00:10:33 -!- Koen_ has quit (Quit: The struct held his beloved integer in his strong, protecting arms, his eyes like sapphire orbs staring into her own. "W-will you... Will you union me?"). 00:16:00 <^v> woooooooooooooooooooooooooooot http://puu.sh/7Xvui.png 00:16:31 <^v> ]xhhhhooooooooohhhhhhxooooooooxooooooxjjjxhoooohhhxhohhhhhhhxhhhhjjjhhhxhhhhooooooooohhhhhhxjjjxxjjjjjjjxjhhhhxjhhhhhhhhjjjhh~ 00:16:42 <^v> will add to wiki, but gtg soon 00:25:32 -!- Frooxius has quit (Read error: Connection reset by peer). 00:25:45 -!- copumpkin has quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…). 00:28:03 -!- Frooxius has joined. 00:59:05 -!- conehead has quit (Quit: Computer has gone to sleep). 01:27:05 -!- ^v has quit (Read error: Connection reset by peer). 01:27:16 -!- ^v has joined. 01:33:03 -!- Phantom__Hoover has quit (Ping timeout: 268 seconds). 02:09:01 -!- nooodl has quit (Quit: Ik ga weg). 02:17:01 seems like every time i turn around my iptables configuration has been deleted. :( 02:23:36 sounds like a really boring poltergeist 02:32:23 -!- nisstyre has joined. 02:41:36 might be the upgrade installer. dunno. making an extra copy of the rules in case it happens again 02:45:04 -!- ter2 has joined. 02:49:03 -!- tertu3 has quit (Ping timeout: 252 seconds). 02:58:51 fizzie: you should add the snarxiv score indicators 02:59:16 (I think 30% was "9th year grad student") 03:00:00 Also, this is genuinely harder than snarxiv 03:00:01 Do you know how to recognize the music of a record by observing the grooves? 03:01:44 In other news, I'm sad CMONFLICTOR isn't a real opcode 03:07:09 -!- Sorella has quit (Quit: It is tiem!). 03:13:23 -!- conehead has joined. 03:14:49 zzo38: that's essentially how CDs work. I'm surre one could readily build an optical needle 03:15:22 and...it's already been done 03:15:33 http://en.wikipedia.org/wiki/Laser_turntable 03:15:57 that's so zeerust a concept i can't handle it 03:16:06 I read a book that someone could recognize a lot of orchestral music on a record, by only observing the grooves, not using any special equipment. However, only post Mozart music, but still that is a lot of possibilities 03:16:44 i imagine there are repetitive patterns, especially if you're listening to some form that's really regimented 03:17:51 I would certainly believe that it might be possible. 03:18:54 probably amplitude variation over time would be enough. i don't doubt there are many more who can differentiate pieces looking at the waveform 03:19:09 Yes, maybe it is. 03:19:18 I don't know exactly, since I don't know how to do it myself. 03:20:47 It says here: "Arthur Lintgen ... apparently possesses the remarkable ability to identify the music on a phonographic record without referring to its label ... by looking at the grooves on the surface of the disc ... the music must be classical, orchestral, and post-Mozart ... this leaves a body of recorded music that runs into thousands of records." 03:21:07 -!- conehead has quit (Read error: Connection reset by peer). 03:22:28 James Randi was commissioned to test it. He figured it out, even multiple recordings of the same music, and an extra cut on the disc, and others. However, two of them were something different: rock music, and a lecture. He called the rock music "gibberish" and the lecture "a vocal solo of some kind". I suppose this is to be expected if you don't understand that kind of "music"... 03:22:52 It is clear to me that this is nothing "paranormal". 03:23:44 well, seeing as how nothing people actually can do can be called "paranormal"... 03:26:03 So it would seem. 03:27:47 There are other reports of things, much of which are exposed as hoaxes, frauds, or something else, but some which don't know such thing quite yet, and which would seem to be "paranormal". I wasn't there, so I don't know if these reports are just made up, but you cannot disregard things out of hand. Once people thought, stones cannot come from the sky, but now they figured it out. So, you cannot just ignore something with no good reason. 03:28:27 "The prototype revealed an interesting flaw of laser turntables: they are so accurate that they play every particle of dirt and dust on the record, rather than pushing them aside as a conventional stylus would." 03:28:38 Extraordinary claims require extraordinary evidence, but, doesn't mean it can ignore everything 03:28:49 Jafet: Interesting idea. 03:29:01 Jafet: but all SERIOUS audiophiles store and play their records in an industrial clean room, so it's no big issue 03:29:03 I didn't know, but yes of course if it a laser, clearly that is how it works! 03:29:47 Would be fun to use a record of 4'33" 03:30:09 I don't think industrial clean rooms tend to have good acoustic qualities 03:30:31 why not? theoretically you can design your clean room any way you want to 03:30:32 it's not musicians' fault that semiconductor manufacturer have no taste. 03:44:34 Then, you need to design, the musical clean room. 03:44:54 i already did. 03:46:02 emptying spit valves is gonna suck, i can already tell 03:47:07 sorry, no wind or brass instruments allowed...you can't play them with a clean suit on 04:12:41 -!- ^v has quit (Ping timeout: 252 seconds). 04:15:52 -!- ^v has joined. 04:46:26 -!- tertu3 has joined. 04:49:46 -!- ter2 has quit (Ping timeout: 268 seconds). 05:20:45 -!- ter2 has joined. 05:21:03 -!- tertu3 has quit (Read error: Connection reset by peer). 05:25:00 -!- john_metcalf has quit (Read error: Connection reset by peer). 06:18:40 -!- tromp has quit (Remote host closed the connection). 06:19:17 -!- tromp has joined. 06:23:45 -!- tromp has quit (Ping timeout: 255 seconds). 06:35:10 -!- ^v has quit (Quit: http://i.imgur.com/DrFFzea.png). 06:39:26 -!- nisstyre has quit (Quit: WeeChat 0.4.3). 06:51:18 Play non-brass music, then. 06:51:28 (If you like non-brass music) 07:11:49 -!- augur has joined. 07:23:33 -!- zzo38 has quit (Ping timeout: 240 seconds). 08:09:09 -!- ter2 has quit (Ping timeout: 252 seconds). 08:10:11 -!- ter2 has joined. 08:20:53 -!- ter2 has quit (Ping timeout: 252 seconds). 08:21:07 -!- ggherdov_ has quit (Changing host). 08:21:07 -!- ggherdov_ has joined. 08:21:07 -!- ggherdov_ has quit (Changing host). 08:21:07 -!- ggherdov_ has joined. 08:27:50 Jafet: I was thinking that there should have been some sort of a complicated scoring mechanism with multipliers and all that. 08:42:02 -!- oerjan has joined. 09:06:42 -!- stuntaneous has joined. 09:09:21 -!- Vorpal has joined. 09:20:09 "I heard it through the grapefruit / not much longer would you be mine. / Oh I heard it through the grapefruit, / Oh and I'm just about to lose my mind." 09:20:39 (Ate a grapefruit for breakfast.) 09:23:24 quintopia: Maybe some kind of a bellows-driven trombone. 09:23:43 Vorpal: http://zem.fi/2014-04-05-opquiz 09:24:47 fizzie: your interpretation makes the losing mind part so much clearer. 09:26:08 Good point. 09:38:48 hi 09:39:09 lo 09:39:13 fizzie, nice photos again 09:39:32 -!- Tritonio has joined. 09:39:33 fizzie, I like the "Pixels most distant from the mean, smoothed." a lot 09:40:27 fizzie, also nice opcode quiz 09:41:29 fizzie, really? There are real opcodes with underscore in them? 09:41:59 At least as accepted by NASM. 09:42:03 huh 09:42:12 Well so far I have 19/23 correct 09:42:16 Not too bad I think 09:42:37 VFNMADD213SS is real? Really? 09:42:46 There should be like a link to what each real one does 09:43:32 fizzie, also, no ARM? 09:43:37 Yes, I would've liked that, but it would've been hard to collect, since they're all spread out. 09:44:04 And ARM opcodes were in an overly complicated from in the Binutils sources, all kind of custom %x expandables. 09:44:11 fizzie, ARM and AVR are probably the only other two I have a chance of 09:44:17 of managing* 09:45:10 Yeah, about 80% success rate is what I manage on x86 09:45:54 The underscores are at least official Intel nomenclature, as per the Intel Architecture Instruction Set Extensions Programming Reference. 09:47:12 Hm 09:47:19 Yeah x86 is crazy 09:47:37 I think the 80% correct was a fluke, now it seems to stay around 70% 09:48:03 AVR I think I could manage, mostly because it is such a simple instruction set 09:48:35 And VFNMADD213SS is "Fused Negative Multiply-Add of Sclaar Single-Precision Floating-Point Values"; VFNMADD213SS xmm0, xmm1, xmm2 does x0 = x2 - x0 * x1 where xN is the single-precision float in the low 32 bits of xmmN. 09:48:52 The fuck intel! 09:49:21 (The numbers -- 132, 213 and 231 are all valid -- denote the way those three operands go.) 09:49:38 Hm 09:50:03 There's some kind of logic in there, though I can't quite decode it immediately. 09:50:20 SS is probably scalar single or something like that 09:50:36 The opcode for VFNMADD213SS is VEX.DDS.LIG.128.66.0F38.W0 AD /r. 09:51:04 The hell? Is this an alias for a lot of prefixes basically? 09:51:34 It's some way of denoting all the parts of the VEX prefix, yes. 09:52:05 Which flag is support for this sort of stuff in /proc/cpuinfo? 09:52:10 "VEX.[NDS/NDD/DS].[128,256,L0,L1,LIG].[66,F2,F3].0F/0F3A/0F38.[W0,W1,WIG] opcode [/r]" is the full format. 09:52:40 "CPUID Feature Flag" column lists "FMA" in the manual. 