00:03:56 -!- wyldstallyns has joined. 00:27:54 -!- Patashu has joined. 00:30:56 -!- Patashu has quit (Remote host closed the connection). 00:31:17 -!- Patashu has joined. 00:33:36 -!- Patashu has quit (Client Quit). 00:34:49 -!- Patashu has joined. 00:42:31 -!- augur has quit (Remote host closed the connection). 00:42:40 Hello; do you want................. 01:19:52 In the Dungeons&Dragons game I will make it my character's job to write down everything about the haunted castle and so on, in a book to keep in the royal archives and in the library, so that next time anyone need to go in there, they can know what to expect 01:22:40 -!- pikhq_ has joined. 01:23:19 -!- pikhq has quit (Ping timeout: 276 seconds). 01:47:09 -!- itidus20 has joined. 01:50:13 -!- itidus21 has quit (Ping timeout: 248 seconds). 01:52:39 messages? 01:52:39 itidus20: You have 1 new message. '/msg lambdabot @messages' to read it. 01:52:49 @messages 01:52:49 itidus21 said 15h 54m 6s ago: elliott still has a box of beasts to use on zot:5 02:18:20 -!- Tod-Autojoined has changed nick to TodPunk. 02:28:23 -!- azaq23 has joined. 03:21:37 -!- calamari has joined. 04:20:40 oh my god someone actually tried to play the death waltz 04:25:15 ? 04:40:05 -!- calamari has quit (Quit: Leaving). 04:46:23 `log waltz 04:46:56 No output. 04:46:59 `log waltz 04:47:16 2011-08-26.txt:23:47:46: [20:52] Johnson Plays Waltz Citing Beethoven's "Your From Round Pittsburg?" Use Escalating Overalls You Damn Jackasses... Just kidding!" 04:47:55 `log waltz 04:48:02 2010-07-11.txt:07:19:52: I missed the Death Waltz bit where the upper staff gets a note below the bottom staff on ledger lines :D 04:48:23 `log waltz 04:48:30 2012-05-03.txt:04:47:55: `log waltz 04:48:41 `pastelog waltz 04:48:48 http://codu.org/projects/hackbot/fshg/index.cgi/raw-file/tip/paste/paste.24809 04:48:58 Oh, right 04:49:14 monqy, there was an update at ??:10 04:49:29 Or so 04:57:25 http://www.youtube.com/watch?v=sCgT94A7WgI 05:00:06 coppro, embedding disabled on request when I try to use youtuberepeat 05:00:13 And it crashes when I try to view it normally 05:00:18 So, FCUK 05:00:22 noted 05:00:26 * Sgeo opens Firfox 05:00:28 Firefox 05:00:33 -!- ais523 has joined. 05:01:03 Oh, I've seen this 05:03:27 coppro, did they skip bar 7 correctly? 05:03:38 hard ot tell 05:03:46 unsurprisingly, it's difficult to follow along to the music 05:04:18 although I think they forgot to repeat the first "bar" 9 times 05:06:21 I don't see that 05:06:50 -!- ais523 has quit (Ping timeout: 272 seconds). 05:13:47 -!- oklopol has joined. 05:42:44 -!- augur has joined. 05:47:42 kmc: Did you see monqy's self-portrait of me? 05:48:22 yes 05:48:28 i think you asked me that yesterday 05:48:33 is this the one with bonzi buddy? 05:49:36 elliott also drew a self-portrait of me with Bonzi Buddy. 05:49:41 But monqy drew a different one. 05:49:51 11:38 < elliott> 19:18 08:05:37: Patashu: http://dl.dropbox.com/u/13786158/shachaf.png http://dl.dropbox.com/u/13786158/eliot.png 05:51:53 It's pretty great. 05:52:00 thonqy 05:52:38 nice 05:53:01 how do you make a self-portrait of someone else 05:53:11 wyldstallyns: Just like monqy did it. 06:07:47 -!- variable has quit (Read error: Operation timed out). 06:15:33 -!- ais523 has joined. 06:18:40 they're very good 06:24:26 @tell elliott I just found out about http://www.zincland.com/7drl/kobold/ from TVTropes. Have you tried it? 06:24:27 Consider it noted. 06:25:14 has anyone not tried smart kobold? 06:25:28 hlep 06:25:34 "whats smart kobold" 06:25:36 monqy, I haven't heard of it until now 06:25:45 how did you live with yourself 06:25:49 bah, next ThinkPad X series will have a chiclet keyboard 06:25:53 http://www.ubergizmo.com/2012/05/lenovo-thinkpad-with-ivy-bridge-leaked/ 06:26:01 -!- zzo38 has quit (Remote host closed the connection). 06:26:55 chiclet? 06:27:37 CHICLET KEYBOARD? MORE LIKE CHIC KEYBOARD, KORRECT? 06:27:48 https://en.wikipedia.org/wiki/Chiclet_keyboard 06:28:04 -!