = a few centuries, yese 18:45:57 hah, IBM's offering companies $8000 for each Sparc processor they replace with a Power processor 18:45:59 GregorR, what is the name for those triangular things you put under shelves to mount them on the wall 18:46:07 in Swedish it is "konsoll" 18:46:08 lifthrasiir: :D 18:46:11 ais523: O_O 18:46:16 but I don't remember if it is console on English as well 18:46:20 i forgot people still use non-x86s in non-embedded environment 18:46:29 on servers, mostly 18:46:32 I'm not sure what you're referring to. 18:46:47 AnMaster: they're normally called shelf brackets in English 18:46:53 ais523, aha 18:46:55 It might be that Swedish furniture technology is advanced to the point that you have words for things we don't even have. 18:46:59 not that I often need to call them anything 18:47:32 ais523, nor do I. I don't think I have any shelves mounted that way in the house... 18:47:48 I do, but have never felt a need to name the things holding the shelf up 18:49:30 ais523, well, if you are going to buy some it might be useful to know 18:50:13 ah, first google result for "shelf brackets": http://www.wickes.co.uk/Shelf-Brackets/Metal-Brackets/icat/tsmetalbrack 18:50:16 GregorR: yeah ikea is like the LHC 18:50:19 for furniture 18:51:32 ais523, right. Was thinking of larger ones in wood though. Usually triangular. Sometimes ornamented(sp?) 18:51:39 well, yes 18:51:45 they're all shelf brackets either way, though 18:51:51 it seems wickes only make the metal ones 18:53:13 AnMaster: remember when I showed you a cooler which you thought was mad because it was big and heavy and would fall? 18:53:15 AnMaster: cpu cooler 18:53:22 AnMaster: here's the cooler I'm currently planning on using: http://www.silentpcreview.com/files/images/prolima-megahalems/12.jpg 18:53:26 ehird, vaguely 18:53:35 women and children, shield your eyes 18:53:35 * AnMaster looks 18:53:58 ehird, it looks like it would put quite a large load on the mobo 18:54:12 especially if the mobo is mounted vertically 18:54:14 " women and children, shield your eyes" // must just be a giant penis you attach to your CPU 18:54:19 GregorR: lawl 18:54:28 AnMaster: the attaching system is pretty solid 18:54:39 ehird, Sure. But is the PCB! 18:54:41 (That's what she said) 18:54:58 AnMaster: "A backplate is placed beneath the CPU socket and bolts with threads on both sides are secured to it." "Two aluminum side bars are placed over the bolts and nuts are used to tighten them. There are a second set of holes on these bars for use with the LGA1366 backplate." "The heatsink is then placed on top of the CPU and a crossbar is fitted above the mounting plate. Large spring-loaded bolts are then screwed into the side bars. This is the only 18:55:00 step that requires any tools — the rest is done by hand." 18:55:05 ... Ok ... I never understood what this "That's what she said" was about. 18:55:17 is it just random xkcd reference, or? 18:55:18 > my $foo = "Bar"; my $Bar=4; say $::("MY::$foo"); 18:55:20 say requires an argument at line 1, near " $::(\"MY::" 18:55:23 AnMaster: it's a sexual innuendo 18:55:26 I see. 18:55:27 ehird: I just copied that example from the spec 18:55:33 AnMaster: It's meant to suggest that something is sexual innuendo when it really isn't. 18:55:38 aha 18:55:45 AnMaster: "That's so small." "That's what SHE said [in bed, referring to your penis]." ← example 18:56:00 AnMaster: Since I said the cooler must look like a giant penis, I decided that " AnMaster: the attaching system is pretty solid" must be innuendo. 18:56:08 I wonder if anyone's ever tried that riposte to a woman? 18:56:11 ehird, backplate hm.. Hope it isn't a conductive one. 18:56:20 AnMaster: those megahalems are 820g with the crossbar, bolts and fan clips, apparently 18:56:26 790g for just the heatsink 18:56:35 so not that heavy 18:56:41 that's in terms of weight, presumably 18:57:07 err, duh? 18:57:22 ais523, isn't it upper case G for acceleration iirc 18:57:25 I thought so, I wasn't quite sure there wasn't a second meaning though 18:57:39 AnMaster: yes 18:57:42 right 18:57:52 790 G would be insane :D 18:58:02 depends on what for 18:58:11 it might be reasonable propulsion caused by a nuclear explosion 18:58:14 of a small object 18:58:17 hmm, apparently with one fan on the megahalems running at only 800 RPM, with full load of a top of the range core 2 quad (NOT i7), it runs at 62c 18:58:31 so probably 65-70c for an i7 at full load 18:58:33 ais523, anything involving a (former) human being 18:58:40 is megahalem a trade name, or does it have a technical meaning? 18:58:44 question is if you can drop the fan 18:59:01 AnMaster: "Megahalems" is the brand of those giant heatsinks for the Core i7s, which use the Nehalem architechture 18:59:05 what 18:59:13 Did you mean: ais523 18:59:18 AnMaster: what 18:59:22 er 18:59:22 yes. 18:59:24 AnMaster: Did you mean: Did you mean: ais523: 18:59:25 right 18:59:33 ais523, wut 18:59:39 you missed the colon 18:59:56 hmm wait 19:00:08 the megahalems are a lot less hot in silentpcreview's review 19:00:16 ais523, no. Shouldn't there be a dot at the end 19:00:28 30-40C 19:00:28 AnMaster: ehird said "AnMaster:" 19:00:37 which is what you were trying to correct in the first place 19:00:38 ais523, I perform word splitting on : 19:00:39 so I'm assuming that dropping the fan would be easy 19:00:49 AnMaster: I give context in corrections 19:00:56 you'd get 30-40C or so on average and higher on load 19:00:58 which sounds fine 19:01:06 just like diff -u >> diff for patching programs 19:01:26 ais523, ok. I should have done / Did/s/: /: $/ 19:01:27 then 19:01:30 agreed 19:01:31 err 19:01:38 s/\$/^/ 19:01:39 even 19:01:55 wait, what? 19:01:58 .. 19:02:01 replace end-of-string with start-of-string? 19:02:05 does that even make sense? 19:02:07 ais523, in the regex 19:02:18 ais523, I escaped $ 19:02:21 thus replacing 19:02:25 ais523, ok. I should have done / Did/s/: /: $/ 19:02:26 with 19:02:27 ugh, I can't believe I'm actually considering applying my own gigantic heatsink 19:02:28 I must be barmy 19:02:28 ais523, ok. I should have done / Did/s/: /: ^/ 19:02:31 then 19:02:32 that replaces 19:02:36 Did you mean: ais523 19:02:37 with 19:02:40 I don't get why you want a ^ in the replacement anyway 19:02:40 Did you mean: ^ais523 19:02:49 thus giving the context 19:02:50 :P 19:02:50 yes, but surely ehird didn't mean "^ais523" 19:03:04 ais523, It is the context. 19:03:06 oh, you're trying to anchor the context to the start of the line/ 19:03:11 yes... 19:03:12 lines start with : not ^ on IRC 19:03:12 "In keeping with the spirit of the [H] we are once again doing hardware heat measurment. This means drilling a very small path into an expensive CPU to place our thermocouple in. This is by far the best way to test coolers and the only way here at the [H]." Holy shit, they're insane 19:03:19 http://enthusiast.hardocp.com/image.html?image=MTIzMjU1MjM1MlEyZ1NFQzg5ckJfMV81X2wuanBn 19:03:21 so s/^/:/ in your most recent metacorrection 19:03:22 I shit you not 19:03:26 They cut open a cpu and put a wire into it 19:03:34 ais523, not the way the client displays it 19:03:34 and 19:03:43 Did you mean: :ais523 19:03:45 AnMaster: for client display, you want > 19:03:47 would have looked silly 19:03:57 unless your client displays IRC messages as regexps? 