00:00:01 <ehird> http://www.reghardware.co.uk/2009/04/17/pedisedate/ ← AAAAAA
00:00:19 <ais523> wow, my .mozilla is rather large
00:00:25 <AnMaster> ais523, you don't know what's on your disk?
00:00:29 <ais523> that's what caused me a problem over on the Sun computers
00:00:39 <ais523> AnMaster: not /everything/, I have loads of packages installed
00:00:42 <AnMaster> ais523, *.sqlite files in could be large in it IME.
00:00:58 <ais523> ~/.local is nearly a GB, I wonder what's in there?
00:01:12 <ehird> My main issue with my new system is partitioning between the two drives.
00:01:25 <AnMaster> ais523, try (after quitting firefox) cd ~/.mozilla/<path to profile>; for i in *.sqlite; do sqlite3 $i VACUUM; done
00:01:27 <ais523> and ~/nwn is 3 GB, but that's not surprising
00:01:32 <AnMaster> that usually frees quite some space
00:01:34 <ais523> AnMaster: no need, I'm not low on space yet
00:01:43 <AnMaster> ais523, it speeds up firefox too
00:01:50 <pikhq> ehird: Moral of the story, use LVM. Repartitioning is easy. :)
00:01:55 <AnMaster> ais523, and reduces RAM needed.
00:02:02 <AnMaster> ais523, needs to be redone every few weeks.
00:02:11 <ais523> why doesn't firefox do it automatically?
00:02:15 <pikhq> Have to unmount to shrink, probably, but nice and easy to grow the filesystem.
00:02:16 <ehird> pikhq: I'm scared of LVM though,.
00:02:34 <ehird> pikhq: Also, that doesn't impress me.
00:02:52 <AnMaster> the tools are a pain to use. But it is very flexible.
00:02:54 <ehird> pikhq: I deleted a dormant partition and resized my main one to fill its space, all from within Disk Utility, instantly taking effect.
00:02:57 <AnMaster> That is in fact the issue of it.
00:02:58 <pikhq> The main reason you have to unmount to shrink is that most *filesystems* don't support live shrinking.
00:03:39 <AnMaster> pikhq, indeed. IIRC jfs or some does support it however?
00:03:39 <pikhq> I can also move partitions to a different disk without unmounting.
00:03:41 <ehird> My magical calculation gods tell me that my OS (i.e., non-home) currently uses 70GB.
00:04:17 <ehird> OS X is rather the big.
00:04:24 <ehird> Especially when you have as much crap as I.
00:04:26 <pikhq> I've got 32GB used, and that's because I've got a lot of junk I don't need installed.
00:04:33 <pikhq> Oh, and that's including full source.
00:04:40 <AnMaster> pikhq, is it JFS that too? which command?
00:04:54 <ehird> Still, brushing 70GB confirms that I'm right not to go for the 80GB X25-M.
00:05:06 <ehird> I'd rather have a ton of unused space than have to, you know, delete stuff.
00:05:08 <pikhq> That's lvm. pvremove.
00:05:28 <AnMaster> pikhq, um.. For *shrinking file system*
00:05:50 <oklopol> ais523: about being low-end, i think my father still uses his 300mhz win95 machine
00:06:02 <AnMaster> I'm sure I saw some fs supporting online shrink
00:06:09 <ais523> I still have a Win95 machine
00:06:12 <ais523> I haven't booted it up in a while
00:06:13 <ehird> oklopol: I had one of those a while back
00:06:19 <ehird> I thought I'd be awesomely retro with it
00:06:21 <ais523> but I'm insisting on keeping it because some things just aren't compatible with XP
00:06:23 <ehird> Instead I threw it away
00:06:38 <ais523> ehird: how are you going to develop 16-bit Windows applications now?
00:06:56 <ehird> ais523: by wishing I'd never been born? Isn't that how it's always been done?
00:07:24 <AnMaster> ais523, why would you want to do that?
00:07:38 <oklopol> i could easily do with a 300mhz computer
00:07:43 <ais523> AnMaster: because the 32 bit APIs don't let you access the internal system speaker?
00:07:54 <AnMaster> ais523, also there is still some support *.com for 16-bit emulation on xp
00:07:55 <oklopol> except for them internets.
00:07:58 <ehird> that would be rather painful
00:07:59 <ais523> sure, you can beep, but you can't mess with the pitch or duration
00:08:08 <ais523> AnMaster: yes, but it isn't perfect
00:08:18 <ais523> the beep APIs don't work
00:08:24 <oklopol> ehird: i mainly just need irc... but i guess that would require a bit more
00:08:25 <ais523> although admittedly they were deprecated in Windows 3.1
00:08:30 <AnMaster> ais523, write a kernel driver?
00:08:37 <ais523> but more to the point, the replacement that Microsoft suggested for it doesn't work in XP either
00:08:43 <AnMaster> ais523, Or switch to Linux or *BSD
00:08:49 <ais523> it hangs for about 5 minutes whenever you load a MIDI file
00:08:50 <AnMaster> where you can mess with the internal speaker easily
00:08:54 <ais523> AnMaster: what do you think I did, eventually?
00:09:04 <ais523> I developed Windows applications for /years/
00:09:08 <oklopol> well actually you could just have a slower connection and there would be no problem
00:09:10 <AnMaster> <ais523> it hangs for about 5 minutes whenever you load a MIDI file
00:09:28 <ais523> that's Windows XP using MCI as the API
00:09:36 <ais523> it worked fine in previous versions
00:09:36 <ehird> ais523: i want dna maze
00:09:47 <ais523> ehird: I can send you a Window executable if you like
00:09:54 <ehird> that would not be useful :P
00:10:01 <ais523> the sources don't compile any more, unfortunately
00:10:14 <ais523> they relied on a proprietary Borland extension to the Win16 API
00:10:22 <ais523> which Borland don't support any more for obvious reasons
00:10:37 <AnMaster> I have a copy of turbopascal for DOS somewhere. Just FYI.