09:52:55 I mean the VEX stuff in general 09:53:21 flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc aperfmperf eagerfpu pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic popcnt tsc_deadline_timer aes xsave avx lahf_lm ida arat epb xsave 09:53:21 opt pln pts dtherm tpr_shadow vnmi flexpriority ept vpid 09:53:37 Heh, that is a lot, also that is "xsaveopt" as a single entry in the line-break 09:53:38 "avx", probably. 09:53:45 That's what introduced the VEX prefix, AIUI. 09:53:58 Well it is there I guess 09:54:02 (Or AVX512F if you want to use the "VFMADD132SS xmm0 {k1}{z}, xmm1, xmm2/m32{er}" form.) 09:54:17 It's a Sandy Bridge 09:54:39 "flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc aperfmperf eagerfpu pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 fma cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx ... 09:54:45 ... f16c rdrand lahf_lm abm ida arat xsaveopt pln pts dtherm tpr_shadow vnmi flexpriority ept vpid fsgsbase tsc_adjust bmi1 hle avx2 smep bmi2 erms invpcid rtm" 09:55:11 Ivy Bridge? 09:55:17 Given that it has more stuff 09:55:28 Haswell, even. 09:55:51 ("avx2" is a tipoff, technically.) 09:56:20 "flags: vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 syscall nx rdtscp lm constant_tsc up rep_good nopl pni monitor ssse3 lahf_lm" 09:56:22 Ah 09:56:25 What is ssse3 09:56:42 Jafet: It's what came in-between SSE3 and SSE4. 09:56:49 (Supplemental Streaming SIMD Extensions 3.) 09:57:00 Jafet, is that your entire flags? Fairly old I guess? 09:57:16 Though I think I can beat you, even if it is a bit of cheating: 09:57:18 Features: swp half thumb fastmult vfp edsp java tls 09:57:22 (That is a RPi) 09:57:25 That's under VT-x 09:57:29 Ah 09:57:54 flags: fpu de tsc msr pae cx8 sep cmov pat clflush mmx fxsr sse sse2 ss ht syscall lm constant_tsc up rep_good nopl pni ssse3 cx16 movbe hypervisor lahf_lm dtherm 09:58:00 (Intel Atom 230, not virtualized.) 09:58:37 What is dtherm? 09:59:16 No idea. Could be something Atom-specific, I guess. 09:59:35 Oh, no, it's also in these bigger flag lists. 10:00:13 Yeah they aren't sorted, so it is kind of hard to find 10:00:27 I guess they are sorted somewhat chronologically. 10:00:45 Btw, I have an old P3 that I could boot to check the flags, but I'm not sure I care, the power isn't connected and so on 10:00:52 CPUID bit order, presumably. 10:01:22 Which would be reasonably chronological 10:01:37 This system does not have a "flags" entry in cpuinfo at all, does that count as a list of length 0? 10:02:40 fizzie, RPi? I think we can use the "Features" then 10:02:43 like I did above 10:02:53 There's no "Features" either. 10:02:57 What architecture? 10:02:59 "system type: 96368MVWG" "cpu model: BCM6368 V3.1" 10:03:04 MIPS? 10:03:05 Yes. 10:03:11 Hm 10:03:18 Oh, there's "ASEs implemented:" but it's empty. 10:03:22 And I don't know what an "ASE" is. 10:03:31 But it sounds like a list of some kind of extensions. 10:05:00 dtherm is apparently the "Digital Thermal Sensor Capability" bit (CPUID function 06h, bit EAX[1]); as far as I can tell, it's what the coretemp thing reads. 10:05:50 Ah 10:07:17 "Features: swp half fastmult vfp edsp neon vfpv3" this phone has a list one element shorter than the RPi. 10:07:55 (Also somewhat different; no thumb, and noen + vfpv3 in place of java + tls.) 10:08:21 I'm pretty sure it can run Thumb code, so I don't know why there's no Thumb flag. Maybe it's a kernel thing. 10:09:13 "bogomips: 4548.57" 10:09:21 Best unit ever. 10:09:43 398.33 BogoMIPS on the MIPS box. 10:09:55 Hah 10:10:04 BogoMIPS: 2.00 10:10:08 that is the RPi 10:10:31 bogomips: 6621.52 is my desktop 10:10:37 Sounds like a bogus result on the RPi. 10:10:42 Quite so 10:10:44 what the hell are bogomips 10:11:05 I have 249.96 on the phone, 1993.93 on this Android tablet and 6799.32 on the desktop. 10:11:16 Is that, like, more bogus than bogomips 10:11:20 myname, number of NOPs per second as measured early during boot by the kernel to calibrate some internal delay loop iirc 10:11:35 Not sure why it is 2 on the RPi though, that seems wrong 10:11:51 i have to check that :D 10:12:05 https://en.wikipedia.org/wiki/BogoMips 10:12:28 I seem to recall I had some system where the BogoMIPS number changed according to cpufreq scaling. Doesn't seem to happen on this desktop, though. 10:13:41 fizzie, constant_tsc probably counteracts it (the feature that results in that flag that is) 10:13:53 Pretty sure it scaled back on older computers yes 10:26:36 -!- yorick has joined. 10:26:45 "The reasons (there are two) it is printed during boot-up is that a) it is slightly useful for debugging and for checking that the computers caches and turbo button work, and b) Linus loves to chuckle when he sees confused people on the news." 10:27:46 Good old turbo button. 10:28:14 Yeah 10:28:37 Is that the button that made your CPU slower 10:28:47 Quite often, yes. 10:29:42 Though I vaguely recall one system that was wired to be faster (16 MHz vs. 8 MHz) with the button pressed. 10:30:44 I guess BogoMIPS is a bit like glxgears then, useless except to tell that your system isn't completely out of whack 10:33:29 "158 frames in 5.0 seconds = 31.595 FPS" 10:33:36 int-e, elliott could you make lambdabot join #cs-york for a bit? 10:33:48 We should really have GPU virtualization by now. 10:34:13 I think Nvidia did have some stuff towards that direction? 10:34:22 For their computing-oriented hardware, anyway. 10:34:44 @join #cs-york 10:34:45 Jafet, we should, but most people who virtualize don't game 10:34:51 At least on the same machine 10:34:55 int-e, thanks 10:34:58 Two different markets 10:36:18 i recall glogbot accepts channel invites, how bad would it be to do the same for lambdabot 10:37:10 Nvidia Grid is fairly new, I think 10:37:30 Also, they loves segmentation 10:38:05 oerjan: I don't know. Are there honeypot channels here where joining gets you killed? ;-) 10:38:08 (i assume the main worry would be if lambdabot gets too overworked) 10:38:15 * int-e is too paranoid for that. 10:38:31 int-e: well yes but you have to be _on_ them to invite, no? 10:38:48 i'm talking about the technical invite which only a channel op can do, afaiu 10:38:49 probably 10:41:31 oerjan: I don't know. Are there honeypot channels here where joining gets you killed? ;-) <-- think so yes 10:42:12 oerjan, I think you only need to be op of +m is set? 10:42:28 Vorpal: oh 10:42:43 oerjan, if you /part, we can test I guess 10:42:58 -!- fungot has left. 10:43:07 [13:43:54] -!- #esoteric You're not a channel operator 10:43:10 Hm okay 10:43:13 (When trying to invite fungot.) 10:43:16 * oerjan #esoteric :is already on channel 10:43:19 -!- fungot has joined. 10:43:26 I guess the checks happen in a weird order 10:43:30 i was about to part when i saw fungot do it 10:43:30 oerjan: ( though its experimental and instrumenting) ' it) 10:43:37 from that kernel onward the BogoMips rating for then current Pentium CPUs was twice that of the rating before the change. The changed BogoMips outcome had no effect on real processor performance. 10:43:41 hahahaha 10:44:58 I imagine the patch name was "improve bogosity of bogomips" 10:46:30 http://www.amazon.com/dp/B00CA5MLCS Enterprise ready GPU virtualization. At least they're cheaper than some Xeons. 10:47:05 Jafet, are those Telsa series with no actual display connectors? 10:47:08 Or real GPUs? 10:47:30 Telsa. Cute. 10:47:43 So useless for non-HPC then, right 10:47:46 (For a moment I was wondering whether that was the real name.) 10:48:42 Tesla? Or what 10:48:44 Honestly, there are too many codenames for all the GPUs and CPUs, who's expected to track them? 10:49:04 Vorpal: Something discovered recently: if you have a channel with modes +if (invite-only, forward on; the latter needs the former), and the person trying to join the channel is either banned or already on the target channel of the +f, the error message given is the "cannot join, invite-only" one instead. 10:49:33 We have some Tesla cards in the local computing cluster of the university. 10:49:40 int-e, Tesla is a product line. Like i5 and i7 or GeForce and Quadro 10:50:03 Vorpal: no display connectors, but they are intended for graphics 10:50:23 Jafet, I thought Tesla were intended for GPGPU? 10:50:34 Also, a better example is Celeron 10:50:39 "10 compute nodes gpu[001-011] are HP SL390s G7 for gpu computing. Same configuration as above but they are 2U high and have 2x Tesla 2090 card each." 10:50:40 Oh? 10:50:42 I think some new CPUs are still Celerons. 10:50:57 Really? I thought i3 replaced that segment 10:51:17 Vorpal: as I said, who's supposed to keep track. I've heard of "Tesla" and not of "Telsa" but took a moment to make the connection, because there was nothing obviously wrong with the latter name. 10:51:27 No, i3 is the lowest rung of the high-end Core series. 10:51:40 int-e, oh I didn't even notice my typo 10:51:45 Intel makes plenty of smaller CPUs. 10:51:56 int-e, not even when you repeated it 10:52:10 I didn't notice the typo either, and the conversation was all manner of confusing. 10:52:16 fizzie, same 10:52:22 Vorpal: the human brain is a marvellous thing ;-) 10:53:14 bbl 10:54:08 https://en.wikipedia.org/wiki/Celeron I think every Intel generation since P6 is represented here 10:55:51 -!- ais523 has joined. 10:56:45 * int-e notes with joy that KICK and INVITE have their operands swapped. 10:56:49 (Sometimes "smaller" just means "we turned half the cache off") 10:57:19 err, parameters to stick to the RFC wording. 