19:04:03 -_- 19:04:16 ^Hello, .*!$ 19:04:31 * AnMaster actually lols at this convo. 19:05:46 well, I was deliberately trying to make it absurd 19:05:56 you succeeded. 19:05:59 well done. 19:09:39 "The fan completely blocks any RAM from being installed in the first slot and even the second slot blocks the retention clip from properly holding the fan. " 19:09:40 Ouch. 19:09:46 Hope my mobo's big enough to avoid THAT. 19:12:25 -!- BeholdMyGlory has joined. 19:14:15 -!- AnMaster_ has joined. 19:16:14 lifthrasiir: test with mandelbrot.b 19:16:16 not lostkng 19:16:20 lostkng is big but not intensive 19:16:38 is that http://www.menuetos.net/mandel.txt ? 19:16:54 lifthrasiir: http://swapped.cc/bf/ 19:17:00 http://swapped.cc/bf/files/mandelbrot.b 19:17:14 indeed same. thank you for suggestion. 19:17:21 also, ew, menuetos :-P 19:24:01 -!- AnMaster has quit (Connection timed out). 19:26:52 -!- AnMaster_ has changed nick to AnMaster. 19:27:17 power outage 19:27:18 :( 19:27:19 > eval "my $x = 3;"; 19:27:20 Scope not found for PAST::Var '$x' in 19:27:20 so 19:27:23 what did I miss 19:27:31 and 19:27:36 what were we talking about 19:27:43 I completely forgot 19:28:26 AnMaster: I forwarded you an e-mail earlier today, FWIW. 19:28:34 Deewiant, let me check.. 19:30:02 " Your interpretation would be considered undefined at best. The specs say that the child should execute before the parent next executes but does not specify immediately before, only that it must execute an instruction before the parent executes its next instruction. In fact it does not even state whether the child ip needs to execute in the same tick as the t command that created it, or execute it 19:30:02 for the first time in the next tick. Both CCBI and Rc/Funge-98 execute it the next tick, but again, which tick it executes in is undefined and only really has importance when using TRDS. " 19:30:03 hm 19:30:31 specify it in 108! 19:30:34 if we are going to go undef like that almost half of the GOOD in mycology will turn into UNDEF 19:30:49 No, not really :-P 19:31:14 I'm pretty certain that process creation taking more than one tick would be a DS9K implementation of Befunge-98 19:31:35 That's not what that's about 19:31:44 oh, what is it about then? 19:32:08 AnMaster: who wrote that? 19:32:17 almost cetainly mike riley 19:32:22 ehird, do you agree with it or not 19:32:23 although there isn't any direct evidence for that 19:32:24 ;P 19:32:25 AnMaster: yes 19:32:25 just an implication from contetx 19:32:28 *context 19:32:33 he writes better not on IRC then 19:32:40 ehird, Riley 19:32:40 Well, the original message was that if you have, say IPs [ 1 2 3 ], then if 2 forks to create 4 you should get [ 1 2 4 3 ] and not [ 1 2 3 4 ] 19:32:47 maybe his irc client inserts the ,,,, ... and ehhehehehehs 19:33:04 Deewiant: yep, that's right 19:33:11 ehird: No, it's undef 19:33:18 yes but it's the right option :-) 19:33:20 * AnMaster tries to figure out DS9K interpretations for the fingerprint acryonyms 19:33:47 FPSP: Fire-Powered Shattering Pulser 19:33:54 think EMP 19:33:58 And Mike's "it does not even state" is about whether, when the child IP is created, it executes for the first time in that same tick or the next one 19:34:01 but with more fire and explosion 19:34:32 NCRS: NuClear Range Select 19:34:33 :) 19:34:52 MVRS: Multiverse tools, including D: destroys the current universe. 19:35:26 darn 19:35:27 mvrs has no D 19:35:55 REXP: Rectal Excavation eXtra Powertools 19:36:04 Used in hospitals to control a rectal excavation machine. 19:36:28 "SOCK": Securely Operated Checkless K-Bolts 19:36:38 what is K-bolt 19:36:38 SUBR: Submarine control system. 19:36:54 AnMaster: a weapon of some sort; it sounds like a long, thin bolt that's highly explosive or somethin 19:36:54 g 19:37:02 AnMaster: I think they're explosive bolts used to hold parts of rockets together 19:37:03 it doesn't actually exist, but that's what it sounds like 19:37:06 ais523: what, really? 19:37:08 ais523, ah 19:37:10 i just invented them! 19:37:10 which are blown up to separate the stages 19:37:16 ehird: they may be called something else 19:37:19 well explosive bolts exist 19:37:20 I'm not sure 19:37:23 http://www.google.com/search?client=safari&rls=en-us&q=K-bolt&ie=UTF-8&oe=UTF-8 19:37:23 well,* 19:37:26 yeah, I just invented them ;-) 19:37:37 TRDS: controls the *real* TARDIS. 19:37:54 ehird: wow, that was really weird 19:37:55 UNIX: Castration utilities. 19:37:59 I followed your link, and only got one advert 19:38:02 ais523: huh 19:38:04 then I removed the client=safari, and got lots 19:38:08 ha 19:38:13 ehird, err. That isn't a backronym 19:38:14 yep 19:38:19 if i s/safari/firefox/ i get more 19:38:24 i get 2 with safari 19:38:26 AnMaster: I konw 19:38:26 with client=epiphany, I get 2 19:38:28 know 19:38:31 ehird, FAIL 19:38:34 AnMaster: No... 19:38:38 AnMaster: How is that fail? 19:38:39 ais523, client=lynx 19:38:40 and client=lynx gives 6 19:38:40 try it 19:38:43 ah 19:38:45 I know trds means tardis. 19:38:45 I tried it before you suggested that 19:38:52 ais523, mosaic 19:39:00 ie? 19:39:02 and what ads... 19:39:04 AnMaster: 4 19:39:04 ;P 19:39:13 ie6 shows five 19:39:20 google customize! 19:39:24 ie8 shows seven 19:39:26 (extension for firefox) 19:39:35 and ie7 shows none at all 19:39:44 wait, or 1 on the refresh 19:39:49 I suspect it's just giving random numbers of ads 19:40:03 nope, ie7 never gives more than one ad 19:40:28 and ie8 is not giving more than one now either, even after about 8 refreshes 19:40:32 wtf is Google up to? 19:41:07 ais523, random I suspect 19:41:08 -!- oerjan has joined. 19:41:28 ais523 or is it statistical significant? 19:43:49 it probably depends on other things like the IP 19:44:02 I mean, Google have a user-agent, why would they care about the useragent specified in the address bar? 19:44:04 that makes no sense 19:44:37 -!- tombom_ has joined. 19:44:40 why is the user-agent even in the address bar 19:44:45 that doesn't make sense 19:47:14 AnMaster: done with JS I think 19:47:17 to avoid server side load 19:47:58 ehird, err. How does it have to parse less that way 19:48:10 AnMaster: it doesn't 19:48:14 AnMaster: but they're already parsing the query string 19:48:20 ah true 19:48:21 they're probably just throwing away the user-agent header 19:48:25 and besides, it's easier to read 19:48:26 ie6 19:48:27 than 19:48:34 Internet Explorer Mozilla AWESOMEFEST (aaa;;:43847 NETSCAPE) 19:48:37 Mozilla (compatible) 19:48:39 isn't it 19:48:45 AnMaster: that's part of it 19:48:47 yeah 19:48:52 it is fully 19:48:54 funny* 19:49:03 yes I know why and so on 19:49:07 -!