00:10:38 <ais523> AnMaster: computer game, that I've been developing since 2001
00:10:44 <ais523> AnMaster: so do I. and it still runs.
00:10:57 <AnMaster> no idea where I have it though
00:11:02 <ais523> not on this computer, though, on the Win95 computer
00:11:05 <AnMaster> I also have a working copy of MPW!
00:11:33 <ais523> I have a copy of Windows Media Player from before when it was rubbish
00:11:37 <AnMaster> horrible thing with a pseudo shell thingy
00:11:41 <ehird> "Expensive extras include 2GB of memory, a SATA-based array of .5 terabytes of storage"
00:11:54 <ehird> ais523: look up "Media Player Classic"
00:12:03 <ehird> it's old WMP interface, modern codec support etc
00:12:08 <ehird> AnMaster: My first mac was this one
00:12:15 <ais523> sounds interesting, but nowadays I just use Totem or Rhythmbox
00:12:26 <ais523> depending on what I want it for
00:12:49 <AnMaster> http://developer.apple.com/tools/mpw-tools/
00:12:57 <AnMaster> they must have forgot that page
00:13:17 <ehird> just left for history
00:13:20 <AnMaster> also: http://en.wikipedia.org/wiki/Macintosh_Programmer%27s_Workshop
00:13:52 <ehird> MPW was supposed to be quitegood
00:14:15 <AnMaster> http://en.wikipedia.org/wiki/ResEdit
00:14:58 <ehird> http://www.ralentz.com/old/mac/humor/mpw-c-errors.html
00:15:44 <AnMaster> "...And the lord said, `lo, there shall only be case or default labels inside a switch statement'" <-- someone hating duff's device?
00:16:05 <ehird> is illegal, I guess
00:16:12 <coppro> duff's device doesn't use non-{case,default} labels
00:16:25 <ais523> ehird: no it isn't, I think it's switch(foo) { } case 1: that's illegal
00:16:30 <ais523> and that that's warning against
00:16:43 <ais523> i.e. it's a misplaced only, telling people off for putting a case outside a switch
00:16:45 <AnMaster> I'm not going to boot my old mac now
00:17:00 <ehird> I'm not going to have a heart attack now.
00:17:38 <ais523> as in, the entire source tree
00:17:46 <ais523> I went and compiled clang from source, as it wasn't packaged
00:17:50 <ais523> and the rest of LLVM at the same time
00:18:00 <ais523> yes it is, I just du'ed it
00:18:05 <AnMaster> ais523, let me boot the computer that has the llvm source tree on it *sec
00:18:12 <coppro> I have several GB of gcc on this computer
00:18:15 <ais523> that's after compile, so it has all the binaries in too
00:18:22 <AnMaster> that is how long it takes from button to ssh being up
00:18:23 <ais523> coppro: have you messed with gcc internals, ever?
00:19:00 <ais523> I should get back to gcc-bf some time
00:19:01 <AnMaster> ais523, I'm running du -sh on the llvm svn checkout
00:19:27 <ehird> I wonder what to call my new box.
00:19:32 <ais523> llvm/Debug/bin and llvm/Debug/lib are what are taking up most of hte space, it seems
00:19:45 <ais523> so it's a debug build of the whole of llvm that takes up the space
00:19:49 * AnMaster use phoenix and tux for his boxes
00:19:54 <AnMaster> ais523, this is a release build though
00:19:58 <ehird> Cheetahs are the ones that run really really fast, silently, right?
00:20:02 <ais523> ah, that must be why it's smaller
00:20:06 <ehird> So cheetah, or ninja or something :P
00:20:07 <AnMaster> ehird, sounds like a OS X release
00:20:15 <ais523> ninja is a good name for a box, actually
00:20:23 <ehird> ais523: have you ever heard a cheetah make noise while running?
00:20:29 <ehird> also, yes, but it seems a lil cliche
00:20:47 <ais523> who calls a system "ninja"?
00:20:54 <ehird> ais523: many people, I'd guess
00:20:56 <AnMaster> "tux" for a linux box is cliche though!
00:21:18 <ehird> I wonder if there's a way to cache small writes to an HD
00:21:20 * AnMaster is writing this on the local terminal of tux.lan
00:21:24 <ehird> It writes to somewhere faster first
00:21:27 <ehird> then copies to HD in the bg
00:21:32 <ehird> to avoid pauses due to HD writes
00:21:40 <ehird> e.g. in IM/IRC clients when writing to log
00:21:45 <AnMaster> ehird, yes. memory caches. The OS does them.
00:21:46 <ais523> ehird: that's every filesystem ever that does that
00:21:55 <ais523> ext4 in particular caches quite a lot, it's come under fire for that
00:21:56 <AnMaster> ais523, not FAT under DOS iirc?
00:21:58 <oklopol> ehird: what these guys said
00:21:58 <ehird> Then why were the pauses a proble mwith SSDs?
00:22:08 <ais523> AnMaster: not by default, but everyone installed smartdrv or something
00:22:12 <ehird> ais523: It looks more and more like I should use ext4 for the OS thang.
00:22:28 <ais523> AnMaster: FAT under DOS didn't cache reads either
00:22:29 <ehird> Still undecided for the home one (everything from small text to multigb files)
00:22:32 <oklopol> annoying how i can shine only every 2 hours or so, and even then i have tons of competition :P
00:22:34 <ais523> so smartdrv made a system a lot faster
00:22:35 <ehird> Suggestions welcome
00:22:38 <coppro> coppro, and my GCC tree
00:23:03 <oklopol> AnMaster: maybe you should get a better client
00:23:16 * coppro hates the evils of most IRC clients' tab-completes
00:23:19 <pikhq> AnMaster: Pvmove, I meant. Really.