10:58:14 So implementing this is not too hard (at most 5 lines of code), I just wonder whether I want to have that feature. 10:58:37 POWER TO THE PEOPLE 10:58:49 this is the internet 10:58:51 Or non-people, as the case may be 10:58:55 I don't trust people ;-) 10:59:06 Do you trust fungot 10:59:06 Jafet: its pretty easy to use. that always does the right if 90 is straight down then it should 10:59:19 fungot has not done me any harm yet 10:59:19 int-e: no niinku pl niinku fnord niinku fnord fnord fnord 10:59:26 ^style 10:59:26 Available: agora alice c64 ct darwin discworld enron europarl ff7 fisher fungot homestuck ic irc* iwcs jargon lovecraft nethack oots pa qwantz sms speeches ss wp youtube 10:59:53 fungot never talks to me like that. 10:59:53 Jafet: heh. ( fnord tsh) ( seems ( ( prefix notation) confusing)) to change it to create a whole breed of new fnord languages. 10:59:56 Ah, it's the recursive style. 11:00:16 No, that would be "fungot". 11:00:16 fizzie: dpkg -i is ok to count your own artificial language when asked how many languages the fnord speak 11:00:19 ^style irc 11:00:19 Selected style: irc (IRC logs of freenode/#esoteric, freenode/#scheme and ircnet/#douglasadams) 11:00:34 Which is still dominated by the f word. 11:00:34 The small bits of Finnish come from ircnet/#douglasadams, where there occasionally was a bit of it. 11:01:15 No statements of the f are part of the "irc" style, since the model was trained before the bot existed. 11:01:38 Oh, you meant the other f word. 11:01:42 Yes, there's that mapping. 11:02:39 `addquote fungot has not done me any harm yet int-e: no niinku pl niinku fnord niinku fnord fnord fnord 11:02:39 Jafet: fnord it. :p it could be managed with the right hand 11:02:40 1183) fungot has not done me any harm yet int-e: no niinku pl niinku fnord niinku fnord fnord fnord 11:03:10 so what does "niinku" mean? 11:03:58 Oh it's just a filler, like, that. 11:04:00 It's a contraction of "niin kuin", which is approximately something like "as if". 11:04:04 Or "like". 11:04:18 The original source of the "niinku" line seems to be a DRUGZ reference. :/ 11:05:18 (Most of the "fnords" were different Finnish nicknames for hashish.) 11:09:54 ^style drugs 11:09:55 Not found. 11:11:03 a couple of those are mysterious to me. Agora, for example. 11:11:27 ^style sms 11:11:27 Selected style: sms (National University of Singapore SMS corpus, 2011-08-20) 11:11:37 fungot, is that as bad as I imagine it would be? 11:11:37 int-e: can i use short form to. thanks 11:11:48 fungot, please don't 11:11:48 int-e: only one place can sit to bishan, take the barcap... all thebest!! :) its time bah, cause i was thinking of coming to soc. btw how to go downfron bugis. u lookfor place 11:12:05 ^style irc 11:12:05 Selected style: irc (IRC logs of freenode/#esoteric, freenode/#scheme and ircnet/#douglasadams) 11:12:13 could be worse. 11:12:27 That was a reasonably tame example of the SMS style. 11:12:32 It's p. bad occasionally. 11:13:01 worse than a f n o r d loop? 11:13:31 Matter of taste, I guess. I don't think it has those at all. 11:13:39 But more of that "u lookfor" kind of stuff. 11:14:03 http://www.cartoonbrew.com/ideas-commentary/sony-demands-removal-of-open-source-indie-short-sintel-from-youtube-98182.html 11:22:30 Wow, a corporate video host stops hosting a video. Much censor. 11:22:43 back 11:23:16 I guess Sintel is sort-of "famous" in that it was one of the first entirely open-source-rendered CGI videos 11:23:26 it's mostly a tech demo 11:23:44 * int-e notes with joy that KICK and INVITE have their operands swapped. <-- Yes, there are more cases like that in IRC iirc. I wrote an IRC bot once and I noticed quite a lot of strange things 11:24:50 Some other post said Sony added Sintel to their "4k demo reel" kind of thing, and speculated that led to the (presumably automatic) takedown request. 11:25:38 fizzie, fnord was basically unique words right? 11:25:43 Right. 11:26:13 For the old models; the new ones, trained with VariKN, mostly have just discarded them, I think. 11:26:54 Hm 11:27:02 fizzie, what did you use the train the new models? 11:28:22 Uh, VariKN. 11:28:45 This computer combined with Debian Stable has a really interesting bug btw. If I pause playback in vlc, sound will stop working on resume, I have to do a jump in the video/audio stream to reset it. Either backwards or forward, doesn't matter how long either. 11:28:58 fizzie, err I meant the old ones 11:29:31 Oh. It doesn't have a name, it was just a custom piece of C++ (of all things) doing a fixed-length unpruned N-gram. 11:30:20 (Rather crummy code, too.) 11:31:06 Well it is C++, good look not being crummy 11:33:34 http://sprunge.us/SLKb?c++ I guess it's not that bad, but it's not that good either. 11:34:02 Also what's really silly is that the only code to test a generated model was in the same program, so I needed to rebuild the whole thing to generate sample sentences. 11:34:03 That is interesting indentation 11:34:11 Or is that sprunge messing up tabs? 11:34:27 Yeah seems like it 11:35:15 -!- Phantom_Hoover has joined. 11:38:06 It uses the time-tested approach of "just throw strings" to C++ exceptions. 11:38:22 (Or, rather, "just throw string literals".) 11:38:54 Pretty sure that is not considered good practice 11:39:10 it's better than throwing malloc'ed strings 11:39:12 That is, secretly, what most C++ exceptions are made of 11:39:25 (although even that is better than a mix of throwing malloc'ed strings and string literals) 11:39:48 On the other hand I have never been a fan of exceptions in general. There is often a better solution 11:39:49 Throwing malloc'd strings is horrible. You should be using new. 11:40:14 Vorpal: they are great for cross-function control flow 11:40:19 Jafet: I was trying to think of the most horrible sort of exception I could off the top of my head 11:40:48 although I have a Java program that throws integers (specifically speaking, a class deriving from Exception with only one int member of its own) 11:40:52 because it was translated from C 11:41:13 I have a factoring program that throws any factors discovered 11:41:18 Jafet, How? 11:41:35 Generally I prefer co-routines for that 11:41:45 (Assuming the language allows that of course) 11:41:54 (Also assuming you mean the same thing as I do) 11:42:06 can be simulated pretty easily, even in c 11:42:19 I know, but not elegantly really 11:42:43 yeah, that or non-re-entrant 11:42:47 The program has several layers of function calls, which makes simulating coroutines unpleasant. 11:43:03 nortti, also thread safety 11:43:25 I could use a lambda object as a continuation, but it gets annoying to pass it around. 11:50:19 -!- stuntaneous has quit (Ping timeout: 264 seconds). 11:52:34 `run allquotes | tail -1 11:52:34 1183) fungot has not done me any harm yet int-e: no niinku pl niinku fnord niinku fnord fnord fnord 11:52:54 `run sed -i '1183s/ No output. 11:52:59 `run allquotes | tail -1 11:52:59 1183) fungot has not done me any harm yet int-e: no niinku pl niinku fnord niinku fnord fnord fnord 11:53:49 `? qdbformat 11:53:50 qdbformat is: message; * nick action; two spaces between messages; all elisions marked with [...] other than irrelevant intervening messages; for messages separated by elision, one space on each side, not two 11:54:44 hth 11:57:54 ​`addquote oerjan says:`? qdbformat[...]oerjan says:hth 11:58:54 What is this evil quote format? 11:59:32 It allows us to achieve great wisdom.pdf 11:59:42 (that's my guess) 11:59:53 * int-e is too lazy to check the repo. 12:00:01 -!- Tritonio1 has joined. 12:00:19 the format is considerably older than the pdf hth 12:00:35 tdnh 12:00:38 -!- Tritonio has quit (Ping timeout: 240 seconds). 12:01:29 Because it does not answer this question: Where else is the format exploited? 12:02:31 nowhere afaik 12:03:11 it's simply a measure of consistency. it also helps a bit when quotes themselves contain quotes. 12:04:15 not in every case though. 12:04:20 `quote django 12:04:21 262) django is named after a person? thought it would be a giraffe or something \ 307) `quote django ​352) django is named after a person? thought it would be a giraffe or something thankfully only one \ 308) `quote django ​352) django is named a 12:15:39 -!- boily has joined. 12:20:38 -!- nooodl has joined. 12:29:18 -!- Tritonio1 has quit (Quit: Tritonio1). 12:40:53 -!- MindlessDrone has joined. 12:46:30 `run quote django | wc -l 12:46:30 8 12:47:05 `cat bin/allquotes 12:47:06 ​#!/bin/sh \ nl -w 1 -s ') ' quotes 12:47:47 `run tail -n 1 quotes 12:47:47 ​ fungot has not done me any harm yet int-e: no niinku pl niinku fnord niinku fnord fnord fnord 12:47:51 I see 12:57:26 now that HackEgo seems to be alive and well, I should get back to updating the Wisdom... 12:57:52 does it have logs yet? 12:58:09 `pastlog acciaccatura 12:58:39 ​/hackenv/bin/pastlog: 2: cd: can't cd to /var/irclogs/_esoteric \ ls: cannot access ????-??-??.txt: No such file or directory 12:58:55 nope :-( 12:59:04 your logs are in a different castle 13:00:42 lexande: hellexandello. my name is boily. you don't have a Wisdom Entry describing you. prepare to be wisdomified. 13:04:55 What happened to the logs? 13:05:10 they're still there, they're just not inside HackEgo's VM 13:05:11 elliott: you are vile. you rmed the *elcome*s. 13:05:15 so HackEgo can't see them 13:05:21 `welcome what about this one 13:05:21 Ah 13:05:21 ​/home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: welcome: not found 13:05:26 wow, even that one? 13:05:36 Did the VM get reset? 