- tombom has quit (Read error: 60 (Operation timed out)). 19:49:08 -!- tombom_ has changed nick to tombom. 19:49:16 but it is still funny. 19:50:02 AnMaster: i've seen worse; a particular idiot thought that including the highest Acid standards test they pass would be best... even though the acid tests are 100% edge cases, and are DELIBERATELY INVALID to test browsers' handling of invalid pages 19:50:10 as in, just having the UA be "Acid 2" or "Acid 3" 19:50:27 the acid tests aren't even official w3c! 19:50:40 ehird, err acid 1 doesn't have any invalid bits iirc 19:50:46 acid 2 and 3 does yes 19:51:01 AnMaster: barely anything doesn't pass acid 2 these days 19:51:04 but my point stands 19:51:12 yes it would be silly 19:51:13 I agree 19:51:22 My Firefox (3.5b4) doesn't pass Acid2 19:51:32 Deewiant, uh... 19:51:33 I haven't tried in safe mode, though, so it could be an extension 19:51:36 3.0.x does 19:51:47 I guess it might be a beta bug 19:51:52 yar, regressions 19:51:56 acid tests don't matter anywhere 19:51:59 Deewiant: Acid2, or Acid3? 19:52:01 or an extension 19:52:02 a load of talk and nothing useful or worthwhile 19:52:05 The nose is a pixel or so too big and the chin is one block too tall 19:52:08 ais523: 2 19:52:19 Even the latest alphas don't pass Acid3. 19:53:08 With extensions on, my browser doesn't pass Acid1. 19:53:23 Deewiant, you have odd extensions then 19:53:44 I refuse to allow pages to set the font size of monospaced fonts. 19:54:02 Deewiant, err why 19:54:19 Because they are invariably either way too small or way too big, because they assume a different font than I specify. 19:55:44 -!- FireFly has quit ("Later"). 19:57:18 what font's that 19:57:29 DejaVu, of course. 19:57:57 ew 19:59:27 -!- FireFly has joined. 20:00:02 I refuse to allow pages to set the font. 20:00:22 You've got a choice between sans serif, serif, and monospace. 20:00:28 Use it wisely. 20:00:41 pikhq, image-text ;P 20:00:49 * AnMaster runs 20:00:58 AnMaster: Someone using that gets shot. 20:01:07 agreed 20:01:20 20:00 pikhq: I refuse to allow pages to set the font. ← My designer sense is tingling, and it wants to throttle you. 20:01:35 Please assume a relaxed position so that it may commence. 20:02:05 I used to have that on but it broke too many pages 20:02:14 IMO, designers should design pages so that with any reasonable sane settings for the user, it'll display as the user wants 20:02:59 IMO designers shouldn't touch the 'font-foo' CSS settings at all. 20:02:59 too small is a big issue yeah 20:02:59 OTOH, a font with different metrics fucks up the whole vertical/horizontal rhythm entirly. 20:03:01 *entirely 20:03:10 I have minimal sizes set 20:03:21 Which is not a problem unless you actually, y'know, design. 20:03:24 to reduce that issue somewhat 20:03:29 HTML is not PDF/PS. 20:03:29 When it all breaks down. 20:03:35 Deewiant: Agreed wholeheartedly. 20:03:44 Irrelevant in this case, however. 20:03:45 ehird, you can't assume the user will have any of the fonts 20:03:52 AnMaster: That's why you specify fallbacks. 20:04:03 you can only assume there is some sans-serif, some serif, some monospace 20:04:16 no, you can't 20:04:18 you can't assume anything 20:04:26 ehird: I've done this because far, *far* too many 'designers' misuse it. 20:04:41 ehird, well true. But lynx doesn't support CSS anyway :P 20:04:44 pikhq: They ruin it for everyone else :P 20:04:48 so all you can assume is "a font" 20:04:55 AnMaster: braille 20:04:59 AnMaster: elinks supports CSS. ;) 20:05:02 ehird, true. 20:05:13 pikhq, irrelevant for lynx 20:05:52 ehird: What I'd like is a way to refuse allowing sites to set the font, with a whitelist for sites that aren't really retarded with font selection. 20:06:25 pikhq, make an extension.. called nofont 20:07:15 Temtping. 20:07:21 Very damned tempting. 20:07:26 Temtping? 20:07:30 funny typo 20:07:41 Opposite hands. ;) 20:08:01 still funny 20:08:47 I'd rather have yesfont 20:09:15 Since you'd have to whitelist the site anyway to check whether it's retarded or not :-P 20:09:58 -!- olsner has joined. 20:11:53 Tempt-ping; it sends seductive ICMP messages. 20:12:52 hah 20:13:16 Deewiant, why not have a setting for "default allow/disallow" 20:13:19 and call it... 20:13:28 boolfont 20:17:27 Because that complicates things 20:17:33 You have to keep track of both a blacklist and a whitelist 20:17:42 Deewiant, yes that allows exceptions to0o 20:17:44 Unless you just want to wipe it or invert its meaning whenever the setting is changed :-P 20:17:44 too* 20:21:52 -!- Hiato has quit ("Leaving."). 20:23:42 -!- kar8nga has joined. 20:26:58 Deewiant, just add the complement to the list when the option is changed ;P 20:27:24 Suure :-P 20:40:01 !underload (S):(u)(:*)(:*)::**^^(re :-P)**~^ 20:40:01 Suuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuure :-P 20:40:20 heh, a link someone pasted on Slashdot to try to explain the Daily Mail to Americans: http://www.google.co.uk/search?q=site%3Adailymail.co.uk+%22ministers+are+considering%22 20:40:49 oerjan: I'm sure that program could be simplified 20:41:00 actually, that must have been deliberately obfuscated 20:41:05 or you would have used more than one S 20:41:22 :) 20:41:47 once i realized (S)S could be rewritten, the rest was irresistible 20:41:55 fair enough 20:42:35 oh of course (:*)(:*) = (:*): 20:42:42 yes 20:42:47 and :::** will boggle people somewhat 20:44:03 !underload (S):(u)(:*):::**^^(re...)**~^ 20:44:04 Suuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuure... 20:44:17 Spiffy. 20:44:42 hm in fact that's a nice compression of 256 20:44:51 (:*):::** 20:44:53 agreed 20:44:57 no, wait 20:45:04 needs one ^ 20:45:04 (:*):::**^ is the compression 20:45:14 the second ^ is using the 256 to duplicate the u 20:45:15 I so hate church numerals ;P 20:45:20 !underload (:*):::**^ 20:45:27 !underload (:*):::**^S 20:45:27 :*:*:*:*:*:*:*:* 20:45:29 I think 20:45:31 hm 20:45:32 maybe not 20:45:33 Hrm. 20:45:44 Church numerals... 20:45:52 Make my head hurt. 20:45:52 pikhq, iirc yes 20:45:56 agreed! 20:46:04 I like Church numerals 20:46:25 I think we need yet another odd number system 20:46:25 -!- tombom has quit ("Peace and Protection 4.22.2"). 20:46:34 I liked that idea I had some days ago 20:46:50 that the interpreter selected the minimal possible base to interpret the number in 20:46:54 ais523: They're brilliant, they just screw with my head. 20:46:58 so 30 was in base 4, 45 was in base 6 20:47:00 and so on 20:47:04 !underload (X):*:*:*(~S:^):^ 20:47:04 XXXXXXXXError: Stack underflow in ~ 20:47:14 +ul (X):*:*:*(~S:^):^ 20:47:15 XXXXXXXX ...S out of stack! 20:47:15 ^ul (X):*:*:*(~S:^):^ 20:47:15 XXXXXXXX ...out of stack! 20:47:24 huh 20:47:35 they seem to be disagreeing about what in particular underflowed 20:47:39 exponential growth? 