- variable has joined. 06:29:38 -!- ais523 has quit (Ping timeout: 245 seconds). 06:31:33 monqy has been UPDATEd 06:33:02 Sgeo: it used to mean a particularly shitty kind of rubber membrane keypad seen on tv remotes etc 06:33:30 now it's used to refer to laptop keyboards where individual keys are separated by empty space 06:33:38 even though the key mechanism is (in general) much better than those shitty membranes 06:33:48 the thinkpad chiclet keyboard allegedly has the same mechanism as the classic thinkpad keyboard 06:34:09 I'm using a Dell chiclet keyboard. 06:35:13 -!- ais523 has joined. 06:36:03 ooh, I have made a discovery! 06:36:22 the wireless around here is really buggy (apparently it's a router issue), and normally won't connect at all, or only for short periods of time 06:36:33 but it's more likely to connect immediately after the computer is rebooted 06:36:57 and it seems that I can replicate that same high success chance by a rmmod/insmod on the wi-fi driver kernel module 06:37:05 this could seriously save me a lot of time if it works 06:37:21 ais523: I once considered making an alias for "sudo sh -c 'rmmod iwlwifi; modprobe iwlwifi'" 06:37:27 But it never leaves my command line history. 06:37:32 So I didn't get around to doing it. 06:38:05 yeah i have a script for that 06:38:19 on linux you should look into whether your card tries to support 802.11n 06:38:21 and if so, disable it 06:38:28 that improved reliability a lot on one machine i tried 06:39:15 My card keeps developing exciting new bugs. 06:39:43 Under Ubuntu it was relatively stable once I got everything set up. Here in Debian it seems to break in subtly different ways on every few-week upgrade. 06:39:49 This is the reverse of kmc's experience, I think. 06:40:22 haha 06:40:26 debian unstable? 06:40:36 i've never used ubuntu on my own machine for a long stretch of time 06:40:50 just in VMs and on other people's machines here and there 06:41:14 i think even debian unstable doesn't update the kernel that often? 06:41:14 Testing. 06:41:17 hm 06:41:27 you could pin the kernel to a specific version 06:41:34 It might have to do with the fact that I'm using a monstrosity of a system around the kernel. 06:44:12 http://newyork.cbslocal.com/2012/05/02/bosch-drowsiness-detection-system-to-make-alertness-tech-more-common/ 06:45:49 because the world demands people to drive while sleepy 06:46:00 shachaf: how so 06:46:29 I didn't feel like getting xmonad set up so I just tried to reproduce my lazy GNOME setup from before. 06:46:33 But GNOME 3. 06:46:47 So now I'm running xfce but with gnome-settings-daemon because I wanted something that gnome-settings-daemon did. 06:46:58 And also I'm running both NetworkManager and WICD at the same time, I think? 06:47:20 that sounds inadvisable 06:47:34 And I have a loop that runs "disable-capslock; xmodmap $REMAP_CAPSLOCK_THINGY" every two seconds, because xfce randomly interferes with GNOME's keyboard settings. 06:47:54 -_- 06:48:07 I should just set xmonad up. 06:48:13 i use wpa-supplicant, nothing higher level than that 06:48:25 I managed to get GNOME 2 to a perfectly usable state. I don't need much out of my WM. 06:48:31 Why did they have to go mess it up? 06:48:56 because GNOME 06:49:50 -!- asiekierka has joined. 06:49:53 did you run into trouble setting up xmonad, or just haven't got around to it? 06:50:12 also what about plain xfce, no GNOME? 06:50:36 Just haven't gotten around to it. 06:50:42 I have a really old xmonad.hs somewhere. 06:50:52 yeah, upgrading those can be a pain 06:50:53 I'm fine with that but gnome-settings-daemon did a few things that xfce didn't, or something. 06:51:06 maybe there is another way to do those things 06:51:21 which doesn't involve getting goatfucked by GNOME 06:51:35 Probably. 06:51:41 But it sounded like so much effort at the time. 06:51:44 this is really kind of a worrying state of affairs 06:51:55 Things like "set up multiple keyboard layouts without messing with Xorg.conf". 