13:06:16 it's a "new" VM, new as in no more than several monthd old 13:06:21 *months 13:06:43 What caused the revert? 13:06:53 I thought it was all in hg so it would be easy to restore a backup 13:07:34 ais523: um elliott deleted the welcomes in a fit of ...something... 13:07:43 well the binaries. 13:07:54 Vorpal: the filesystem wasn't reverted 13:07:58 elliott deleted the welcomes in a fit of binaries??? 13:08:02 just the VM is on a different physical computer 13:08:09 not the same one that's doing the logging 13:08:18 Hm 13:09:58 yes the logs were never inside the repository, just mounted on the same machine 13:11:06 although the codu.org webserver obviously has access to both 13:11:28 (still) 13:12:04 `run df 13:12:05 df: cannot read table of mounted file systems: No such file or directory 13:12:39 `run ls /var 13:12:39 ls: cannot access /var: No such file or directory 13:12:42 `run ls / 13:12:43 bin \ dev \ etc \ hackenv \ home \ lib \ lib64 \ opt \ proc \ sbin \ sys \ tmp \ usr 13:13:58 ais523: the spanish welcome is still there, but elliott tells me all the recurring spanish people are the same anyway 13:18:33 this is a spanish-language channel, after all 13:19:42 elliott: well that welcome has been tailored specifically to make it clear we're not. 13:21:23 it's a shame, really. i was somehow imagining somewhere a venezuelan site with a convenient list of irc channels people could try, which had somehow been made by grepping for es in channel names 13:22:09 oh, that's brilliant 13:22:12 I want to believe that now 13:22:30 not *all* the spanish people are one person 13:22:36 there was just one particularly persistent one over the last few months 13:22:51 I think the "es" hypotesis is plausible. 13:23:30 elliott: are all the venezuelans the same? 13:23:47 um, not sure 13:24:42 good, good, we can continue to believe then 13:29:10 oerjan: If you want to get all technical about it, the HackEgo hg browser is actually running on www2.codu.org, a different VPS (the one HackEgo also runs), and www.codu.org just (AFAIK) proxymates the access. 13:30:02 s/runs/runs on/ 13:30:46 ah. 13:33:40 -!- metasepia has joined. 13:33:50 ~metar ENVA 13:33:50 ENVA 061320Z 26007KT 9999 FEW040 SCT070 08/01 Q1007 NOSIG RMK WIND 670FT 36002KT 13:34:16 ~yi 13:34:16 Your divination: "Obstruction" to "Clustering" 13:39:03 b_jonas: sorry about you getting caught in the spam filter, that's probably the first false positive in the modern era of spam filtering 13:39:20 (graue's spam filter blocked "span" and "div", that was hilarious…) 13:39:28 (…in that it lead to some interesting workarounds) 13:39:46 yeah we had span and div _templates_ instead 13:40:34 and it took some silly workarounds to make _those_ without triggering the filter in the process. 13:41:00 oerjan: a spambot somehow managed to make a page whose URL triggered the spam filter if it appeared in the Referer header 13:41:12 i remember that 13:41:31 given that deleting a page is a POST request, that took a bunch of confusion before I told my browser to turn referers off on Esolang 13:45:07 it seems there isn't that much spam nowadays, there have been no new filter catches since fizzie fixed the logging. well, unless it broke again i guess. 13:45:37 just a thought: fizzie's server is a different physical server from elliott's, right? 13:45:42 so presumably it has a different IP 13:45:47 perhaps the spambots don't use DNS 13:45:49 yes, it's Gregor's 13:46:06 the IPv4 space is easier to bruteforce than the domain name space… 13:46:31 ais523: um how can they post to a webserver without knowing the hostname? 13:46:45 i suppose some might allow it... 13:46:58 webservers often have a default hostname 13:47:02 esolangs.org did not. 13:47:06 or rather, solidity did not 13:47:08 right 13:47:16 And the new esolangs.org does not, either. 13:47:47 (Because the "default" server, the one it hits without a esolangs.org Host: header, is the www2.codu.org one, where there's just the hg browser and nothing else.) 13:49:15 * oerjan now imagines interlinking the wiki and HackEgo 13:50:21 Would you like HackEgo to post all recent changes on-channel?-) 13:50:27 (MediaWiki has some features for that.) 13:50:32 fancy 13:50:50 tempting 13:51:20 (I mean, they're not for IRC in particular, but it has some general mechanisms for getting change notifications out.) 13:51:59 fizzie: I think they were originally for IRC 13:52:09 in fact, IRC's the recommended official way to get a stream of all changes to Wikipedia 13:52:11 if you need one for some reason 13:53:27 fizzie: I tried to set that up once. 13:53:37 there are pre-built bridges from the UDP thing to IRC. 13:53:49 It sounds a bit spammy, but maybe it's not too bad 14:02:04 -!- Sorella has joined. 14:03:35 FireFly: There aren't typically very many changes/day, so there's at least that. 14:09:34 (On the other hand, it's kind of a silly idea, and we all know how serious and, for want of a better word, reserved, this channel is.) 14:10:33 hear hear 14:11:11 ^style europarl 14:11:12 Selected style: europarl (European Parliament speeches during approx. 1996-2006) 14:11:35 fungot: what is your opinion on the meat shortage? 14:11:36 oerjan: in fact, this would result in a conclusion that is perceived as a lame duck in the community institutions the council, to the framework of agenda 2000 but also to the rights associated with citizenship, which must be understood in every country, probably loses more from the commission was behind the policy choices now about what we are discussing this morning because these are very important. 14:12:10 -!- Patashu[Zzz] has quit (Ping timeout: 268 seconds). 14:12:32 well that certainly clears it up. 14:13:37 * oerjan forgot the "excuse me" 14:18:40 fungot: How come you didn't address the channel as "Mr. President"? 14:18:41 fizzie: mr president, ladies and gentlemen, the emperor is not wearing any clothes. these organisations are often under threat, at a revised directive for the duty of care for all those who have battled very fnord against the dual use of the most important details are dealt with objectively and impartially so that all of that here, hardly anybody, or rather on public access to documents which, due to the lack of backbone of the 14:21:36 fungot: what is your foreign policy on the Existence of Canada? 14:21:36 boily: madam president, i have tried to impose its legal point of view to match the objectives and procedures agreed upon by the commission on this issue. the commission can accept the first part of the amendment of directive 90/ 220/ eec, according to the united states 14:27:36 Sounds like you're getting annexed. 14:29:12 fungot: what are your thoughts concerning the situation in Crimea? 14:29:13 FireFly: madam president, commissioner, the copenhagen facility, money is given to improving transparency and a closer relationship with the applicant countries on this day that franz fnord fnord was murdered, together with a series of dictates issued from brussels as to what is being done, and i very much welcome an extension of the single market 14:30:34 wow, that almost seems relevant 14:31:46 I'm sorry fungot, I can't let you do that 14:31:47 FireFly: mr prodi, i ask the president-in-office of the council! we should note that the report only covers areas over which there is discussion everywhere the usa cannot achieve this itself. perhaps with that background it is totally wrong. as far as the framework is concerned. 14:33:18 I honestly hope "money is given to improving transparency" is a synthesis 14:33:28 (Regarding the es hypothesis, #selinux says in its /topic that the swedish linux channel is #linux.se .) 14:34:01 Also, franz ferdinord 14:34:20 Melvar: haha, that's also brilliant 14:34:36 as a joke, or does it come up often enough that they felt it was required? 14:36:35 I don’t know. 14:37:56 Does someone know a good way to pipe /list into grep? 14:38:55 you could hack together something involving netcat, nobuffer, and tee, I guess 14:39:08 not sure if the nobuffer is required, it often is, though 14:40:48 ~duck nobuffer 14:40:49 --- No relevant information 14:41:12 boily: it's something I wrote, but I'm sure loads of other people wrote it too 14:41:44 it basically just adds LD_PRELOAD hooks to get stdio to stop buffering, you inject it into programs that use default stdio buffering in cases where that's unacceptable for what you're doing 14:42:03 (and it does /that/ by hooking fopen and friends to call setvbuf and turn buffering off) 14:42:22 Does that do more that stdbuf? 14:42:34 ? 14:44:01 `stdbuf 14:44:01 stdbuf: missing operand \ Try `stdbuf --help' for more information. 14:44:16 `stdbuf --help 14:44:17 Usage: stdbuf OPTION... COMMAND \ Run COMMAND, with modified buffering operations for its standard streams. \ \ Mandatory arguments to long options are mandatory for short options too. \ -i, --input=MODE adjust standard input stream buffering \ -o, --output=MODE adjust standard output stream buffering \ -e, --error=MODE adjust standard 14:44:27 oh neat, it even has a standard name 14:44:37 and is installed here too 14:44:46 so I needn't have written nobuffer after all :-) 14:44:46 stdbuf is in coreutils. 