20:47:41 EgoBot and thutubot, that is 20:47:48 oh 20:47:48 AnMaster: "underflow", not "overflow" 20:47:51 right 20:48:00 AnMaster: i just wanted to check if it would print anything before the error 20:48:04 ah 20:48:05 hm 20:48:28 so you can use that for printing the whole stack like in the others 20:48:37 ais523, where should it really have underflown 20:48:45 ~ or S 20:49:01 ~ 20:49:11 if a ~ succeeds, the subsequent S can't possibly underflow 20:49:16 because there must be at least two stack elements 20:49:23 so bug in thutubot then 20:49:31 ais523, what was ~ now again 20:49:33 swap 20:49:33 probably 20:49:35 ? 20:49:41 or to be more precise, a bug in its error-handling 20:49:42 and yes, swap 20:50:30 hm... 20:50:38 +ul (x)~~S 20:50:38 ...~ out of stack! 20:50:45 +ul (x)~S 20:50:46 ...S out of stack! 20:50:56 curious :D 20:50:59 -!- kar8nga has quit (Remote closed the connection). 20:51:01 +ul (x)~: 20:51:01 ...: out of stack! 20:51:04 +ul (x)~a 20:51:05 ...a out of stack! 20:51:06 +ul (x)~' 20:51:13 oh, it didn't say ...' out of stack 20:51:15 +ul (test)S 20:51:15 test 20:52:12 +ul (x)~(Hi)S 20:52:12 Hi 20:52:25 +ul (x)~(Hi)SS 20:52:26 Hi ...S out of stack! 20:52:40 +ul (x)~(Hi)~(there)S 20:52:40 there 20:52:42 fungot doesn't say what underflows. 20:52:42 fizzie: " fnord?" she inquired. so, as i thought good, ' as many as five nights fnord warmth, you know.' 20:52:53 It's just a generic "out of stack" message. 20:52:57 I suspect it has some sort of underflow token that's being swapped onto the stack 20:53:03 and you get an underflow error when you try to use it 20:53:13 I didn't do that deliberately, but thutubot's complicated enough that I might have done it by mistake 20:54:46 +ul !(test)S 20:54:47 ...! out of stack! 20:54:59 +ul (x)~!(test)S 20:54:59 ...! out of stack! 20:55:25 I suspect it has some sort of underflow token that's being swapped onto the stack 20:55:25 and you get an underflow error when you try to use it 20:55:26 err 20:55:31 didn't you code it 20:55:37 oh right 20:55:38 seems only ~ gets fooled, and only for the second element 20:55:42 yes, but that's not the same as knowing how it works 20:56:04 ais523, so how did you detect underflow then 20:56:08 so probably ~ just checks for one argument 20:56:46 +ul (x)*(test)S 20:56:46 test 20:56:50 AnMaster: if the program underflows, none of the regexps match, so it falls out of the main loop unexpectedly 20:56:55 ah * has the same bug :D 20:56:58 ah 20:57:10 what does * do then. I don't remember 20:57:19 concatenate 20:57:20 +ul (x)*S 20:57:20 ...S out of stack! 20:57:37 +ul (x)*(test)~(hi)S 20:57:37 hi 20:57:58 +ul (x)~a 20:57:58 ...a out of stack! 20:58:32 those are the only two-argument commands 20:59:44 interesting that * doesn't mess up more though, since it actually combines the two args rather than just swapping them 21:00:37 +ul (x)*(test)~*S 21:00:38 ...* out of stack! 21:00:45 bah 21:00:50 +ul (x)*(x)*S 21:00:51 ...S out of stack! 21:00:56 ah, as I suspected 21:01:13 my current guess is that * and ~ both do the same thing as ! if there's only one element on the stack 21:01:13 ais523, what 21:01:23 which would be weird, but not completely implausible 21:01:57 ais523, what does ! do now again... 21:02:12 AnMaster: pop 21:02:15 ah 21:03:12 +ul (!):( drops)~^( accidentally)SS 21:03:12 accidentally! 21:03:21 oops 21:03:53 looks like it dropped certainly :P 21:04:11 yeah the accident was in the printing part 21:04:15 ais523, is there any underload in underload implementation 21:04:54 +ul ((Underload in Underload)S)^ 21:04:54 Underload in Underload 21:04:55 ^ul (!):( drops)~^( accidentally)SS 21:04:55 accidentally! 21:04:58 hm 21:05:12 oerjan, well that is cheating... I meant a non-eval implementation 21:05:16 SS should have been *S 21:05:23 AnMaster: underload has no input 21:05:27 true 21:05:41 but surely you could embed the input in some way 21:05:44 ^ul (!):( drops)~^( accidentally)*S 21:05:45 ! accidentally 21:05:48 +ul (!):( drops)~^( accidentally)*S 21:05:48 ! accidentally 21:05:58 !underload (!):( drops)~^( accidentally)*S 21:05:58 ! accidentally 21:06:08 AnMaster: there's an Underload interp in BF, and a BF-minus-input to Underload compiler 21:06:13 you could trivially combine the two 21:06:26 ais523, true. 21:06:32 you could code it in binary with : and ^, like i did with the rule 110 automaton 21:06:53 but you couldn't do it as a base64 encoded string. 21:07:05 Since strings in underload are more like atoms 21:07:13 (as far as I understood) 21:07:28 yep, no way of decoding except by running 21:07:37 correct 21:07:52 Underlambda takes that further, incidentally, there's no way to do anything with information in Tier 1 except by running it 21:08:08 if you want to get at the details of the code, you load a library that puts metadata into things 21:08:11 that is irritating. Makes it require more thought to be sure if it is TC or not. 21:08:12 it's a lot cleaner that way 21:08:29 um wait was it : and ^ i used or was that the other option i gave up 21:08:31 AnMaster: it's pretty easy to compile SKI into Underload 21:08:36 * oerjan checks wiki 21:08:38 ais523, true. 21:08:41 but in general 21:08:43 such languages 21:08:50 not really 21:09:00 Underlambda just uses the function from stacks to stacks as its basic data type 21:09:06 so it's very purely concatenative 21:09:12 whereas Underload is slightly dirtier due to S 21:09:27 (in Underlambda, S serialises a function, but the form of the serialisation is implementation-defined) 21:09:31 yeah it was : and ^, the other was : and a i think 21:09:44 ais523, I could compile befunge into underload easily if there is no input... 21:10:04 what, really? 21:10:09 I would have thought it was impossible due to ? 21:10:14 OUTPUT="$(./cfunge $1)"; echo "($OUTPUT)S" 21:10:15 ;P 21:10:16 although I suppose you could stick a PRNG in there somehow 21:10:19 AnMaster: those are as far as i can tell the only pairs of commands whose sequences can be cleanly decoded 21:10:28 also, that fails on infinite loops 21:10:36 ais523, true. But it was a joke 21:10:43 it was an /incorrect/ joke, though 21:10:52 ais523, meh. 21:10:55 besides, what if there are parens in Befunge's output? 21:10:59 you didn't even escape properly 21:11:12 ais523, it wasn't 1.0 duh :P 21:11:28 it was a "initial import" rather 21:11:59 ehird: http://www.tgdaily.com/html_tmp/content-view-42283-135.html 21:12:39 Deewiant: http://www.ramsan.com/products/ramsan-440.htm 21:12:42 600k. 21:12:44 I got you beat. 21:12:52 :-) 21:15:17 !underload (u)(~:S( )S:*~:^):^ 21:15:30 still no infloop printing 21:15:38 +ul (u)(~:S( )S:*~:^):^ 21:15:39 Deewiant, IOPS? 21:15:39 u uu uuuu uuuuuuuu uuuuuuuuuuuuuuuu uuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuu uuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuu uuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuu ...too much output! 