00:23:21 <AnMaster> actually the version reply is not on the client I'm on atm.
00:23:26 <ais523> wow, ~/.strigi is 13 GB
00:23:29 <AnMaster> I have two clients connected to the bouncer
00:23:38 <coppro> I have been called "Coboney" more times than I can think of
00:23:47 <ais523> Strigi's the KDE indexing daemon, i think
00:23:49 <ehird> 00:22 ehird: Still undecided for the home one (everything from small text to multigb files) ← noone have surgerestions?
00:23:57 <ais523> as opposed to Tracker which is used on Gnome Ubuntu
00:24:07 <pikhq> ehird: ext3 works just fine.
00:24:09 <ais523> I installed both Gnome and KDE, I should just uninstall one of the daemons as I don't need both
00:24:10 <coppro> yeah, you can turn that off
00:24:15 <AnMaster> ais523, I hate all indexing daemons apart from locate
00:24:24 <coppro> open strigi:// in Konqueror I think
00:24:30 <ehird> pikhq: For data? Hmm... I don't think I want journaling...
00:24:39 <ehird> pikhq: Think of the 2GB files!
00:24:52 <AnMaster> ais523, other ones are just overhead that impacts both space and performance.
00:25:13 <ais523> anyway, my backup strategy is interesting
00:25:29 <ais523> it's not just designed to help against broken drives
00:25:33 <ais523> but also against an accidental rm -r .
00:25:51 <ais523> for instance, I have Emacs storing all its backups in a separate directory tree from the files being edited
00:25:54 <ais523> that's come in useful before now
00:26:07 <ais523> started ever since I accidentally deleted the Unlambda -> Underlambda compiler
00:26:14 <ais523> I should reconstruct it sometime
00:26:14 <ehird> ais523: i did that anyway, to reduce clutter
00:26:16 <AnMaster> "This label is the target of a goto from outside of the block containing this label AND this block has an automatic variable with an initializer AND your window wasn't wide enough to read this whole error message"
00:27:06 <AnMaster> "You are comparing two structures that have holes in them" <-- um. Comparing structures?
00:27:32 <ais523> AnMaster: prompts if the rm isn't trivial
00:27:34 <AnMaster> ais523, well yes, that seems odd for a compiler to be able to figure out
00:27:58 <ais523> to be precise, -I prompts once if you specify more than 3 files, or use -r
00:27:59 <coppro> I should probably prefer that to -f...
00:28:20 <coppro> does it still ignore read-only?
00:28:26 <ais523> many people have alias rm = rm -I
00:28:44 <ais523> coppro: you need -f as well to ignore read-onlyness
00:29:12 <ais523> but -rIf is completely legal
00:29:57 <ais523> lowercase i is a lot more prompty than capital I
00:30:06 <ais523> it's effectively a darcs-interface rm
00:32:04 <AnMaster> ais523, yeah darcs is irritatingly unsure of it's own capabilities, feeling a need for asking the user all the time
00:32:20 <ais523> darcs is irritatingly incapable of reading minds
00:32:52 <coppro> I wonder what a Perl implementation of darcs would be like
00:33:17 <AnMaster> coppro, less monads, more line noise.
00:33:25 <pikhq> ehird: YOU WANT JOURNALING.
00:33:33 <ais523> pikhq: but even ext3 has journaling
00:33:46 <pikhq> ais523: And ehird said "I don't think I want journaling".
00:33:58 <AnMaster> <ehird> pikhq: For data? Hmm... I don't think I want journaling...
00:33:58 <AnMaster> <ehird> pikhq: Think of the 2GB files!
00:34:35 * pikhq points at his two 21GiB files.
00:34:38 <ais523> on a disk that large, you'll have to spend days fscking if you have a power cut
00:34:44 <ais523> if you don't have journaling
00:35:09 <pikhq> Why I have losslessly compressed copies of Big Buck Bunny & Elephants Dream, I'm not quite sure.
00:35:16 * AnMaster ponders a portable cfunge for Mac OS 9. Just because.
00:35:20 <coppro> Is there a downside to hournaling?
00:35:25 <AnMaster> somewhat like openssh vs portable openssh
00:35:29 <pikhq> coppro: Very slightly slower.
00:35:38 <AnMaster> except, it won't support windows.
00:35:50 <ais523> oh, yes, there are ext2 drivers for Windows, but not ext3 drivers
00:35:51 <pikhq> But it gets you a significantly more robust filesystem.
00:35:53 <ais523> they aren't official, anyway
00:35:55 <AnMaster> ais523, what do you think? Insane enough?
00:36:08 <AnMaster> ais523, anyway does ick work on pre-OS X MacOS?
00:36:12 <ais523> portable /and/ OS9 only?
00:36:14 <pikhq> ais523: And ext3 can be mounted as ext2?
00:36:18 <coppro> the IFS works pretty well on ext3, even when the disk has journaling, as long as the mount/unmounts are clean
00:36:19 <ais523> also, I haven't tested, but I wouldn't be surprised
00:36:23 <coppro> ext3 is otherwise backwards-compatible
00:36:23 <ais523> pikhq: yes if the journal is up-to-date
00:36:47 <pikhq> Well, yes. Sorry, I should've said: assuming clean mounts.
00:36:52 <ais523> the ext2 standard has a flag saying "don't try to mount as ext2", ext3 uses that flag for "unreplayed journal data"
00:36:56 <ais523> which is not a coincidence at all
00:37:27 <pikhq> That's good design right there.
00:37:44 <coppro> I gave up on hoping for Windows mounts when I turned on extents
00:38:22 <AnMaster> how can anyone be insane enough to use that yet for anything important...