14:45:01 I suspect it only sets fd buffering, though, not stdio. 14:45:18 -!- oerjan has set topic: Note that the Spanish Oteric channel is #oteric.es | PSA: fizzie is running the wiki now, contact him for any problems | https://dl.dropboxusercontent.com/u/2023808/wisdom.pdf http://codu.org/logs/_esoteric/ http://tunes.org/~nef/logs/esoteric/. 14:46:00 Jafet: no, it sets stdio buffering 14:46:03 fds don't /have/ buffering 14:46:28 interesting. “stdbuf --help”'s French translation is only halfway done, with random sentences in English. 14:47:07 "BUGS: On GLIBC platforms, specifying a buffer size, i.e. using fully buffered mode will result in undefined operation." 14:47:14 stdbuf's info page states that it doesn't work with tee, because tee changes buffering itself 14:47:19 So yes, stdbuf truly is just nobuffer. 14:47:21 so maybe you don't need a nobuffer with tee 14:53:15 Among other lesser-known coreutils is ptx, which can be used to make a search engine. 14:54:56 what the fungot is that mysterious utility... 14:54:56 boily: mr president, as we have always been a broad measure of agreement between the european union to take the results of the brussels 2 convention, but also with their quality. 14:55:03 Jafet: ptx confuses me a lot 14:55:14 Jafet: ptx puzzles me a lot. 14:55:15 mostly because I can't figure out what it's meant to do even after reading the info page 14:55:28 well, it creates a permuted index 14:55:29 but what is that? 14:56:02 -!- oerjan has quit (Quit: Like an index, but all over the place). 14:56:26 I figured out what it was by running it on a file. 14:58:04 you mean to say you ran a program without knowing what it did? 14:58:12 even if it comes from coreutils, that seems a little risky 14:58:46 The manual clearly says that it reads input files and produces some kind of output 14:59:07 (It creates a strangely-formatted index of the input.) 15:00:03 hmm, I wonder who maintains the coreutils Info file 15:00:14 it is decorative. I ran it on a DCSS morgue file. lots of zombies! 15:00:17 if it isn't Ulrich Drepper, it may be worth working out what the program does and then sending a patch 15:01:06 -!- ter2 has joined. 15:02:42 I believe (and hope) that Drepper does not maintain anything in coreutils. 15:03:28 Who is Ulrich Drepper and why does that name look vaguely familiar? 15:04:31 FireFly: from gnulibc, “with Ulrich Drepper[4] as the lead contributor and maintainer.” 15:04:41 s/nu// 15:05:22 Okay 15:05:33 that [4] looks ominous 15:06:55 of course it is ominous. it is the Number Four of Death! 15:08:25 drepper no longer maintains glibc 15:08:33 Hey, it occurs to me that people here might know: Are there any other -l puns other than libiberty and libowfat? 15:08:33 he was fired or whatever 15:08:51 Drepper is a very clever programmer and maintainer who often gets frustrated because many others are not as clever as he is. 15:09:08 I suspect -loobs is a reverse pun. 15:11:04 I suspect that there are, but can't think of any offhand 15:11:28 unless you count libraries that happen to make real words by coincidence, such as libick 15:11:44 $ locate libick.a | wc -l 15:11:45 35 15:11:51 wow I have a lot of compiled versions of INTERCAL 15:12:20 -!- tromp has joined. 15:12:39 some people have pron stashes on their machines. other collect intercal compilers... 15:13:13 one of them is in a backup of my account on a SunOS workstation 15:13:23 the account probably doesn't still exist, but who knows, it might 15:13:41 I don't have physical nor network access to the computer in question, so no way to tell 15:15:06 That's handy, I didn't know about timeout(1) 15:15:14 I imagine very few people my age have ever had the experience of accidentally writing programs that aren't portable to Unices other than SunOS 15:15:45 "Some platforms don't curently support timeouts beyond 2038" 15:15:56 Y2038K bugs are annoying 15:16:13 in another 23 years, maybe there'll be a huge panic and I'll be called in to help out with the cleanup effort 15:16:22 or maybe the world will end because people became complacent after Y2K 15:16:32 err, Y2K38 15:20:05 I thought the world ended in 2012 15:20:21 we need 2012 the game (a spin-off of 2048) 15:20:23 I thought that theory was conclusively disproved in 2013 15:20:47 ais523: Not conclusively 15:20:55 lifthrasiir: easy, it's identical to 2048, except your goal is to make 2012 (via integer overflow, obviously) 15:21:06 I mean... how do you know that 2012 doesn't exist anymore? 15:21:43 ais523, or modulo 2012, which would be a bit more interesting 15:22:00 `sdate date 15:22:00 ​/home/hackbot/hackbot.hg/multibot_cmds/lib/limits: line 5: exec: sdate: not found 15:22:13 oh come on, what sort of self-respecting server doesn't have sdate? 15:22:41 mroman: so you combine two 1024s to create a 36 15:22:48 then what? 15:23:47 I have no idea what 1024 has to do with 36 15:24:11 (1024 + 1024) % 2012 == 36, right? 15:24:38 anyway, the integer overflow joke is that you can't integer overflow in 2048 because there's not enough room on the board, and if you could, you'd get 0 15:24:39 I meant more the thing where time coexists 15:25:01 oh, I pinged the wrong person 15:25:02 lifthrasiir: ^ 15:25:48 ais523, yeah, 36. I'm actively destroying your joke :p 15:26:18 since it is modulo 2012, the eventual goal would be to make 2012 modulo 2012, aka 0. 15:26:20 but if it's modulo 2012, you can never get 2012 exactly, best you can get is 0 which is equivaent 15:26:30 there isn't enough room to make 0 module 2012 either, though 15:27:18 wheezy has: libace libarch libass libast libawl libax libemma libend libib libice libid libint libion liboath libobby libode liboop liboping liboss libots libow libowfat libs libust, though most of them are probably unintended. 15:29:18 is that in a default install? 15:29:37 checking the repos would be even better, as in more false positives 15:32:00 I have only libass it seems, satisfying that criteria 15:32:31 ais523: http://man.cat-v.org/unix-1st/1/sdate ? 15:32:50 nortti: no 15:33:06 `run ls /lib/lib*.so /usr/lib/lib*.so | sed 's:(/usr)?/lib/lib:l:; s:\.so$::' | sort | comm -12 - /usr/share/dict/words 15:33:07 ls: cannot access /lib/lib*.so: No such file or directory \ comm: /usr/share/dict/words: No such file or directory 15:33:09 also UNIX first edition predates september 1993 15:33:17 $ sdate date 15:33:17 Sun Sep 7523 16:33:03 BST 1993 15:33:24 oh 15:33:32 `ls /usr/share/dict 15:33:32 No output. 15:33:33 523! 15:33:44 -!- boily has quit (Quit: PERMUTATIONALISED CHICKEN). 15:33:46 [wiki] [[User:Fizzie]] M http://esolangs.org/w/index.php?diff=39225&oldid=39224 * Fizzie * (-59) I removed a test, but it keeps happening. 15:33:47 -!- metasepia has quit (Remote host closed the connection). 15:33:49 `ddate 15:33:49 Today is Sweetmorn, the 23rd day of Discord in the YOLD 3180 15:34:50 sdate is a great way to confuse random program 15:34:53 *programs 15:35:04 it's quite common for programs to assume that there can't be more than 31 days in a month 15:35:18 quite reasonable, too 15:36:27 how can you say that on September 7523? 15:36:50 oh 15:36:57 Now I see what sdate is doing 15:37:01 well, as long as one uses gregorian calendar 15:37:26 the sdate calendar is much simpler in many ways 15:37:27 Huh, I thought the above didn't work at all, because I was all scrolled up in the backscroll. 15:40:30 -!- zzo38 has joined. 15:46:07 for l in $(apt-cache search --names-only 'lib.*' | grep -Eo '^lib[aeiou][a-zA-Z]+' | uniq | sed 's/^lib/l/'); do for i in `seq $#l -1 3`; do if ( dict -s exact -d wn -m ${l[1,$i]} >/dev/null 2>&1 ); then echo $l; break; fi; done; done 15:46:43 Most of the matches are the minimum, 3 letters long. 15:50:11 -!- ^v has joined. 15:51:58 ais523: Quite a lot of our university (including a classroom full of UltraSPARC 10 workstations, and both main general-use shell servers) ran on SunOS not more than a decade ago, so I think a lot of our CS students have that experience. 15:54:38 Heh. My main experience with Solaris is that killall will in fact, kill all. 15:55:31 Incidentally, if I make "test" edits of my user page (e.g. after upgrade, to test that everything still works), is it considered good MediaWiki administrative practice to somehow undo those (if that's even possible) so that they don't clutter up RC or page history? 15:56:35 ais523, sdate does some sort of LD_PRELOAD? 15:56:49 Vorpal: pretty much 15:57:42 Hm I suspect it won't break cfunge, I only think I use unix timestamp in that for the DATE fingerprint 15:58:29 Oh, DATE doesn't even have a command to get current date? 15:58:34 only for conversion 15:58:35 right 15:59:17 Oh okay, TIME uses localtime/gmtime, so that would be confused 15:59:22 I assume 16:15:14 Gah, the offline experience of Google Maps Android has gone so downhill. 16:15:16 It used to be so that you had a separate "Offline" tab in "My Places", where you could see a list of what's available offline, and remove/add regions, and when adding you could drag the bounding box what you want to cache. 16:15:20 But in this 7.x version, there seems to be no way at all of seeing what is available offline, and adding new regions is done by the most illogical way I have ever heard of: you zoom the map so that what you want to cache offline is visible, and then search for "ok maps". 