06:52:05 if GNOME is the main UI that's usable by beginners, but is hated and disowned by skilled programmers 06:52:15 down that road lies PHP 06:52:56 but a fair number of the skilled programmers i know run at least some parts of GNOME 06:53:00 so maybe it's not so bad 06:53:24 I'm really at the "don't care" stage. 06:53:36 I don't want too much out of my WM. I mostly use full-screen windows anyway. 06:53:57 But GNOME 3 felt like an early beta or something -- so much was broken and the rest was uncustomizable. 06:55:41 kmc: When are you going to write kmcwm? 06:56:03 Well, that was fun. 06:56:33 Tech writing, had to write instructions on "any topic"... 06:56:41 So, I explained the basics of Magic. 06:58:12 hi Magic 06:58:16 How does Magic work? 06:58:56 9 pages for a basic primer on the subject, apparently. 06:59:07 And by "basic" I mean I even omitted a card type. 06:59:12 (fuck planeswalkers) 06:59:15 haha 07:00:32 Of course, that's 9 pages with fairly defaulty TeX formatting. Large margins, ToC, whole page taken up by a picture of a card... 07:00:36 Eh, whatever. 07:01:53 Also, computer modern because I'm lazy and the professor seems to like everything else I've turned in in it. :P 07:02:27 LaTeX is a great way to get some unearned credibility 07:02:46 Greater still when you're actually writing Pandoc. :P 07:02:52 kmc: Is it as good as a clear plastic bider? 07:03:02 probably the dirtiest trick I ever pulled was, if I had a long sequence of algebraic manipulations, I would put the dodgiest step across a page break 07:03:05 pikhq_: lol do you think the reader of this document will learn Magic as a result? 07:03:17 itidus20: Quite plausibly. 07:03:19 pikhq_: which input format? 07:03:23 dun dun dunnnnnn 07:03:26 kmc: Markdown. 07:03:32 yeah, that's nice 07:03:44 kmc: Dodgy as in "unsure about it" or "don't feel like writing it out"? 07:04:02 It's like writing TeX, but much much easier on me. 07:04:50 unsure 07:05:19 i liked using markdown for beamer slides 07:05:31 but i did run into some limitations (which could have been fixed by writing a pandoc module, probably) 07:05:45 also it takes pdflatex a long time to compile 100 slides! 07:05:57 if i do another presentation I will seriously consider breaking it up and stitching the resulting pdfs 07:06:03 You should've used that slidy thing! 07:06:06 with a makefile 07:06:18 shachaf: *shrug* I've had bad experiences with HTML slideshows in the past 07:06:29 Yeah... I've been running my whole thing through Tup. 07:06:35 what's Tup? 07:06:59 Build automation tool, vaguely like make with better algorithms behind it. 07:07:23 Probably the nicest feature is its monitor, which can automatically rebuild stuff when the source has changed. 07:07:28 Oh, I thought it was the thing that puts multiple Postscript pages on a single printed page. 07:07:36 cool 07:07:39 http://gittup.org/tup/\ 07:07:42 Erm. 07:07:43 Oh, right. 07:07:43 http://gittup.org/tup/ 07:07:45 That thing. 07:07:51 The latter is psnup. 07:07:59 Right. 07:08:04 I was thinking of "2-up". 07:08:24 Typically like using it as I dev for fairly quick feedback on things. 07:08:32 Seeing if something compiles is as easy as :w 07:09:01 I usually do that by hand with inotifywait 07:09:07 "hand" 07:09:36 Shame that TeX building is a royal pain to automate with it, though. 07:09:53 (as TeX spews files all over the place and sometimes needs to run multiple times...) 07:09:57 yeah 07:10:04 also there's little opportunity for partial recompilation 07:10:09 if you have just a flat document 07:10:26 Shame that TeX is what it is. 07:10:40 So good and so bad all at the same time. 07:11:12 Good enough that no one will ever replace it, or something. 07:11:15 yeah 07:11:36 Along one dimension, at least. 07:12:04 Its output is good enough that it'd be really hard to even come close. 07:12:26 I wonder if it's the most popular Pascal program around these days. 07:12:54 Probably. 07:12:59 I know it's the only one on my system. 