00:38:22 <ais523> they massively reduce the amount of metadata needed for very big files
00:38:36 <AnMaster> ais523, wouldn't help for me then
00:38:38 <ais523> AnMaster: some of the more up-to-date distros have it as stable
00:38:42 <AnMaster> I mostly have loads of small files
00:38:46 <oklopol> what are you talking about?
00:38:52 <AnMaster> a few kb at most most of the time
00:39:08 <ais523> there's more than one way to arrange data on a drive
00:39:09 <AnMaster> coppro, the online defrag tool is _still_ missing for ext4
00:39:14 <ais523> Windows uses FAT/FAT32/NTFS
00:39:19 <ais523> but there are a lot of other filesystems around too
00:39:20 <AnMaster> I'm going to stay to ext3 and xfs for the next few years.
00:39:39 <ais523> AnMaster: the online defrag tool is _still_ missing for ext3 too...
00:39:48 <oklopol> ais523: yes but what are these like, journaling sounds like the log-structured file systems tannerbaum talks about in MOS
00:39:51 <AnMaster> ais523, yes but it wasn't even promised for it.
00:40:03 <AnMaster> which is a rather large difference.
00:40:04 <ais523> oklopol: journaling means you record what you're doing before you do it
00:40:15 <pikhq> I'm waiting for kernel versions with ext4 marked as stable to be stable in Gentoo, myself.
00:40:18 <ais523> if the system crashes while you're doing it, the log makes it possible to reconstruct the filesystem
00:40:28 <ais523> rather than having to run a full disk-check the next boot like Windows does with scandisk
00:40:32 <coppro> the biggest mistake I made when updating was that ext4 is not compiled by default into the Intrepid kernel, so my kernel couldn't mount /
00:40:44 <ais523> it is compiled into Jaunty by default
00:40:53 <ais523> yes, but Windows still does the on-boot scan anyway
00:41:00 <ais523> and you'd be surprised how many people are still on FAT32
00:41:08 <oklopol> ais523: right, that's kind of a trivial idea tho, so not very useful to me
00:41:16 <ais523> oh, have they updated now?
00:41:26 <ais523> I'm a bit out of touch on windows filesystem scanning programs
00:41:28 <AnMaster> ais523, the name of the tool isn't scandisk in xp it is autochk
00:41:49 <ais523> well, I've seen XP use scandisk
00:41:53 <ais523> going home now, anyway
00:41:54 <pikhq> Win32 is the NT native API.
00:42:09 <pikhq> It's built into the fucking kernel.
00:42:13 <AnMaster> pikhq, win32 is a NT native program itself.
00:42:20 -!- ais523 has quit (Remote closed the connection).
00:42:44 <AnMaster> pikhq, http://technet.microsoft.com/en-us/sysinternals/bb897447.aspx
00:42:53 <AnMaster> pikhq, just read the first paragraph
00:42:57 <pikhq> In theory, that's how the NT kernel was designed.
00:43:18 <pikhq> Microsoft's formal design and the actual implementation differ, in that Win32 runs in kernelspace.
00:43:46 <AnMaster> pikhq, sure. But there are apps using this native API. Of which autochk is one
00:44:18 <AnMaster> as is described in the second paragraph
00:44:29 <pikhq> Sounds like very poor terminology.
00:44:42 <pikhq> Sounds more like they mean "system calls".
00:45:19 <AnMaster> pikhq, maybe. Maybe not. But I'm just using the official name of it.
00:45:41 <pikhq> Okay, that's what Microsoft calls it.
00:45:53 <pikhq> Yet another thing that MS does dumb.
00:46:02 <AnMaster> pikhq, we shouldn't call this think MS calls "xp" for xp.
00:46:10 <AnMaster> we should call it Windows NT 6.1
00:46:19 <AnMaster> which iirc is the internal version number
00:46:37 <AnMaster> when it comes to the getwinverex (or whatever) win32 api functions
00:46:54 <pikhq> It's Windows NT 5.1, but yeah.
00:47:13 <coppro> "initialization Blue Screen"
00:47:26 <coppro> probably not what they really meant
00:47:30 <pikhq> I'm also of the opinion that Windows 95 should've been called Windows 4.0. ;)
00:47:30 <AnMaster> coppro, um... never used /sos for flag in boot.ini?
00:48:04 <coppro> in the context, they mean the screen that comes up when the disk check runs
00:48:07 <AnMaster> coppro, it shows a blue screen with text instead of the xp logo then
00:48:16 <coppro> not an actual Blue Screen
00:48:17 <AnMaster> and yes it is the same for disk check
00:48:29 <AnMaster> coppro, that screen is actually blue though!
00:48:30 <coppro> Blue Screens are darker
00:50:02 <coppro> yes, I'll agree the screen is blue. But not Blue
00:52:24 <pikhq> Ah, Windows NT. The MS OS with decent design fouled up by poor coders and poorer additions...
00:52:52 <AnMaster> pikhq, to be fair: 2000 and xp weren't *too* bad.
00:53:01 <AnMaster> I mean I consider the operating systems
00:53:14 <AnMaster> 9x and vista I consider "jokes"
00:53:26 <AnMaster> haven't heard enough about 7 yet to be sure
00:53:33 <coppro> actually, ME isn't a joke
00:53:37 <AnMaster> coppro, that is 9x technically
00:54:06 <pikhq> AnMaster: The API still has ended up being really, really screwy, because of its lack of sensible design after the fact.
00:54:31 <AnMaster> pikhq, at least POSIX only got a *few* such quirks
00:54:56 <pikhq> And shockingly few for being 30ish years old now.
00:54:57 <AnMaster> and them I'm specifically thinking about mkstemp() tmpnam() and so on. And signal()
00:56:02 <AnMaster> mkstemp, mkostemp, mkdtemp are more or less sane. tempnam, tmpnam, tmpfile are not.
00:56:11 <pikhq> ReactOS seems rather nice.