16:15:24 Then it flashes briefly something to the tune of "caching map data", but afterwards there seems to be no indication of what exactly it saved. 16:16:01 Incidentally, if I make "test" edits of my user page (e.g. after upgrade, to test that everything still works), is it considered good MediaWiki administrative practice to somehow undo those (if that's even possible) so that they don't clutter up RC or page history? 16:16:03 not really 16:16:12 fizzie, downgrade if you can, that is what I did 16:16:26 I mean, just compare: https://support.google.com/gmm/answer/2650377?hl=en&ref_topic=2649131 https://support.google.com/gmm/answer/3273567?hl=en&ref_topic=3273087 16:16:38 fizzie: (re the comment Phantom_Hoover quoted) no, although sometimes it's considered good form to use a page designed for test edits 16:16:50 those pages can later have the test edits removed in batch 16:17:07 actually, that reminds me of the old AfD page on Wikipedia 16:17:15 or maybe VfD 16:17:55 before it was split into subpages, it had been edited so many times that when an annoyed admin deleted it, it brought the site down because the server couldn't handle deleting that many edits 16:18:13 then when it was subsequently undeleted (after the site came back up), it brought the site down again 16:18:40 ais523, heh 16:18:44 then the developers added a limit that you can't delete pages that have been edited 5000 or more times, so some enterprising admin decided to get a bot to edit the Main Page 5000 times in order to stop it being deleted 16:19:00 the developers stopped that at like 3000 because it was creating a bunch of load in of itself 16:19:03 ais523, also how could it bring the server down if the delete isn't permanent? Isn't it just marking a flag then 16:19:37 ais523, would the main page be in risk of being deleted though? 16:19:38 it's not exactly a flag internally, the edit gets moved to a separate table, and a whole bunch of tables that relate to non-deleted edits get change 16:19:40 *changed 16:19:50 fizzie: notifying people the wiki upgraded is prtetty useful, anyway 16:19:53 and yeah, the main page used to get deleted quite frequently 16:20:13 ais523, why? Shouldn't those admins that did it get fired then? 16:20:14 it was the #1 most common action for disgruntled admins (which is strange, because it doesn't cause a whole lot of damage) 16:20:26 haha, fired 16:20:30 as if MW admins get paid 16:20:31 It’s rather visible though. 16:20:38 and yeah, obviously they were desysopped, unless they could successfully pass it off as a joke and also a lapse of judgement at the same time 16:20:42 or an accident 16:20:47 elliott, well the equivalent. Stripped of their rank, whatever 16:20:53 I think it has been deleted by accident before now 16:21:01 anyway, there are a whole bunch of protections in place nowadays 16:21:39 I think I'm one of the few people who knows how to delete the Main Page through them, and even then, a) I no longer have the perms; b) with the protections in place, doing so would be rather time-consuming, and there'd be a pretty high chance someone would notice what I was doing and stop me 16:22:24 using a bot to help out would make it a bit faster, although testing it would be awkward; if you're going to break the rules to the extent of running an unapproved bot 16:22:32 err, to the extent of deleting the main page 16:22:42 you're probably not going to be worried about running an unauthorized bot 16:23:00 testing it sounds easy, just run a local test wiki 16:23:52 before it was split into subpages, it had been edited so many times that when an annoyed admin deleted it, it brought the site down because the server couldn't handle deleting that many edits 16:24:02 wasn't 'the annoyed admin' ed poor? 16:24:16 According to http://en.wikipedia.org/wiki/Wikipedia:Don't_delete_the_main_page it's protected in the mediawiki software nowadays 16:24:21 not sure, I don't typically remember the names of specific disgruntled admins 16:24:23 FireFly: yes, I know 16:25:02 this is why very few people would be able to delete it, you'd need a knowledge of how the software works in addition to a knowledge of the redundant protections on enwiki in particular 16:25:14 and for all I know, they may have added more since I stopped actively adminning Wikpedia 16:27:32 * Melvar is amused that Wikipedia:Don't delete the main page has the shortcut WP:IDIOT . 16:28:26 those shortcuts are incredibly cheap to create 16:28:37 you could make one right now if you felt like it, and given that they're rather more expensive to delete 16:28:46 if it's remotely plausible people would probably just let it stay 16:29:25 and there are lots of them on en.wikipedia: the very long page https://en.wikipedia.org/wiki/User:B_jonas/Pages_with_shortcuts lists many of them 16:29:55 note that it only lists ones where the target page has a template marking the shortcut 16:29:56 Wikipedia:Village stocks apparently has WP:AAAARGH 16:30:10 and there are lots more where the target page isn't marked 16:30:52 there are shortcuts like P:P -- guess where that goes 16:31:15 or P:N 16:31:28 P: is Portal:, I guess 16:31:30 P:M wasn't Microsoft... 16:31:32 then there's P:Z 16:31:37 but there are numerous portals that could start with P 16:34:26 probably most of them actually do start with P 16:35:16 /most/? 16:35:17 “WP:Oh I say, what are you doing? Come down from there at once! Really, you're making a frightful exhibition of yourself.” 16:36:46 The target's title doesn't make *much* more sense 16:37:32 Melvar: I remember that page being created 16:37:39 I forget the context, but ISTR it was necessary 16:42:55 Does that redirect to the reichstag page 16:44:39 yes 16:54:16 I forget the context, but ISTR it was necessary <-- what? Why? 16:54:25 as I said, I forget 16:54:38 My mind just boggles at the idea such a page could be *required* 16:54:50 like, people were taking something too seriously, and it was causing problems 16:54:58 Okay 16:55:00 Hm 16:55:49 -!- Frooxius has quit (Read error: Connection reset by peer). 16:56:26 I think there were also people who didn't understand WP:POINT 17:05:41 ooh https://en.wikipedia.org/wiki/P:X 17:09:22 P:Q seems to be free 17:14:46 Aw, that’s just an x. 17:24:14 `coins 17:24:15 q-revercoin holdecoin quiseporacoin glypecoin usaumcoin rfollocoin baggedcoin temparncoin l3841coin zowcoin braincoin :coin redgrecoin mempaicoin raintfcoin petrocoin sellcoin thumorecoin toicoin condicoin 17:25:14 `coins 17:25:15 rhotorcoin nconveccoin gracoin smycelcoin kayakcoin bythessenhohcoin dobbocoin brbcoin devercoin rnarycoin memmentcoin sedcoin sorccoin overspidcoin bhacoin convecorritcoin migoritaffcoin objectdrumcoin tbfcoin huntedlyearandeltcoin 17:25:24 `cat bin/coins 17:25:24 words ${1---eng-1M --esolangs 20} | sed -re 's/( |$)/coin\1/g' 17:25:30 right 17:28:51 -!- copumpkin has joined. 17:41:52 Huh, someone was just querying for "DrIvErcAzo.esOLaNGs.oRg" (with that capitalization) when I was debugging a (non-esolangs) DNS thing and looking at tcpdump. 17:42:07 (Spoiler: there is no such host.) 17:42:36 I /said/ that brute-forcing the space of all domain names would be harder than brute-forcing IPs :-) 17:43:27 Huh, and now crawl-66-249-66-216.googlebot.com asked for "drivercazo.esolangs.org", in lowercase. Was that in fact something at some point? 17:43:57 Four Google hits for the name. 17:44:17 perhaps someone linked to it 17:44:26 with that random caps 17:44:32 and googlebot converts to lowercase 17:46:35 Two of the hits did have curious-looking links; for /stranica/drayver-dlya-netronix-mn01.html and /catalog/drayver-dlya-i7500.html. 17:56:38 Oh: https://isc.sans.edu/diary/Use+of+Mixed+Case+DNS+Queries/12418 17:57:16 (There were more mixed-case queries, so I got curious.) 17:58:11 Crafty trick. 18:01:29 fizzie: IME the logs are always filled with the weirdest things 18:01:56 fizzie: https://twitter.com/zucrowgo/status/184029492237438976 18:02:00 I haven't even looked at the web server log. 18:34:27 douglass_: "We have two mugs of tea; one has pictures of geese and one has the MongoDB logo." shachaf: "Oh no, which is the lesser of two evils?" 18:44:29 Geese are evil? 18:54:28 fizzie, interesting, that mixed case thing, but I wonder what happens when some DNS server doesn't maintain the case. Surely stuff will break then, and if you allow for that you basically nullified security. 18:54:36 Geese are vicious! 18:54:44 They bite. 18:55:41 fizzie, You can't even reliably keep track of when it worked in the past of course, since the server software could change, or there could be some BGP anycast going on making it resolve to different servers at different points in time 18:55:48 It probably works most of the time, but still... 18:56:36 The page you linked even said "almost all" hm 18:57:59 The RFC says "all". 18:58:07 Though of course that's just an empirical observation. 18:58:21 I can only find a draft RFC for it 18:58:33 The draft is what I meant. 18:58:33 From 2008 18:58:48 That is quite a long time for it just to sit around as a draft isn't ut? 18:58:51 it* 18:59:53 Long enough that it has expired, yes. 19:00:00 Right 19:00:19 -!- tromp has quit (Remote host closed the connection). 19:00:31 (The validity period is six months.) 