07:13:54 how does tup know which files etc. your compiler read? 07:14:00 does it use something like ptrace 07:14:02 pikhq_: Did you know execl("/usr/bin/tex", NULL); // => Segmentation fault 07:14:06 kmc: I think it uses FUSE? 07:14:12 whaaa 07:14:21 https://github.com/gittup/fuse 07:14:31 On Linux, it creates a FUSE filesystem in a dot directory, and runs the command there. 07:14:42 On Windows, it does some hack similar to LD_PRELOAD. 07:15:00 If LD_PRELOAD and/or ptrace is acceptable why can't it do that for both? 07:15:18 FUSe seems like overkill. 07:15:28 strange 07:15:31 FUSE is quite a bit better than LD_PRELOAD for the purpose. I bet ptrace would work decently, though. 07:15:48 Though, does ptrace work on OS X? 07:15:50 there are various situations where LD_PRELOAD would fail 07:15:55 apparently ptrace sucks on OS X 07:16:00 but there are other Mach interfaces you can use 07:16:09 FUSE works on every remaining UNIX, though. 07:16:14 -!- Deewiant has quit (Ping timeout: 265 seconds). 07:16:23 pikhq_: I guess it's more reasonable in Windows because LD_PRELOAD-hookable functions are the *only* portable way to access the filesystem. 07:16:38 Oh, sorry, every remaining UNIX *but OpenBSD*. 07:16:45 does Solaris have FUSE? 07:16:50 Yes. 07:16:54 (come on guys, even Hurd has FUSE!) 07:17:13 Hurd *is* FUSE, man 07:17:14 -!- Deewiant has joined. 07:17:14 dude 07:17:17 How about actual Unix®? Does it work on Unix®? 07:17:22 i should get Mosh working on Hurd 07:17:29 Yeah, they implemented FUSE by making a wrapper library. 07:17:33 that would be a laff 07:17:51 it will be a supported platform for Debian wheezy! 07:17:53 So much time, so little to do! 07:17:59 shachaf: It doesn't get more actual than the BSDs. 07:18:11 pikhq_: It -- does, doesn't it? 07:18:28 does AIX have FUSE? 07:18:38 I mean AT&T Unix. 07:18:40 No, but can you call that "remaining"? 07:18:49 latest stable release October 2011 07:18:52 according to Wikipedia 07:18:56 shachaf: Nah. 07:18:58 Oh, I thought you meant "remaining after you rule out Linux". 07:19:07 kmc: Bah, humbug. 07:19:10 latest release of HP-UX was last month! 07:19:23 I meant "remaining" as in "not dead or effectively so". 07:19:48 just because your hippie friends can't afford a real UNIX doesn't mean they're dead 07:19:51 ;P 07:20:07 Strictly speaking, that was 2007. 07:20:17 HP is just contractually obligated to provide patches until 2013. 07:20:18 * shachaf runs SCO OpenUNIX. 07:20:28 there was a MINIX release in February 07:20:35 Ooh, MINIX 3. 07:20:36 That's real. 07:20:38 -!- Slereah has quit (Ping timeout: 240 seconds). 07:20:42 Oh, right, MINIX. Yeah, that doesn't have FUSE. 07:20:43 so HP-UX will actually go away in 2013? 07:20:47 As real as HURD, anyway. 07:20:51 Probably. 07:20:54 http://pdos.csail.mit.edu/6.828/2011/xv6.html 07:20:58 AIX will stay around forever. 07:21:05 Hopefully. 07:21:24 * kmc hasn't actually read the xv6 source, JOS is more interesting 07:21:25 Before that, they released in 2003, then 2002, 2001, 2000, 1997, 1996, 1995, 1992, 1991... And so on. 07:21:54 1991 was a good year. 07:21:58 Yeah, that at least *suggests* HP-UX isn't going to last. 07:22:19 Oh, and it's for PA-RISC and Itanium machines only. 07:22:30 -!- Deewiant has quit (Ping timeout: 260 seconds). 07:22:44 AKA "dead" and "only technically alive", respectively. 07:22:50 JOS is pretty neat 07:22:58 you implement copy-on-write fork() in userspace! 07:23:03 the kernel gives you just enough rope to do so 07:23:13 -!- Deewiant has joined. 07:23:14 (the last PA-RISC came out in 2005. Some 3 orders of magnitude ago.) 07:23:19 What's JOS? 07:23:33 in MIT's 6.828 you write portions of an OS named JOS 07:23:43 the materials are all online; you should check it out 07:23:57 Ah, a teaching OS sort of thing. 07:24:04 yeah 07:24:25 Looks neat. 07:24:30 * shachaf should learn more about this topic. 