21:16:05 ^ul (u)(~:S( )S:*~:^):^ 21:16:05 u uu uuuu uuuuuuuu uuuuuuuuuuuuuuuu uuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuu uuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuu uuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuu uuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuu ...too much output! 21:16:10 +ul (u)(~:S( )S:*~:^):^ 21:16:11 u uu uuuu uuuuuuuu uuuuuuuuuuuuuuuu uuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuu uuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuu uuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuuu ...too much output! 21:16:11 hm 21:16:15 ^ul (^^:^^^:^^^^^:^^^^^^:::^^^^^^^^:::^^^:^^^^::^)()~((())~:a~*):a~*~^!(~((!())(!:^(^)*)(!!:^(!^)*))~*^!!^):^(~((()())(:a~*:(*(!^)(:)S)~*~(!*(^)(^)S)~*):a~**((!^)~^!^)(!(^)~^^))~*^( )S!!a:(*)*~(~*)**^~*(()()(!)()(!)(:a~*:(!^(!^((!^)*)(!(^)*))(!^((^)*)(!(^)*)))~*~(!^(!^((!^)*)(!(^)*))(!^((^)*)(!(!^)*)))~*):^)~*^!!!!!!~:^):^ 21:16:17 ^^:^^^:^^^^^:^^^^^^:::^^^^^^^^:::^^^:^^^^::^ :^^^:^^^:::^^^::::^::^^::::::^::^^:^^^::^:^^ ^^:^^^:^::^^:^:::^^:^^^:::::^^:^^^^^:^:^^^^^ :^^^:^^^:^^^^^::^^^^^:^::::^^^^^:::^^^^^:::: ^^:^^^:^^^:::^:^^:::^^^:::^^:::^::^^:::^:::: ^^^^:^^^:^::^^^^^::^^:^::^^^::^^:^^^::^^:::^ :::^^^:^^^:^^:::^:^^^^^:^^:^:^^^^^:^:^^^::^^ ::^^:^^^: ...too much output! 21:16:26 oerjan, how... 21:16:29 AnMaster: I/Os per second 21:16:33 did you type random chars 21:16:42 I mean, how can you track that 21:17:05 AnMaster: that's the rule 110 automaton 21:17:10 ah 21:17:11 prepared 21:17:14 I see 21:17:20 multiline would make it easier 21:17:34 yes but the irc bots don't allow that 21:18:06 (it would require giving a newline in the input, for one thing) 21:18:27 oerjan, EgoBot can load from urls 21:18:35 fungot can't yet. it was planned however 21:18:35 AnMaster: " a far fnord one than mine, said arthur. " why, they're only a pack of cards!" 21:18:41 ask fizzie about ETA for tit 21:18:42 it* 21:18:42 it shows fine in irssi though, due to line wrapping 21:18:56 fungot: Arthur in Wonderland! 21:18:56 ais523: " but he needn't run over me!" or to any loud cry, such as " fnord me!" i said to myself " that's very curious!" she cried. ' yet i must sell my sunday fnord the fnord almost unnatural fnord which arthur met the woman who had won his heart, and outgrabe in despair, took to pointing out places for himself, and feebly asked " is that great yellow fnord fairyland?" 21:19:03 ^style 21:19:03 Available: agora alice* darwin discworld europarl ff7 fisher ic irc jargon lovecraft nethack pa speeches ss wp 21:19:05 ais523, why Arthur 21:19:11 AnMaster: I don't know, ask the bot 21:19:15 "Any loud cry, such as 'fnord me!'." 21:19:16 ah right 21:19:23 AnMaster: well for EgoBot only the first line would be on channel, too 21:19:31 oh right 21:19:40 -!- MigoMipo has joined. 21:19:41 oerjan, you can use it outside the printed bits 21:19:42 right 21:20:14 i considered making a "quine" version but there is an enormous amount of extra noise 21:20:36 oerjan, quine of what 21:20:50 isn't there some short one 21:20:50 the rule 110 automaton 21:20:53 forgot how 21:20:59 not a real quine 21:21:01 oerjan, ah you mean for input 21:21:03 right 21:21:15 but something that would do the automaton between two bots 21:21:31 ouch 21:21:38 oerjan, can you fit that in one line 21:21:40 actually i think they don't allow enough output for it 21:21:40 I doubt it 21:22:03 You can do length-unlimited fungot programs if you want to do a bit of scripting which uses ^str add in a privmsg. I don't think those have a length limit. (You can also use that to make fungot run out of memory, so don't do it.) 21:22:04 fizzie: how it happened. ' that fnord full of them. ' i'm sure mine only works one way,' and then its eyes looked fnord" and an " o." 21:22:43 oerjan, err (!^((!^)*)(!(^)*))(!^((^)*)(!(^)*))) <-- seems repeated there. Can't you use : somehow to cut that down? 21:22:49 not perfect copy 21:22:52 since one is nested 21:22:53 but still 21:23:13 surely it can be made shorter somehow 21:24:05 AnMaster: that's the encoding of the automaton rules 21:24:12 oh 21:24:19 oerjan, still, can't you make it shorter 21:27:31 maybe a bit of tweaking 21:27:55 ((^)*)(!(^)*) = ((^)*):(!)~* 21:28:36 a bit complicated for saving one char :D 21:30:07 in Underlambda, higher tiers have a few charsaving abbreviations 21:30:14 in particular, ` for ~^ and & for ~* 21:31:44 AnMaster: thing is, the representation of a bit in that tree depends not just on what the bit is, but also on whether it is left or right branch 21:31:55 oerjan, meh 21:32:40 ais523, hm... would be funnier it it looked like combined 21:32:47 like .' -> ! 21:32:49 in ic 21:32:50 ick* 21:32:53 yes, but that's less useful 21:33:00 oh, ' is another typesaving character 21:33:03 'c == (c) 21:33:11 lispy! 21:33:11 likewise for '^ == (^) and for any other character 21:33:18 ais523, ~^ 21:33:19 also, I suspect '(a) == ((a)) 21:33:20 err 21:33:24 because there's no other obvious meaning 21:33:25 can't you combine them 21:33:27 in unicode 21:33:32 like ^ over ~ 21:33:34 would be fun 21:33:40 but probably impossible 21:33:45 since both are "over" ones 21:33:47 AnMaster: or rather, every right branch contains an extra ! to get rid of the left branch still on the stack 21:33:58 AnMaster: but that would make the input /longer/ 21:34:11 oerjan, one char saved is one place up in anagolf gained! 21:34:14 ;P 21:34:42 ais523, only if you count bytes rather than chars 21:34:59 oerjan: if the rules were long enough, presumably it would be shorter to write a recursive !-adder 21:35:11 AnMaster: yes, but golfing servers do count bytes 21:35:22 true 21:36:37 ais523: oh and also there is a recursion on the stack, because you look at the last two bits and one bit, each of which have two elements on the stack for accepting next bit 21:36:47 iirc 21:37:20 You can add a combining ^ into the mathematical ~ operator, ∼. That should give a reasonably-rendered ~^-combination. Would look something like ∼̂. Maybe the ascii-~ would work too. Though combining characters tend to render uglily. 21:38:42 fizzie: is that actually a combining circumflex over a tilde? 21:38:54 It should be, but I might've gotten them in the wrong order. 21:38:57 it's a combining circumflex over something that looks like a tilde but isn't 21:39:02 why not just use a genuine tilde? 21:39:15 Because the TILDE OPERATOR is obviously greater than some crummy ASCII tilde. 21:39:40 I'll do both 21:40:01 Besides, the mathematical operators block has, for those situations where a regular ~ is too weak, also the ∾ -- "inverted lazy S = most positive". 