19:00:34 Well I use unbound locally, I could probably turn it on 19:01:50 The expired draft does mention "a small set of rare and/or private label authoritative DNS implementations" that do lowercase names. 19:02:24 Right 19:03:32 geese don't have moral agency so I guess they can't be evil 19:03:37 but they are vicious, though 19:03:39 swans are worse though 19:03:46 swans will break your arm 19:03:48 kmc, are you sure they don't? 19:03:55 they also fight to the death over territory 19:04:09 which upsets people who want swans to be peaceful symbols of love 19:04:19 I think pigeons are a much better peaceful symbol of love 19:04:27 I should probably finally get DNSSEC on for zem.fi one of these days; it's just that IIRC, the free secondary-DNS service I've been using doesn't grok the records. 19:04:30 (Kind of shabby that none of them do TSIG for zone transfers, either.) 19:04:37 one time a flock of geese chased douglass_ and me down a hill and almost into the Charles River Basin 19:05:21 there was that time when I was in the middle of a flock of geese, the ones behind me were trying to force me forwards in order to disturb the ones ahead of me 19:05:30 none of them actually attacked but it was pretty scary 19:05:42 * kmc -> lunch 19:05:46 kmc, pigeons? Peace? Really? When I hear pigeon I think of all the issues they cause in cities 19:05:59 fizzie: did I ever give you alan dipert's email? in case there's ever any domain stuff with esolangs.org 19:06:04 I guess that is a symbol of love, given how many there are 19:06:18 elliott: I don't think I have it, no. 19:06:18 elliott: you missed out the article :-( 19:06:19 Vorpal: when people like them they usually call them "doves" instead of "pigeons", but the terms are basically interchangeable 19:06:38 fizzie: okay, alan@dipert.org then 19:06:50 Now that you've said it, it actually sounds familiar. 19:06:55 kmc, yes, and I see right through that nomenclature and still dislike them :P 19:07:07 although googling suggests adipert@acm.org?? 19:07:19 "the alan dipert" was the usual terminology, IIRC 19:07:33 but alan@dipert.org is the one I received emails from and sent to. 19:07:50 maybe they're different people? 19:08:05 fizzie, speaking of free secondary DNS, what do you use for that? 19:09:35 ais523: they are not. 19:09:44 -!- ^v has quit (Ping timeout: 252 seconds). 19:10:14 Also, IIRC pigeons supposedly don’t taste too bad if they haven’t been eating trash. 19:10:17 fizzie, been looking for one myself 19:10:43 -!- MindlessDrone has quit (Quit: MindlessDrone). 19:11:21 Vorpal: I use twisted4life.com at the moment, but it's kind of clunky. And their server is in Malaysia somewhere, which I guess is good for redundancy, but it's quite slow from many places. Earlier, I've used afraid.org and xname.org. And there's buddyns.com and dns.he.net I've heard of. 19:11:55 fizzie, hm what about more than 2 DNS server? Couldn't you use several. 19:12:13 Sure, but it's even more of a hassle. 19:12:18 True 19:12:31 Some of those come with more than one server, anyway. 19:12:34 I would assume the he.net one is quite reputable 19:13:12 I've been thinking about it, since I already have their IPv6 pipe and all, but they don't have DNSSEC support yet either. (Not sure if that just means for their primary DNS side, though.) 19:14:44 None of those you mentioned seems to support DNSSEC, at least they are not listing it 19:15:22 I seem to recall afraid.org had something. 19:15:47 Their FAQ doesn't say anything about it, though. 19:19:07 -!- ais523 has quit. 19:20:55 (And there's https://acc.rollernet.us/help/dns/secondary.php that I know of that's explicitly DNSSEC-compatible and free.) 19:21:21 -!- nisstyre has joined. 19:23:35 https://forums.he.net/index.php?topic=3139.0 "still some hurdles" 19:24:54 vorpal: I use https://puck.nether.net/dns/ and http://freedns.afraid.org/secondary/ 19:25:07 vorpal: http://www.frankb.us/dns/ 19:25:24 The ones i’m using support DNSSEC. 19:29:32 -!- chaiomanot has joined. 19:39:48 -!- ^v has joined. 19:42:00 ion, hm nice 19:43:09 Huh I just got SMS spam, that was new 19:44:59 I guess CM doesn't have a block list, oh well 19:49:02 fizzie, based on that link ion posted, I guess afraid is good 19:49:09 Doesn't do ipv6 apparently 19:56:27 fizzie, elliott, ion: What is a good VPS provider? I want to do my own domain, web server and mail server basically. 19:56:41 Reasonably priced 19:59:13 I’m afraid i don’t know, i haven’t been looking for a VPS. 20:02:26 I'm running zem.fi on Tilaa, and I think it's good. 20:02:57 (That really-cheap-dedicated-server link someone posted was rather intriguing, too.) 20:03:17 (That's "Tilaa" as in tilaa.nl.) 20:03:52 That redirects to tilaa.com 20:04:14 Also, I wouldn't want to deal with the crap that a dedicated server means 20:04:36 fizzie, Tilaa uses xen I guess? 20:06:25 Oh, I guess .com is their primary domain, then. 20:06:29 And yes, it's a Xen shop. 20:06:31 Hm 20:06:44 I have no idea what a reasonable price is here though 20:07:35 fizzie, I guess they allow mail? And I presume irc too? 20:07:45 (As in bouncer, not server) 20:07:52 You can compare prices with some "mainstream" providers, like Linode or a reserved Amazon EC2 instance. 20:07:57 Hm 20:09:01 And I don't think Tilaa has any restrictions except some generic "must follow local (Dutch) law" clause. 20:09:06 And something about the bandwidth. 20:09:51 It says network traffic is unlimted/free on the configurator at least 20:10:08 Presumably there is some clause about null routing in case of DDoS or similar though 20:10:42 "-- not allowed to run services that continuously use full capacity, like public software mirrors, BitTorrent seeding or web crawling/indexing --" in the "knowledge base". 20:10:47 Rather vaguely worded, but oh well. 20:10:56 Well okay, that seems kind of reasonable 20:11:30 There are other places where you pay for anything above some fixed number, which is arguably at least more consistent. 20:11:39 what does a "reserved" EC2 mean btw? 20:12:40 Also EC2 pricing is convoluted 20:12:59 They have "on-demand" instances where you pay per hour whenever the server is online, and "reserved" instances where you make a single payment, and then get a discounted hourly rate. 20:13:22 (I thought reserved instances had some monthly price, but apparently it's an upfront cost nowadays. May have changed, of course.) 20:13:22 Also I seem to remember reading about EC2 blocking outgoing SMTP 20:14:28 Yeah I can't be bothered to dig through that pricing model 20:14:47 Yes, it's kind of convoluted. Separate prices for data transfer and storage and all that. 20:14:57 At least for Linode you should be able to find a single number. 20:15:13 Yes, and it seems generally a bit more expensive than Tilaa 20:16:24 I used prgmr before Tilaa (like I guess I've mentioned), and I don't have anything really to complain there either, except for some lag for interactive use (but I doubt they can do much about the speed of light). It's a bit less "commercial-looking" than Tilaa. 20:18:12 fizzie, so far Tilaa seems to offer the most bang for the buck 20:18:52 Also why does prgmr list Debian 6.0.3 as an option they will install, but not Debian 7? Debian 6 is *oldstable* 20:19:15 Not impressed by that in the least 20:22:16 Oh, EC2 is only 99.95% uptime? I expected more from them 20:24:52 There are quite a few extremely cheap places selling OpenVZ and the like, if you want to go that way. 20:25:00 Hell no 20:25:10 ooh, that is sneaky, Tilaa excludes "external factors" "such as" "DDoS attacks" 20:25:22 in the uptime guarantee 20:25:38 fizzie, anyway you said Tilaa used xen right? 20:25:53 Anyway why would places using openvz be cheaper 20:26:27 I guess because you don't need to dedicate RAM for your guests then. 20:26:42 Easier to oversell. 20:28:03 Ah 20:29:12 fizzie, yeah Tilaa's uptime guarnatee is kind of shitty, seems good otherwise though 20:30:54 -!- copumpkin has quit (Quit: Textual IRC Client: www.textualapp.com). 20:32:06 fizzie, which distro did you go for on Tilaa btw? 20:32:20 Just out of interest 20:32:26 Debian. 20:32:29 Ah 20:33:26 And yes, their SLA is kind of funny; I guess it only includes their own mistakes and nothing else, since their own planned maintenance as well as acts of unspecified "third parties" are all excluded. 20:33:38 -!- copumpkin has joined. 20:34:33 fungot: What do you think of Tilaa? I know you're not exactly running there, but you've been speaking via them for no particular reason except convenience and vanity. 20:34:34 fizzie: madam president, mr mauro used an example. finally, mr marset campos: we hope that the process in november with the formal presentation of the euro and in view of the fact that the president of the commission, and we will also have to work with you to intervene to stop uncontrolled and illegal immigration on public opinion and the draft joint action as amended, i would ask for the commission and in the draft resolution 20:37:01 fizzie, why not run it on tilaa? 20:37:05 Model files too large? 20:39:14 Jafet: It's a bit late, but "money is given to improving transparency" is indeed a synthesis. (From "In connection with global loans, that is, the Copenhagen facility, money is given to banks which then issue the loans." and "-- priority should be given to improving transparency and a clear demarcation of powers between the European Union and the Member States.") 