07:24:31 I can't figure out why AIX is still alive, but then, this is IBM. 07:24:34 they call it an "exokernel" 07:24:44 it was never clear to me exactly what this means 07:24:58 They support a 50 year old ISA. 07:25:23 they designed that ISA at the outset knowing they would support it for 50 years :D 07:25:29 System/360? 07:25:34 yeah 07:25:55 ... Oh, wow, I can believe they did. 07:25:57 to be fair, they've extended it a lot, and now include support for 3 address sizes 07:25:58 it was meant from the beginning to be a kind of virtual thing that would upgrade to newer hardware 07:26:03 32-bit addressing in *1964*?!? 07:26:05 lol 31 bit architecture 07:26:18 pikhq_: it started as 24-bit 07:26:42 oh wait, I'm confusing 360 and 370 again 07:26:49 olsner: I'm discussing the ISA, not early implementations thereof. 07:26:52 maybe i am too 07:27:16 The ISA allowed 32-bit addresses even though the early chips didn't work when you went beyond 2^24. 07:27:45 This is ridiculous. 07:27:49 just like x86-64 today 07:28:42 shachaf: the primitive IPC message in JOS is a machine int plus optionally a page mapping 07:28:46 Also, you could operate on 128-bit values. 07:29:17 (though I think early ones couldn't do arithmetic or anything like that on them) 07:29:35 early IBM machines had hardware support for pre-decimal British £sd money 07:30:00 Wow. 07:30:15 in fact there were several different formats for storing same 07:30:20 which you could select with a front panel knob 07:30:40 I wish my computer had knobs. :-( 07:31:46 And blinkenlights? 07:32:46 fun stuff about z/Architecture is that you have something like 19 address spaces you can switch between and access with special instructions 07:33:30 http://upload.wikimedia.org/wikipedia/commons/4/4d/IBM_1401_Control_Panel.jpg 07:33:35 "IBM PENCE BSI SHILL" 07:33:42 home, primary, secondary, plus 16 "Access Registers" 07:34:00 64-bit, 31-bit... Presumably also 24-bit, just in case someone needs to run really weird FORTRAN, and? 07:34:11 Oh, address spaces, not schemes. 07:34:24 shachaf: you can attach some knobs and blinkenlights quite easily 07:34:39 i had an analog CPU meter for a while 07:34:48 I never asked for blinkenlichten. 07:35:01 This computer has three pseudo-buttons. 07:35:04 this one http://www.youtube.com/watch?v=vgZEfExBBiY 07:35:07 One of them sends Super-X. 07:35:29 One of them dims the screen by slightly less than the regular Fn-dim-screen button. 07:35:55 The third one pops up a big ⊘ on the screen. 07:36:29 shachaf: sounds incredibly useful 07:36:46 Doesn't it just? 07:36:56 Also I need to get my laundry from Outside. 07:37:01 kmc: That's awesome. 07:37:19 there used to be a Ksplice blog post about how i made it 07:37:20 * kmc pouts 07:37:27 but it's just an arduino and a resistor and a small python script 07:37:49 everyone on reddit was like "LAME Y DIDN'T U TAP CPU POWER PINS N00B" 07:37:57 Doesn't change the awesome. 07:37:59 it's an ammeter, you see 07:38:11 Tap CPU power pins? Eeeek. 07:38:29 my laser cutter has a meter on the front panel 07:38:31 labeled "Amperemeter" 07:38:33 -!- Deewiant has quit (Ping timeout: 260 seconds). 07:38:38 Analog is the new digital. 07:38:39 in that ugly serif font used for Latin letters in Chinese fonts 07:38:46 (the same font is on the rest of the front panel) 07:38:54 it doesn't say anywhere what component it's measuring the amperes of 07:39:30 also the lid interlock switch doesn't work, I should do something about that 07:39:42 Ah, ugly-large monospace. 07:40:14 shachaf: an awesome suckless CPU meter, done right, that celebrates craftsmanship 07:41:07 whoa, dude 07:41:26 -!- Deewiant has joined. 07:42:30 shachaf: I think it's cool to have an OS where the lowest level idea of how to send data around involves zero-copy transfers of page mappings 07:42:48 there are probably some "real" microkernels that do this too 07:42:58 Isn't that what L4's known for? 07:43:17 maybe 07:43:34 yeah i guess so 07:43:39 kmc: I came across a job ad that looked vaguely interesting today, but then I saw that it was for Ksplice. 