00:56:37 <pikhq> Shame they've got to deal with the Win32 API quirks, but they seem to be doing more with the NT kernel's ability to support multiple APIs.
00:56:44 <AnMaster> Actually that's a lie. Genera wins over Plan9
00:56:54 <pikhq> ReactOS 1.0 might make me enjoy using a Windows-esque OS.
00:57:33 <pikhq> Assuming that I can use it mostly using the POSIX personality, of course. :p
00:57:34 <AnMaster> Error 500: The qualifier "enjoy" can not be used for the "windows" noun.
00:57:58 <pikhq> s/personality/subsystem/
00:58:25 <AnMaster> pikhq, "personality" is Linux terminology right?
00:58:50 <AnMaster> as in linux32 on multilib 64-bit systems
00:59:00 <pikhq> Unix-ish terminology.
00:59:04 <AnMaster> Linux tux.lan 2.6.27-gentoo-r8-1 #2 Thu Apr 16 19:43:25 CEST 2009 x86_64 AMD Sempron(tm) Processor 3300+ AuthenticAMD GNU/Linux
00:59:08 <AnMaster> Linux tux.lan 2.6.27-gentoo-r8-1 #2 Thu Apr 16 19:43:25 CEST 2009 i686 AMD Sempron(tm) Processor 3300+ AuthenticAMD GNU/Linux
00:59:30 <pikhq> Though linux32 on x86_64 isn't using a personality, just changing a very small handful of system calls.
00:59:57 <AnMaster> pikhq, actually it is. Running a 32-bit app isn't though
01:00:14 <AnMaster> int personality(unsigned long persona);
01:00:45 -!- oklokol has joined.
01:02:15 <AnMaster> pikhq, linux32 is a symlink to setarch btw
01:02:40 <AnMaster> http://rafb.net/p/N8i5es35.html
01:03:35 <pikhq> AnMaster: Shell script that calls setarch, you mean.
01:03:47 <AnMaster> lrwxrwxrwx 1 root root 7 1 dec 09.27 /usr/bin/linux32 -> setarch
01:03:56 <AnMaster> it detects what name it is invoked with
01:05:02 <pikhq> It *was* a shell script once. I'm sure of it.
01:07:11 <AnMaster> "To avoid performance problems you should keep the following guidelines in mind, which are separated by architecture:
01:07:11 <AnMaster> IA-32, Intel® 64, and IA-64 architectures:
01:07:11 <AnMaster> Do not access or create data at large intervals that are separated by exactly 2n (for example, 1 KB, 2 KB, 4 KB, 16 KB, 32 KB, 64 KB, 128 KB, 512 KB, 1 MB, 2 MB, 4 MB, 8 MB, etc.)."
01:07:37 <AnMaster> why is accessing two structures separated by 1 KB of memory a bad idea exactly..
01:15:24 <pikhq> Um. What in the world does Intel's caching setup look like?
01:15:30 -!- Sgeo has joined.
01:16:32 <AnMaster> oh maybe that is used for associativity at L1?
01:17:02 <pikhq> I guess I could see that making sense for L1.
01:17:35 <pikhq> I can understand some of the specific intervals...
01:17:42 <AnMaster> yes for L1 it would, not for L2 though
01:17:43 <pikhq> Once you're dealing with page boundaries, that is.
01:17:58 <AnMaster> pikhq, that only happens at 4 kB and above
01:18:32 <AnMaster> pikhq, but a L1 wouldn't be 4KB
01:19:57 <AnMaster> you are talking about a few cache lines
01:20:10 <AnMaster> on AMD64 they are 64 bytes iirc
01:20:19 <AnMaster> not sure how large they are on intel CPUs
01:25:57 <oklokol> AnMaster: Do not access or create data at large intervals that are separated by exactly 2n (for example, 1 KB, 2 KB, 4 KB, 16 KB, 32 KB, 64 KB, 128 KB, 512 KB, 1 MB, 2 MB, 4 MB, 8 MB, etc.)." <<< obviously those would all be in the same cache slot
01:26:15 <oklokol> that's like basic cacheology
01:27:48 <oklokol> usually the actual "set" part of a cache is small
01:28:11 <oklokol> and you have multiple small sets hashed into based on the address
01:28:44 <coppro> I understand how caching works less than I understand timecube
01:29:57 <oklokol> logically the cache should be a set
01:30:15 <oklokol> that it would just be a set of data currently in cache, and stuff would just either be in the set, or not be
01:30:20 <oklokol> but in practise, this is not doable
01:31:28 <oklokol> so what you do is first take certain index bits from the address (you can use any hash, but clearly it can't take much time to compute, because the idea is to save time) and that address gives you a set to use
01:31:37 <oklokol> and that set will be fully random access
01:32:11 <oklokol> dunno if that cleared it up, point is what AnMaster said seems kinda trivial, no reason to question intel's caches because of it.
01:32:23 -!- coppro has quit (Read error: 60 (Operation timed out)).
01:34:29 <oklokol> i guess the basic idea of timecube is trivial too, the 24h day can be split into 4 pieces.
01:34:49 <oklokol> or maybe that was not the most important part, dunno.
01:35:11 <AnMaster> oklokol, yes I know how a cache works
01:35:26 <oklokol> okay, then why the confusion?
01:36:30 <oklokol> i'll just assume you didn't know, i'll feel intelligent and you don't need to answer :)
01:37:04 <pikhq> I don't pretend to know anything about how CPUs are actually implemented.
01:37:23 <pikhq> My knowledge stops a bit before assembly gets converted to bytes.
01:37:32 <AnMaster> oklokol, I think there are better ways to do it.
01:38:40 <oklokol> AnMaster: and i mean that as a question, all i know about caches is a few slides for a processor architecture course.