21:40:59 Heh, there's also a Unicode codepoint with the official name "NOT TILDE", but sadly it's just a tilde-with-a-slash (≁) and not something that's really not tilde, like, say, a duck. 21:41:01 ~̂∼̂ 21:41:14 First one is an ascii tilde, second is a TILDE OPERATOR 21:41:18 the one on the left looks better to me, that's the ASCII tilde based on how it looks 21:41:29 They render pretty similarly here. 21:41:36 the only difference here is that the OPERATOR is antialiased 21:41:42 possibly a different font 21:41:51 * ais523 wonders why everyone is guessing that Jeff Atwood's password is "orange" 21:42:46 ais523: orange u glad that isn't your password? 21:43:01 oerjan: where does that joke come from, I never understood it 21:43:14 it's a knock knock joke iirc 21:43:30 substitution of "orange" for "aren't" isn't particularly obviously funny 21:43:36 although I've seen it several times in jokes 21:43:44 Here's the ASCII tilde with a combining tilde both above and below: ~̰̃. 21:43:46 for me it's become funny due to the repetition 21:43:54 AH 21:44:09 ais523: http://www.listphile.com/Best_Knock_Knock_Jokes_of_All_Time/Orange_Banana 21:44:42 ais523: also, knock knock jokes are _supposed_ to be puns like that 21:44:54 I still don't get the pun 21:45:02 hm the banana part isn't though 21:45:03 I know they're supposed to be puns 21:45:09 but orange doesn't sound anything like aren't 21:45:19 nor does aren't sound like something orange-related 21:45:24 ais523: "aren't you glad i didn't say banana" 21:45:25 ais523, so, any underlambda implementation yet 21:45:32 if not..., any spec 21:45:37 ais523: i suppose it depends on dialect 21:45:38 I might try it tomorrow 21:45:52 for the lower tires at least 21:46:06 I can't even imagine how "orange" could sound like "aren't" 21:46:16 ais523: aren'tchu 21:46:36 you have to mingle the two words 21:46:49 I still don't get it 21:46:55 * ais523 wonders if he's being meta-trolled 21:47:06 I get it 21:47:28 all the vowels are different, nearly all the consonants are different... 21:47:36 ais523: oh you're british, maybe you don't have an r sound in aren't? 21:47:39 Here's the ASCII tilde with a combining tilde both above and below: ~̰̃. <-- what about ~ with ^ above 21:47:49 and 21:47:57 how do you make those combining ones 21:47:57 AnMaster: loads of those have been pasted earlier 21:48:05 AnMaster: and via character map 21:48:09 well 21:48:10 you mean, you don't have a GUI program for htat? 21:48:13 what about the combiner one 21:48:14 ... 21:48:21 AnMaster: the combiners are there too 21:48:22 ais523, know one for KDE? 21:48:36 I've usually been using gucharmap, which is a gnomey sort of program. 21:48:36 not offhand 21:48:44 I'm sure there's a KDE thing too. 21:48:53 ais523: i suspect it works best in a texas accent, or something midwest US like that 21:48:55 It might even have a built-in character-select-o-tron, for all I know. 21:49:13 Just remember to stick combining characters after the thing you want them to combine with. 21:49:57 kcharselect, maybe. 21:50:15 will compile it and try it later 21:50:29 * AnMaster has a sparse KDE installation 21:50:51 any americans around who can tell us whether "aren't" really sounds anything like "orange"? 21:51:28 different in Oxford English I think 21:51:34 -!- MigoMipo has quit ("I must sleep!"). 21:51:56 *Maybe* in some exaggerated parody of Texas English... 21:52:48 but not in smug asshole english 21:52:58 thank God 21:53:00 so GWB gets it perfect? :) 21:53:05 *ly 21:53:44 GWB? 21:53:56 oh, if it needs an American accent, possibly I can imagine how it works 21:54:11 * oerjan should have seen that AnMaster comment coming 21:54:40 AnMaster: famous ex-president with bad speech 21:55:02 ah, Bush 21:55:03 right 22:03:26 ORANGE YOU GLAD I DIDN'T SAY BANANA? 22:03:29 Nobody remembers that? 22:03:41 GregorR: oerjan linked me to it, I'd never seen it before 22:04:05 Oh I was just responding to any americans around who can tell us whether "aren't" really sounds anything like "orange"? 22:04:31 GregorR, yay back to square one 22:04:43 that was the start of the discussion duh 22:04:50 Oh :P 22:04:53 lawlehcoptahs :P 22:05:00 ugh 22:05:03 I have a program looking for all potential 26-letter pangrams right now. 22:05:06 GregorR: ais523 being british couldn't imagine how that pun actually works 22:05:16 GregorR, pangram? 22:05:33 Fail Brittania (couldn't resist the pun urge :P ) 22:05:54 GregorR: there was a 26-letter pangram as the answer to a massive puzzle on Agora that lasted five weeks 22:05:55 Brittans fail the waves? 22:05:58 AnMaster: A sentence containing every letter of the alphabet, e.g. The quick brown fox jumps over the lazy dog. 22:05:59 that sounds weird. 22:06:02 This sentence contains 43 'a's, 9 'b's, ..., and 16 small jumping rats. 22:06:12 GregorR, that is a huge search space 22:06:19 Yup :P 22:06:24 oh not self-describing ones 22:06:32 "Zing! Vext cwm fly jabs Kurd qoph." 22:06:43 oerjan: I don't think there's a 26-letter self-describing pangram 22:06:45 And the trick is not to use lame ones like that :P 22:06:47 GregorR, you need to check if they are gramtically valid. 22:06:48 those Kurds are always getting it 22:06:58 AnMaster: Yeah, that's gonna be the PITA part X-P 22:06:59 ais523: well not in english. maybe Rotokas. 22:06:59 oerjan: actually, it's just a Kurd who drew the qoph 22:07:18 someone should write a minimum-length pangram in Chinese 22:07:22 GregorR, what about: The brown fox quick over jumps the lazy dog. 22:07:39 AnMaster: That's not a _26-letter_ pangram. 22:07:49 there's a minimum-length pangram in Japanese kana, IIRC, which makes a lot of sense and even scans correctly, IIRC they use it to order the alphabet 22:07:49 oh right 22:07:51 26 letter 22:07:53 ais523: @_@ 22:07:55 indeed 22:08:03 GregorR, any example of such as 26 one 22:08:05 AnMaster: Which is to say, contains ever letter exactly once. 22:08:13 or isn't it known yet 22:08:15 "Zing! Vext cwm fly jabs Kurd qoph." 22:08:16 GregorR: Been watching Code Geass or something? 22:08:21 err 22:08:24 I've seen two or three, but they're all hyper-lame. 22:08:26 are those valid words 22:08:28 ... 22:08:31 Yes. 22:08:35 AnMaster: yes, although some of them are rather obscure 22:08:45 Obscenely obscure :P 22:08:45 ais523, too obscure for aspell even 22:08:50 so what does it mean 22:08:50 that's one of the few isogram pangrams which actually makes grammatical sense 22:09:00 Nobody outside Wales would ever use "cwm" 22:09:06 AnMaster: "zing" is the noise a bullet makes as it whizzes past you 22:09:11 ok... 22:09:14 the other ones 22:09:15 I thought that was Welsh. 22:09:21 like all words but "fly" there 22:09:28 (I don't know of any other languages that use w as a vowel...) 22:09:30 pikhq: It's a Welsh loan word, yes. 22:09:37 ... 22:09:42 what does it mean GregorR 22:10:04 it's a sort of valley shaped like a semicircle 22:10:10 ok... 22:10:14 so the other words then 22:10:25 Vext cwm Kurd qoph 22:10:36 Vext is an old spelling of "vexed" 22:10:37 fly jabs I think I know, unless they mean something odd 22:10:40 which means confused, or annoyed 22:10:57 qoph is a letter in the Hebrew alphabet 22:11:01 More like angry, IMO 22:11:08 and a Kurd is a member of the Kurdish ethnolinguistic group 22:11:14 Deewiant: I thought so too, but I looked it up 22:11:18 Or I guess "annoyed" is it 22:11:20 ais523, ah right 22:11:22 It's just stronger 22:11:27 ais523: "Confused"? Really? 