20:40:00 Vorpal: Well, they're slightly large. But there's no really strong reason why not, except laziness. 20:40:19 fizzie, how large are they, combined? 20:40:25 The chroot jail for fungot is 646M. 20:40:25 fizzie: mr president, the report proposes to introduce competition in air transport services entails risks which have been declared adopted and i do it willingly. madam president, i am fully behind the support which the european parliament. 20:40:42 Oh okay, that is just program, interpreter and model? 20:41:06 I guess the interpreter + libraries would account for a couple of megabytes at most 20:41:11 Actually, it's not even that, since I kludged in cfunge the "chroot at startup" feature. 20:41:30 It's only the Funge-98 sources and data. 20:42:10 fizzie, hey, you should send a patch upstream ;P 20:42:29 And apparently some DNS resolver support files, though I don't think those are used for anything at the moment. 20:42:30 Personally I would be using an umlbox like what HackEgo uses 20:42:44 An umlbox might be nice too. 20:42:55 I guess that would be out of question on a VPS though? 20:43:09 Or I guess since it is xen, you could do it... 20:43:10 HackEgo runs on a VPS, so I don't see why not. 20:43:17 Hm okay 20:43:30 I mean, it's just a process almost just like any other. 20:44:06 I thought it used special stuff from the kernel 20:44:08 But maybe not 20:44:14 -!- maurer_ has joined. 20:44:39 There was some sort of a performance thing. 20:44:42 I forget the details. 20:44:52 Xen is "just" like having your own machine. 20:44:55 except for the ways in which it's not. 20:45:08 Such as you couldn't do hw virutalization inside it yeah 20:45:33 UML had the "Skas mode" that needed a patch, but I can't remember if that got replaced by something newer. 20:46:36 The cfunge process has a RSS of 6988k and a VSZ of 27592k, so I guess it would fit okayish even on my so-small-they-no-longer-sell-it 256M RAM Tilaa instance. 20:47:35 fizzie, oh, what does it cost you? 20:47:52 Also, I forget what VSZ is 20:48:02 Address space cost, I think. 20:48:04 And 4.77 EUR/month, IIRC. 20:48:18 -!- copumpkin has quit (*.net *.split). 20:48:19 -!- maurer has quit (*.net *.split). 20:48:20 -!- jconn has quit (*.net *.split). 20:48:22 -!- Phantom_Hoover has quit (*.net *.split). 20:48:23 -!- itsy has quit (*.net *.split). 20:48:24 -!- lambdabot has quit (*.net *.split). 20:48:31 (That's including VAT; the prices quoted on their page were, at least before, excluding it.) 20:48:34 fizzie, Oh, so cheaper than any available option, 20:48:44 -!- copumpkin has joined. 20:49:05 Yes. But it's also "only" 6 gigs of disk, too. 20:49:22 I made up a implementation of a "CREATE FUNCTION" command on SQL. 20:49:32 fizzie, ah 20:49:43 Er, make that 10 gigs, apparently. Huh. I was sure it was six. 20:49:47 I don't know, if that server supports this SQL extensions or not. 20:51:54 `run uname -a 20:51:54 Linux umlbox 3.13.0-umlbox #1 Wed Jan 29 12:56:45 UTC 2014 x86_64 GNU/Linux 20:52:05 That's quite new. 20:54:09 If you are not going to store any movies and/or pornography, then probably 6 GB is sufficient. 20:54:30 -!- lambdabot has joined. 20:54:47 It's a bit few also for things like photos. 20:55:32 Our collection of vacation photos shared with family and friends seems to be 21G at the moment. 20:55:41 -!- Koen_ has joined. 20:56:39 -!- Phantom_Hoover has joined. 20:59:55 Bah, UML documentation is so fragmentary and (in many places) old. The HOWTO in Documentation/ of the kernel sources is from 2002, the sf.net page at http://user-mode-linux.sourceforge.net/ has just links to "The UML Wiki" with an "It works!" default page in it and "The Old UML Site", and so on. 21:01:14 http://blogs.igalia.com/berto/2006/09/13/user-mode-linux-and-skas0/ well, that's not new either, but I think that's the "skas mode with no patched kernel on the host" thing I thought existed. 21:01:40 Because I don't think there are skas3 patches for anything newer than 2.6.x. 21:03:24 Ah 21:05:04 -!- iano has joined. 21:05:26 -!- iano has left. 21:24:45 Vorpal: Incidentally, I added AVR in that quiz. 21:27:55 fizzie, huh, which variant of AVR is that 21:28:01 I coded for the ATMega128 iirc 21:30:02 fizzie, also it likes giving me the fake opcode "BR" a /lot/ 21:30:06 I got it 3 times in a row 21:30:36 Guess the instruction set is too small? 21:30:46 It's just "grep AVR_INSN avr.h | cut -c 11- | sed -e 's/,.*//' | sort | uniq". And it is quite small, yes. 21:31:36 Well I got 89% at best 21:31:46 More commonly around 85% 21:32:09 http://sprunge.us/aZaF it does seem quite fond of "br". 21:32:26 (That's a "uniq -c" on 100 generated opcodes.) 21:32:52 Possibly a shorter model order might've been better for this. 21:33:58 -!- oerjan has joined. 22:08:37 -!- SpaceWizard has joined. 22:19:43 -!- nisstyre has quit (Quit: WeeChat 0.4.3). 22:26:14 -!- conehead has joined. 22:39:35 -!- Frooxius has joined. 22:42:42 gtristan: NULL values in columns take up one byte of disk space each, I think. 22:42:50 -!- zzo38 has left. 22:44:39 Vorpal: counterpoint: pigeons are awesome, screw you 22:44:50 some pigeons have made a nest behind a plastic fake owl at my house <3 22:45:00 the kind meant to keep birds away 22:46:09 Pigeons are not my favourite bird 22:46:28 what is your favourite bird 22:46:43 They don't seem to have been made to the same scale as any of the other birds that appear in my garden 22:46:53 kmc, I dunno, the blackbird, maybe? 22:47:04 [wiki] [[Talk:Universal Machine]] N http://esolangs.org/w/index.php?oldid=39226 * EzoLang * (+102) Possible duplicate 22:47:04 Or a coal tit? 22:47:13 Hold on that's new 22:47:38 When did HackEgo start doing that 22:47:57 this morning, maybe 22:48:13 yeah. 22:48:22 The song of the Australian Magpie is other-worldly 22:50:10 b_jonas: EzoLang seems to have a point 22:50:58 Taneb: how about seagulls 22:51:01 seagulls are good 22:51:07 for example i'm the best 22:51:08 -!- nooodl has quit (Ping timeout: 240 seconds). 22:51:24 Well, the seagulls around here feel out of place 22:51:27 * oerjan likes seagulls. better than crows, anyhow. 22:51:32 I'm 25 miles from the coast 22:51:37 oerjan: i heard that seagulls have a bad reputation in norway 22:53:06 shachaf: well people don't like to have them too close, all the time. 22:53:15 They have a bad reputation here 22:53:18 Pigeons too 22:53:20 they're noisy and vindictive during breeding season 22:53:48 And the geese at my uni are notorious 22:53:57 -!- nooodl has joined. 22:55:49 not much geese in trondheim that i know of, we see them flying in their V formation during migration 22:56:44 however somehow the noise of seagulls doesn't grate on me the same way crows do. 22:57:30 humans also tend to be noisy and vindictive during breeding season, which is all of the time. 23:09:36 i ended up getting chased by one of the geese at warwick 23:09:51 this was admittedly after i'd followed it around for a bit until it walked into a corner 23:10:30 never corner a goose 23:10:31 rookie mistake 23:11:10 Phantom_Hoover, how does it feel going to an inferior university in the all-important metric of duck to student ratio? 23:11:10 shortly after that i realised i was no longer following the goose; the goose was following me 23:11:23 Taneb, terrible 23:11:47 also there was a guy who got kicked out for, amongst other things, chasing a swan about the place last year 23:12:10 Someone tried to sell a swan to someone I know 23:12:33 I'm... not sure how they got it that far from the lake 23:16:22 copumpkin: the play was good 23:16:33 oh I'm glad! 23:16:42 I missed the second day of hac nyc anyway 23:16:49 but that too was pretty good 23:17:21 I need to get me to some of these 23:17:35 Wouldn't be too hard to do hac yc 23:17:57 Taneb: just go to zurihac 23:18:03 and/or bayhac 23:18:08 shachaf, I really want to but hoooteeel roooooms 23:18:18 Can I like stay with you in your Zurich apartment 23:18:35 i don't have a zurich apartment 23:18:52 i can't predict my zurischedule 23:19:11 -!- zzo38 has joined. 23:22:37 http://en.wikipedia.org/wiki/Guinea_(British_coin) 23:22:43 i like how batshit insane this is 23:22:52 With one of those I could afford a Zurich apartment! 23:23:01 'let's have a coin worth £1.05!' 23:30:03 Taneb: are you moving to zurich 23:30:26 Not in the near future 23:30:40 I still have the majority of an undergraduate degree to complete 23:31:07 But after that? Who knows! 23:31:11 I'll go where the wind takes me 23:32:06 i'd just remind you the earth is something like 4/5 ocean hth 23:32:44 I know people with boats, it's cushty 23:32:51 I'm 25 miles from the coast 23:33:06 i imagine that Taneb is p. wind-resistant 23:33:12 so it's probably ok 23:33:15 hm 70.8% says this link 23:37:16 i live p. close to the uk's pole of inaccessibility 23:37:18 it's not great 23:38:44 In holiday-time I live near England's pole of inaccessibility from roads 23:39:22 -!- bigbugkik has joined. 23:42:25 Hang on 23:42:29 Did... 23:42:35 Did I just use the word "cushty" 23:44:29 -!- Frooxius has quit (Read error: Connection reset by peer). 23:52:48 -!- Phantom_Hoover has quit (Remote host closed the connection). 23:53:15 -!- nooodl has quit (Ping timeout: 255 seconds). 23:56:09 -!- yorick has quit (Remote host closed the connection). 23:58:21 -!- nooodl has joined. 23:59:38 -!- oerjan has quit (Quit: new cushtoms).