01:38:48 <AnMaster> oklokol, even if you exclude the clear way of making all RAM register file speed (which has a downside in the form of the cost) it would be trivial to base it on a bit more than just least significant bits of the address
01:38:55 <oklokol> (not really the main topic obviously)
01:39:05 <AnMaster> you could take a bit or two from higher up
01:39:47 <oklokol> the lsb'er bits you take the better
01:40:07 <AnMaster> oklokol, since many common access patterns have a 2^n unit stride.
01:40:10 <oklokol> taking a few from higher up wouldn't change the current issue, but overall it would be a worse solution
01:41:03 <oklokol> the bits would be the same?
01:41:15 <AnMaster> not if you take the right ones due
01:41:37 <oklokol> you take the lsb'est bits you can, usually
01:41:52 <AnMaster> but rather a few bits above, like the ones important for a common stride of 4 kb
01:41:55 <oklokol> from that to right is the address inside the block
01:42:02 <AnMaster> you very often operate at two pages
01:42:07 <AnMaster> equally from the start of both
01:42:24 <AnMaster> that is in fact very very common
01:43:02 <oklokol> so you're basically saying caches shouldn't have the basic block structure at all?
01:43:15 <oklokol> that you store stuff in blocks of size 2^n
01:44:08 <oklokol> anyway, that's what the set part of the cache is for, that you can do something like that
01:44:17 <AnMaster> I'm just saying having same associativity for the first cacheline of any two different pages is rather bad
01:44:42 <oklokol> right, and as a correction you suggest what?
01:44:48 <oklokol> taking a few bits from MSB?
01:45:17 <AnMaster> well possibly a different scheme
01:45:37 <AnMaster> lets do unary xor on the address. intercal style
01:46:20 <oklokol> or wut. usually the idea is when you have an address TT..TII..IBB..B, the II..I part is used as the hash, to determine the set, then TT..T is used to find the correct slot in the set, or to see it's not there, then BB..B is used to address within the block.
01:47:03 <oklokol> AnMaster: anyway it's true it might make sense to use a more random hash, that i'm not disagreeing with.
01:47:06 <AnMaster> how often have you not done memcpy between two page aligned adresses.
01:47:28 <oklokol> sure, but that's not a problem.
01:48:39 <oklokol> i mean there aren't going to be conflicts if you just use a few pages of stuff and you need to use the same part of each page at a time
01:49:21 <oklokol> the problem is when you have many pages, and use the same part of each
01:49:31 <oklokol> even if you don't actually explore the other parts of the page
01:49:34 <AnMaster> oklokol, what about array[1024][1024]
01:49:42 <AnMaster> you are going into issues there
01:49:45 <oklokol> AnMaster: things like that compilers can optimize.
01:50:17 <AnMaster> oklokol, I'm doing random-connected access in it though
01:50:58 <AnMaster> oklokol, I need power of two for fast range checks (bit shift)
01:51:00 <oklokol> that may actually be a weak spot.
01:51:20 <oklokol> usually you just stride one way at a time
01:51:26 <AnMaster> oklokol, I also need power of two for various other reasons
01:51:30 <oklokol> but funges can have local patterns of both kind
01:52:00 <oklokol> well anyway, i don't claim to have given this stuff any thought, and i most certainly don't know anything about it :)
01:52:05 <AnMaster> total size needs to be a multiple of 64 bytes for exampl
01:53:53 <oklokol> also i'm not exactly that interested in computer architecture in general, kind of an inexact science
01:58:06 <oklokol> i remember deciding to go to sleep at 23:00
01:59:13 -!- serbz has joined.
01:59:16 -!- serbz has left (?).
03:14:47 <Sgeo> Is there a way, given discrete points that have connections north, south, east, and west to other discrete points, and each point knows only of itself and its direct neighbors, to make sure that the whole structure makes sense?
03:17:24 <oerjan> do they know the distances between them?
03:17:40 <oklopol> what does that mean makes sense
03:18:31 <oerjan> i assume he means that they can be put as coordinates on a flat map...
03:19:33 <Sgeo> The distance between a point and its neighbor is one
03:20:18 <oklopol> Sgeo: just plot them into an array
03:20:29 <oerjan> in that case it should be simple: pick a point, give it coordinates (0,0), trace through the whole structure assigning coordinates, and checking that any point you revisit matches
03:20:54 <Sgeo> I want points to be able to add themselves to the structure just knowing one other point
03:21:17 <oerjan> each point needs to know its coordinate of course
03:21:41 <oklopol> Sgeo: it may have to traverse an arbitrary amount of links to know whether a square is free
03:21:41 <oerjan> the only remaining problem i see is if two points get the same coordinates
03:21:56 <oerjan> which is equivalent to what oklopol said
03:22:35 <oerjan> yeah, and also to find its _other_ new neighbors
03:22:52 <oerjan> say if (0,0) and (0,2) already exist and you add (0,1) from (0,0)
03:23:12 <oerjan> then it could require a long path to check that you need to connect to (0,2) as well
03:23:30 <oklopol> so basically there's no local way to do it, i guess that was the question
03:23:39 <oklopol> as it is algorithmically trivial otherwise
03:25:36 <oklopol> so, anyone know a good chess ai library?
03:25:38 <oerjan> so you are going to need some overarching structure.
03:26:22 <oerjan> each point is a square inside a 2x2 square inside a 4x4 square etc. until whatever fills all of it
03:26:23 <oklopol> you mean like, if you only add next to existing nodes, oerjan?
03:26:50 <oerjan> then you should need only O(log n) time to fix things
03:27:12 <oerjan> also, i don't know anything about chess ai libraries
03:27:26 <oklopol> oerjan: also he could just have the NxN array..................
03:27:46 <oklopol> then again i guess that was specifically forbidden
03:27:50 <oerjan> yeah but that could require much more memory
03:28:07 <oklopol> welllll i guess, but the distances were one
03:28:10 <oerjan> a hierarchical structure should only have logarithmic overhead
03:28:12 <oklopol> so maybe less than you'd think?