22:11:35 so. what does the entire sentence mean 22:11:40 Of course "vexed" means confused. 22:12:07 you didn't explain cwm 22:12:09 Confused flying valleys jab those instances of the letter qoph that are (quite inexplicably) Kurdish. 22:12:11 AnMaster: Zing! An angry fly that lives in a semicircular valley jabs the Hebrew letter qoph, as drawn by a Kurd 22:12:14 oerjan: I did, just earlier 22:12:17 Latin vexare - harass/annoy 22:12:24 GregorR: it's ambiguous, my meaning is slightly more sensible 22:12:26 Oh, I'm sorry, yes, fly as in a noun, bleh 22:12:28 ais523, ah 22:12:30 that's odd 22:12:34 Yes, ais523 is right. 22:12:42 -er 22:13:06 anyway 22:13:15 what about one using space at most once ;P 22:13:18 impossible I bet 22:13:47 so 22:13:52 how did anyone come up with that one 22:14:00 and second question 22:14:10 how the hell do you pronounce "cwm" 22:14:36 ais523, ^ 22:15:02 AnMaster: the w's a vowel, it sounds like about half an oo 22:15:12 oo as in book 22:15:14 or 22:15:16 what 22:15:20 yes, as in book 22:15:29 so like a single o 22:15:30 + 22:15:32 no 22:15:35 hm 22:15:37 IPA /kum/ 22:15:49 * GregorR wonders how AnMaster pronounces "book" :P 22:15:52 Deewiant, doesn't help me if I don't know how it is supposed to sound 22:16:00 Learn IPA :-P 22:16:07 GregorR, not like a double single o no 22:16:19 I just can't figure out how to pronounce half an o 22:16:31 AnMaster: Only the Welsh truly know. 22:16:32 Deewiant, is that u the Swedish u? 22:16:53 No, the Swedish u is /ʉ/ 22:17:00 oh ok 22:17:05 you mean u with blur on it 22:17:06 :P 22:17:20 Well, y'know, those Swedes speak in a blurry way. 22:17:29 too smal font 22:17:32 small* 22:17:42 I can't think of a Swedish word with /u/ :-/ 22:17:52 works great of ASCII and åäöÅÄÖ but not anything else really 22:17:54 Anyway, it's like the Swedish o 22:18:11 Deewiant, so Swedish o is /u/? 22:18:30 Yeah, I think it always is 22:18:33 hm 22:18:38 oh like that 22:18:47 that isn't like half of book 22:18:48 ... 22:19:01 No it's not 22:19:04 IIII'm a monoglot and I'm OH-KAY, I sleep all night and I work all day. 22:19:07 book is /bʊk/ 22:19:13 Deewiant, so ais523 was just misleading me then 22:19:16 right... 22:19:25 A 26-letter self-describing pangram: A, B, C, D, E, F, G, H, I, J, K, L, M, N, O, P, Q, R, S, T, U, V, W, X, Y, Z. 22:19:30 More likely he's confused about the vowel sounds in his own language :-P 22:19:50 Deewiant, I still can't figure out cwm, I mean... what is the c there. is it s or k sound 22:19:52 or some other 22:19:55 /kum/ 22:19:58 ah right 22:20:04 you said that above duh 22:20:06 C is only S before E, I and Y. 22:20:09 W is not E, I or Y. 22:20:14 Deewiant, so like "kom" in Swedish? 22:20:18 :D 22:20:37 Right, Swedish o is *not* always /u/ 22:20:43 And the letter "w" is almost always the long "oo" sound, I think. Like, um... 22:20:51 Deewiant, yeah it is more like å in there 22:20:52 What? 22:20:56 i vaguely recall welsh c is always k 22:21:10 "Goose". 22:21:21 Deewiant, eh? 22:21:42 Nah, I was confused 22:21:54 ok.... So I have something like 4 totally different descriptions of how cwm is pronounced now 22:21:55 ... 22:21:58 ! 22:21:59 /kum/ 22:22:21 Ideally, everyone would understand me if I called a goose a gwç. 22:22:27 heh, "The letter "k" was in common use until the sixteenth century, but was dropped at the time of the publication of the New Testament in Welsh, as William Salesbury explained: "C for K, because the printers have not so many as the Welsh requireth". This change was not popular at the time." 22:22:36 :D 22:22:36 Deewiant, and that is like "kom" but with the other o sound? As in "gol"? 22:22:52 AnMaster: in IPA, /u/ is the "oo" of "goose". 22:23:00 Huh? 22:23:01 so in other words, it isn't spelt "kwm" because old printers didn't have enough ks? 22:23:12 ... 22:23:19 Isn't "goose" /gʊ:s/ ? 22:23:20 ais523, now you are just being absurd. 22:23:29 AnMaster: I was replying to oerjan 22:23:32 * AnMaster wonders why "om" at end of words in Swedish make a short o sound. 22:23:39 oh right 22:23:43 Okay, maybe I'm weird. 22:24:05 ais523: so it seems :D 22:24:16 oerjan, awesome! 22:24:40 No, "goose" isn't /gʊ:s/, unless either http://en.wikipedia.org/wiki/Help:IPA_for_English is lying or my knowledge of English is very fail. 22:24:56 Wiktionary's IPAfication of "goose" is /guːs/ actually. 22:25:12 Hmm, perhaps I am confused 22:25:14 The word "foot" is /fʊt/. 22:25:33 Yeah, that's definitely a different sound 22:25:37 But I think /u/ is not what I thought it was 22:25:57 so spelled in English it would be koom? 22:26:07 (like in Koom valley?) 22:26:15 dictionary.com lists it as koom 22:26:26 uh uh 22:26:30 Yeah, it would probably be "koom". 22:26:36 I think I hit a hidden TP reference there... 22:27:00 Okay, I just don't get how Finnish "kuu" is /ku:/ while English "food" is /fu:d/ 22:27:03 I guess "koom valley" in the Discworld is a pun on cwm 22:27:03 then 22:27:05 Rhyming with "boom", not the first half of "woman". 22:27:26 Too bad I don't know any Finnish. 22:27:32 Is it Indo-European? 22:27:37 No, it's Finno-Ugric. 22:27:43 anyway, I love the word "cwm", because it's normally a giveaway that someone is attempting a pangram 22:27:53 Uralic! Wow. 22:28:03 Deewiant, isn't it the only language in that family? 22:28:08 Like Hungarian. 22:28:10 No. 22:28:20 I freely admit I probably wouldn't be able to differentiate between 'food' /fu:d/ and 'foot' /fʊt/ based on only the wovel; I mean, sure, there's a difference, but in the grand scheme of life, the universe and everything, it's not so big. 22:28:21 ah 22:28:26 AnMaster: Finnish, Estonian, Hungarian and a bunch of very-tiny languages 22:28:37 saami 22:28:38 Meänkieli, Saami, Mordva, etc. 22:28:41 I freely admit I probably wouldn't be able to differentiate between 'food' /fu:d/ and 'foot' /fʊt/ based on only the wovel; I mean, sure, there's a difference, but in the grand scheme of life, the universe and everything, it's not so big. <-- it is rather obvious to me 22:28:54 Deewiant, Mordva? 22:29:01 AnMaster: Very-tiny, like I said. 22:29:05 ah 22:29:11 I could differentiate between "wide" and "white" based only on the vowel. 22:29:19 Saami I heard of of course 22:29:29 * AnMaster hates when you get double "of" in English 22:29:40 Put a comma in between 22:29:44 Everyone hates when you get double anything in English. 