07:43:45 haha 07:43:46 where? 07:44:05 i hear people are still pretty happy there 07:44:08 and you could move to boston 07:44:11 you could also not move to boston 07:44:50 If it's so great why aren't you there? 07:44:59 it's always better on holiday 07:45:04 that's why we only work when we need the money 07:45:31 And on holiday we port mosh to GNU/HURD. 07:45:36 yeah 07:45:38 maybe 07:46:23 I need to get my laundry. 07:46:32 I wonder how the cost of a 4kB memory copy compares to the overhead involved in mapping a new page 07:47:12 yeah 07:47:49 on x86 you mostly have to flush the TLB when context switching anyway 07:48:00 Completely? 07:48:04 yeah, so some of that mapping overhead can be made to disappear into things that you have to do "anyway" 07:48:09 but it could hurt when the two processes are running simultaneously on different cores 07:48:16 then you need to do an IPI to tell the other guy to flush his TLB 07:48:17 not fun 07:48:27 but you can also do new mappings / increased privileges lazily 07:48:37 However, 4kiB memory copy also messes with cache. 4kiB is a lot of cache. 07:48:59 kmc: at the cost of at least one fault in the future 07:49:01 on architectures with tagged TLBs, it might be more painful relatively speaking 07:49:04 yeah 07:49:13 -!- Deewiant has quit (Ping timeout: 245 seconds). 07:49:41 I mean, my L1 is 512k... 07:50:04 -!- Deewiant has joined. 07:53:48 I guess that's what those non-temporal move instructions are for 07:54:27 the TLB flush is a constant cost, while memory copy is not 07:54:30 so there is probably a break-even point 07:55:14 I wonder whether 4kB is still a reasonable page size these days. 07:55:31 It seems like there are a lot of things that huge pages are reasonable for. 07:55:40 Hence why Linux now supports them. 07:55:47 Right. 07:55:56 And I guess it does the transparent huge pages thing these days. 07:56:00 Yeah. 07:56:03 At least in theory. 07:56:09 Automagic 2MiB pages if you turn it on. 07:56:58 hi linux 07:57:07 monqy: You should draw a self-portrait of kmc. 07:57:22 oh dear 07:57:27 -!- azaq23 has quit (Max SendQ exceeded). 07:58:39 you can't draw self portraits of other people 07:58:44 PowerPC supports like 1 GB pages, doesn't it? 07:58:54 x86 does too, some models anyway 07:58:54 you can't write someone else's autobiography 07:58:57 I think 16GB? 07:59:11 PowerPC also has a crazy half-hard-filled "inverted page table" TLB 07:59:13 There was some architecture that supported that, anyway. 07:59:30 x86-64 does 1GiB. 07:59:38 * kmc added huge page support to JOS, but only for the kernel direct mapping 07:59:44 x86 could do 4TB pages by using a flag in the same manner it does 2MB/1GB pages 08:00:05 er, 512GB pages 08:02:25 Linux also has a thing to transparently merge pages with the same content 08:02:58 Really? 08:03:04 How does it find them? 08:03:06 for pages that aren't all-zero too? 08:03:24 i think so 08:03:35 shachaf: by scanning all of memory periodically 08:03:40 i think it's not enabled by default... 08:03:50 but can improve memory consumption dramatically on machines hosting a large number of similar VMs 08:04:17 Ah, hmm. 08:06:03 It's enabled by default *but* only scans pages with the appropriate madvise() set. 08:06:26 http://en.wikipedia.org/wiki/Kernel_SamePage_Merging_(KSM) ? 08:06:27 ah yeah 08:06:30 MADV_MERGEABLE, apparently. 08:08:04 -!- ais523_ has joined. 08:08:10 -!- ais523_ has quit (Changing host). 08:08:10 -!- ais523_ has joined. 08:08:10 -!- ais523 has quit (Disconnected by services). 08:08:23 -!- ais523_ has changed nick to ais523. 08:12:23 -!- ais523 has quit (Read error: Connection reset by peer). 08:12:28 -!- ais523_ has joined. 08:12:35 -!- ais523_ has changed nick to ais523. 08:18:07 kmc: Was there a verdict about mosh mouse? 08:18:22 we were just talking about that in #mosh 08:18:43 the verdict is that it's a mess and i don't want to do it 08:18:44 https://github.com/keithw/mosh/issues/101 08:18:54 Oh, yes, you were. 08:19:02