03:28:25 <oklopol> algorithmically less i think.
03:28:42 <oerjan> while an array might need n^2 for n points
03:29:12 <oerjan> and also, may need to extend the array by copying
03:29:34 <Sgeo> Waiving the requirement for it to make sense might be more fun, actually
03:30:15 <oerjan> no, i don't think it's O(1) because adding 1 cell can add an entire new row and column, every time
03:30:25 <oklopol> Sgeo: wait sorry also i could actually tell you.
03:30:42 <oerjan> or, rather, *row _or_ column
03:31:23 <oklopol> Sgeo: the meaning is somewhat as follows, amortized O(f(n)) means on average the function takes O(f(n)) steps
03:31:39 <oklopol> but only somewhat, the actual definition i cannot really explain in human terms atm
03:32:01 <oerjan> Sgeo: amortized O(1) space means it could allocate in large blocks, but would still only allocate one new cell per step on average
03:32:29 <Sgeo> What's non-amortized?
03:32:43 <oklopol> oerjan: i think it's O(1) if you make the whole thingie double as big when you need one more in size
03:32:47 <oerjan> non-amortized would be an upper bound for every single step
03:33:19 <oerjan> oklopol: nope, a zig-zag diagonal pattern grows by n cells every step on average, no matter how you do it
03:34:02 <oklopol> oerjan: i guess that's somewhat true.
03:35:22 <oklopol> i mean, there's a subtle hint of truth in it.
03:37:00 <oklopol> Sgeo: actual definition of amortized cost of O(f(n)) is if you do n operations, you will not have done more stuff than a non-amortized O(f(n)) function would've done for the same n operations.
03:37:16 <oerjan> Sgeo: for example a good garbage collection algorithm should use time at most O(allocated space), but if it's not very good at being incremental it can still cause a large pause. so amortized time is good, but may still be too much non-amortized for real-time purposes.
03:37:44 <oerjan> *a large pause at a single allocation
03:39:13 <oklopol> this channel is too small for the two of us, it sometimes feels.
03:39:49 <oklopol> i should probably make it smaller, it's like size 18 because i used to use this computer laying on my armchair
03:39:52 <oerjan> oklopol: i just thought that if we both explain it, it will be easier to understand
03:39:59 <oklopol> which is more like a leathery bed than a chair
03:40:07 <oklopol> oerjan: ah, yes probably :P
03:41:22 <oerjan> courier 10 point, here
03:42:10 <oklopol> actually was just 16, but still pretty massive, especially as this is a tiny screen
03:42:12 <oerjan> i don't think i've changed it
03:42:44 <oklopol> i usually just change to courier, size doesn't matter
03:43:13 <oerjan> i maximized window height though
03:43:56 <oklopol> you mean you double clicked it so it's fullscreen?
03:44:14 <oklopol> top to bottom and size doesn't matter... where are all the penis jokes
03:44:15 <oerjan> i like to be able to see at least two windows at a time
03:45:44 <oklopol> let's /part this place and go eat some cod
03:46:08 <oerjan> the timing is somewhat wrong for getting cod here
03:47:11 <oerjan> maybe i should go to a fish restaurant tomorrow, except the only one i know of has just gone bankrupt
03:47:21 <oerjan> it's the middle of the night
03:47:24 <oklopol> what's wrong with your town
03:47:53 <oerjan> also, this is not that close to the center, i would need a taxi
03:48:38 <oklopol> i was supposed to say something clever about you know one place burning and another going bankrupt, but then i lost my brain of thought.
03:48:38 <oerjan> i'm sure i could get some food at 7/eleven if i did so, but i doubt they have cod :D
03:49:05 <oerjan> the bankruptcy is probably due to the crisis
03:49:14 <oerjan> the fires are a more long-term problem
03:49:19 <oklopol> some everywhereable fast food restaurant?
03:49:47 <oerjan> financial crisis. if you read or watched news you might have heard about it
03:50:00 <oklopol> well sure i have, but i didn't know it actually existed
03:50:38 <oerjan> hasn't hit norway hardest, but i still notice an unusual number of abandoned store fronts
03:51:27 <oerjan> i vaguely recall not finding 7/11 listed in finland when i checked
03:51:32 <oklopol> continuing my earlier message - got distracted - i just assumed mankind was having their period or something
03:52:16 <oerjan> in norway at least, they are about the only thing with a 24 hour open policy
03:52:26 <oklopol> and we don't have a financial crisis either. at least i don't. i don't believe in that sort of nonsense
03:53:54 <oklopol> we have tons of things with a 24hop
03:54:14 <oklopol> but they just serve food, no cod
03:54:26 <oerjan> i don't quite recall the english word for such small shops
03:54:47 <oerjan> they sell mostly snacks, some other food
03:55:16 <oerjan> i - don't recall hearing that word
03:58:56 <oerjan> i don't know the exact opening hour laws in norway, they seem to get gradually relaxed most of the time
03:59:33 <oerjan> they are stricter for big shops, and for anyone selling alcohol
04:00:01 <oerjan> while small shops selling mainly food can be open at any time
04:00:35 <oklopol> how about shops that just buy alcohol?
04:01:00 <oerjan> i'll try and remember if i see any :D
04:01:52 <oklopol> "hello do you buy alcohol here?"
04:02:09 <pikhq> In the US, 7/11 is just one of many places that are open 24/7...
04:02:30 <oerjan> i suppose their main problem would be that only a few people could legally sell to them :D
04:02:35 <pikhq> Freaking Walmart tends to be open 24/7. Y'know, that land of everything that's cheap crap.
04:03:07 <oklopol> pikhq: i love your country :|
04:03:47 <oerjan> big stores must be closed at least on sundays here
04:04:20 <pikhq> And, since it's the US, it's got more parking lot than building.