22:29:55 kerlo, true 22:29:58 kerlo: fool versus food 22:30:03 Is that not a very different vowel 22:30:06 But sometimes we just have to accept that that that is is. 22:30:16 Deewiant, foot vs. food is more different 22:30:19 Deewiant: those vowels sound pretty much the same to me. 22:30:26 Foot is different from both. 22:30:28 AnMaster: Yes, but I'm not interested in that. 22:30:31 But sometimes we just have to accept that that that is is. <-- parser failure on three of them 22:30:47 two I can handle 22:30:49 but not three 22:31:13 AnMaster: but but but it's easy 22:31:16 Heh, typical Wikipedia style: "The term Finno-Ugric is somewhat controversial today[citation needed], with many historical linguists[who?] feeling --" 22:31:20 Noun phrase: that that that is is 22:31:23 Sentence: that that is is 22:31:37 AnMaster: /msg 22:31:48 Pronoun: that 22:31:52 Restrictive clause: that is 22:32:48 If you understand the phrase "that that is is", you just stick a complementizer in front of it and get the noun phrase found in "accept that . . .". 22:32:50 night in 1 minute 22:34:21 Spanish is less ambiguous: Pero a veces sólo tenemos que aceptar que lo que es es. 22:35:22 and norwegian even less: ... akseptere at det som er, er. 22:35:27 kerlo: "you" versus "rule"?! 22:36:04 -!- puzzlet has quit (Remote closed the connection). 22:36:08 -!- puzzlet has joined. 22:36:11 Both supposedly /u:/ 22:37:30 Aha 22:37:31 http://en.wikipedia.org/wiki/Advanced_%28phonetics%29 22:37:40 In English, the back vowel /u/ is farther forward than what is normally indicated by the IPA letter . This fronting may be shown explicitly, especially within a narrow transcription: [u̟]. Whether this is as far front as the central vowel [ʉ], or somewhere between [u] and [ʉ], may need to be clarified verbally. 22:37:51 No shit it's farther forward, it's a mile away :-P 22:38:20 You have one big-ass mouth, then. 22:38:25 those english always sticking out their tongue 22:38:27 (Do not move the dash.) 22:38:31 big ass-mou 22:38:32 dammit 22:38:33 :-D 22:40:03 fizzie: But seriously, am I the only one who thinks that "food" and fi:"kuu" have completely different vowel sounds 22:40:30 kuuld be 22:40:52 I'm not sure I'd go so far as to say "completely", but different, yes. 22:41:04 Of course my understanding of 'food' might be the wrong. 22:41:10 I honestly thought they'd be different even in broad transcription 22:41:15 I am reasonably sure I know what fi:kuu sounds like. 22:41:22 "you", then 22:41:33 I assume you know what that sounds like, in general 22:42:27 Although it does obfuscate a bit with the /j/ rolling into it 22:43:21 doncha comfuse yer prunciation, ye haer 22:44:10 I can plausibly believe a "you" that's exactly like fi:juu, with identical wovel to fi:kuu, but I'm not so sure I know what "you" should exactly sound like; I think if I were to say "you" and "food" they'd have somewhat different sounds, anyway. 22:44:30 I mean "plausibly believe that someone, esp. a Finnish someone, would say it like that". 22:44:46 Certainly a Finnish someone might, but no English one IMO :-P 22:45:13 ... Kuu? For food? 22:45:31 fi:kuu is made out of cheese. 22:45:35 That's kinda funny, since in Japanese, "Kuu" is a informal way of saying "to eat"... 22:45:49 (It is the same as en:moon.) 22:50:44 We could switch to a two-sound phonology -- say a Hamming-encoded bitstring formed out of 0 = IPA y, 1 = IPA ɑ, duration-insensitive -- that'd certainly make the acoustics side of speech recognition easier, for one thing. 22:51:02 With distinct stops between bits, of course. 22:51:16 No, I don't think we could. :-P 22:52:02 Deewiant: Well, y-ɑ-ɑ-ɑ-y-ɑ-y-y-y-y-ɑ-ɑ-y you. 22:52:23 Yay for me! 22:54:15 ASCII-ise that question mark? 22:54:45 Which question mark 22:55:42 The IPA thing that's not a y. 22:55:47 Ah, ɑ? 22:56:02 pikhq: http://en.wikipedia.org/wiki/Open_back_unrounded_vowel 22:56:05 It's the a variant without the hook-at-the-top thing. 22:56:37 A bit alpha-like. 22:56:52 Actually "latin small letter alpha" seems to be the Unicode name. 22:57:02 Ah. 22:58:35 -!- olsner has quit (Read error: 104 (Connection reset by peer)). 22:58:56 It's also encoded as A is X-SAMPA. 22:59:33 X-SAMPA is such a pain 22:59:46 Those X-SAMPA "b_<" and friends look like botched horizontal-smileys. 22:59:49 [jU\"fO@r\i@] 23:00:11 /yuˈfɔriə/ 23:01:11 Yeah, X-SAMPA is a pain. 23:01:21 What's more a pain is the notable lack of terminals that do Unicode right. 23:01:33 And they're the only programs that don't on my system. 23:03:09 Clearly you're using the wrong terminals. 23:06:00 rxvt-unicode is misnamed. 23:06:10 Must be rxvt minus unicode. 23:07:14 well, - _is_ minus 23:07:48 pikhq: urxvt doesn't do unicode? 23:08:31 It sure hasn't for me. 23:23:26 -!- BeholdMyGlory has quit (Remote closed the connection). 23:28:28 -!- FireFly has quit ("Later"). 23:30:17 It's been doing unicode very well for me. Well, I guess the combining-character rendering could use some work. But other than that, and even that stuff rudimentarily works. 23:34:24 One might need to give it reasonable fonts, though; I'm not sure what the default font search list looks like. I've been using a "URxvt.font: xft:DejaVu Sans Mono:size=8, xft:Kochi Gothic:size=8" (and identical boldFont) resource line. 23:35:37 fizzie: 8?! 23:36:37 Well, and I haven't really tried anything fancy, like switching direction with the right-to-left mark; but certainly it's been reasonably good in rendering fancy characters. 23:37:54 ehird: I started with 10, I think; used that about a year, switched to 9, used that one more year, and recently switched to 8. 23:38:01 fizzie: use 54 23:38:04 It's some sort of adaptation technique, I guess. 23:38:15 Maybe I can graduate to 7 soon. 23:38:47 pango's got incredible Unicode rendering 23:39:01 coppro: that's what SHE said! 23:40:08 ehird: doesn't work in that context 23:40:20 ais523: that's the point 23:40:23 :-P 23:40:27 I'm using size-9 font atm, I deliberately sized it down from the default 23:40:37 low dpi screen? 23:41:00 Heh, urxvt FAQ; "here is a more complete set of non-standard colors" (to replace the ones that look mostly like a VGA screen) "They have been described (not by me) as 'pretty girly'." 23:42:00 This one is low-ish in the DPI rankings; the reasonably common 94dpi. 23:43:56 fizzie: ITYM 96dpi 23:44:16 This house has some CRTs, my 100dpi screen, and the ole 84dpi screen 23:49:48 No, 94, officially; pixel pitch of 0.270 mm, translating to (25.4 mm/in) / (0.270 mm/dot) = 94.074... dot/in. I guess it might not be "reasonably common", though. 23:53:43 Hadn't noticed before that they're making "normal" laptops nowadays with 16" 1920x1080 displays (which means ~140dpi) -- I did know about the really high-DPI screens in some VAIO models and other "high-end" things, but that 16" one was a cheapo-laptop, some <1000 eur price. 23:54:14 mine's 1280*800 23:54:24 fizzie: macbook pros have 17" 23:54:39 iphone has 160dpi display 23:55:21 9 a i o pb qs rn te whyd xv 23:55:25 But that's the small. 23:55:41 Last looked at Apple laptop specs when getting the iBook G4. 23:56:19 (That one was 12" 1024x768, meaning a bit over 100 dots.) 23:56:31 Sleepity-sleep anyway. 23:56:34 bye