04:04:25 <oklopol> oerjan: maybe people can sell alcohol in stores that can themselves sell alcohol, you know, if it's like an area of effect thing.
04:05:18 <oklopol> pikhq: well, i like throwing balls on parking lots
04:05:49 <oklopol> well, i mean those indoors ones.
04:06:22 <oerjan> trondheim does have some mall-like places with lots of parking in the city outskirts. (as well as some smaller ones with no parking in the centre.)
04:07:02 <oerjan> (ok not _entirely_ no parking, but still...)
04:07:46 <pikhq> Mall? No, just a department store.
04:07:51 <pikhq> One that's larger than most malls.
04:07:59 <oerjan> i'm not quite sure on the distinction
04:08:30 <oerjan> there are department stores too, like Ikea
04:08:48 <oerjan> er, if that word means what i think
04:09:01 <oklopol> it does if it means what i think.
04:09:42 <oklopol> google definitely doesn't know what i think it is
04:10:16 <MizardX> Do you know what you think it is?
04:10:22 <pikhq> Walmarts tend to have about 10,000 square meters of floor space.
04:11:26 <oerjan> now if i had a sense of size, i might be able to tell how big 10,000 square meters is, but i don't.
04:11:36 <oklopol> oerjan: take its square root
04:11:49 <pikhq> Oh, sorry, that's just a "Walmart Discount Store". Most of them are Walmart Supercenters, which average *18,000* square meters of floor space...
04:12:00 <pikhq> Those also contain a full supermarket.
04:12:04 <oerjan> now i just need to know how long 100 meters are. whoops.
04:13:54 <oklopol> oerjan: like three blue whales
04:15:37 <oklopol> i'd be so big and i could go so deep you know.
04:15:39 <oerjan> i see Obs! listed under wp's hypermarket article, that's the closest such place from here
04:16:19 -!- coppro has joined.
04:19:24 * pikhq observes that those appear more akin to malls...
04:22:06 <pikhq> Oh, Wikipedia's being dumb.
04:22:11 <oerjan> they have sort of grown, the one close to here was expanded a lot and now includes many shops besides the main Coop Obs! chain
04:22:16 <pikhq> It shows links to malls *containing* an Obs!.
04:22:49 <oerjan> i guess that's what they are
04:23:26 <pikhq> http://en.wikipedia.org/wiki/City_Syd Take a floor of that. That's a small Walmart store.
04:23:39 <oerjan> I'm talking about City Lade btw, didn't know it had its own article
04:23:47 <oerjan> and City Syd is at the other end of town
04:24:18 <Sgeo> http://tr.froup.com/tr.pl?1604
04:24:19 <oerjan> (the first one is within walking distance from where i live)
04:26:43 -!- pikhq has quit ("Off I go to the non-Internet.").
04:30:00 <Sgeo> They're noodles! You don't compress and encrypt them, you eat them!
04:44:12 <Sgeo> http://tr.froup.com/tr.pl?1653
04:44:35 <Sgeo> "The cardinality of the set of counties we're banned from is greater than the cardinality of the set of counties"
04:45:23 <coppro> looks too much like CVS
04:47:14 <coppro> http://www.circleversussquare..com
04:49:21 <oerjan> you mean, CVS looks too much like T&R.
04:53:36 <coppro> but there's no dates on T&R so I can't tell
04:53:50 <coppro> also I read CVS first, so it wins
04:55:04 <oerjan> basically, i read T&R a year ago, so it's at least that old
04:56:09 <oerjan> "Triangle and Robert is a webcomic by Patrick Shaughnessy that ran from August 1999 to September 2007."
04:58:05 <oerjan> "Triangle and Robert won Best Minimalist Comic in the 2001 Web Cartoonist's Choice Awards.
05:09:39 <Sgeo> "We need wackiness." "Robert, we're already designing a system to launch HAM into ORBIT with a RAILGUN."
05:11:53 <Sgeo> "We need a heat-resistant ham glaze, not the destruction of a six-mile radius." "It was not a radius. The damage was limited to a single twelve-mile line segment." "What's this now? I just meant in general." "Then mind not."
05:21:34 <Sgeo> http://tr.froup.com/tr.pl?1783 has a bit of an overview (spoilers, obviously)
05:23:40 <Sgeo> Well, not really
05:29:02 <Sgeo> "Oh, we're escaping? How do I help?" "We're escaping FROM you."
05:39:28 -!- bsmntbombdood has quit (Read error: 110 (Connection timed out)).
05:40:17 -!- bsmntbombdood has joined.
05:47:59 -!- Sgeo has quit ("Leaving").
06:30:30 -!- oerjan has quit ("leaving").
07:23:08 -!- bsmntbombdood has quit (Read error: 110 (Connection timed out)).
07:50:49 -!- bsmntbombdood has joined.
07:58:04 -!- olsner has quit ("Leaving").
07:58:20 -!- olsner has joined.
07:59:59 -!- clog has quit (ended).
08:00:00 -!- clog has joined.
08:17:32 -!- coppro has quit (Read error: 110 (Connection timed out)).
08:17:41 -!- tombom has joined.
09:25:38 -!- olsner has quit ("Leaving").
10:49:54 -!- neldoreth has quit ("Lost terminal").
10:52:36 -!- neldoreth has joined.
11:00:24 -!- GreaseMonkey has quit ("YES -> thor-ainor.it <- THIS IS *DELICIOUS*!").
12:31:58 -!- Judofyr has joined.
12:48:25 -!- sebbu3 has changed nick to sebbu.
13:10:47 -!- FireyFly has joined.
13:21:00 -!- FireyFly has changed nick to FireFly.
13:47:04 -!- Gracenotes has quit ("Leaving").
14:25:48 -!- BeholdMyGlory has joined.
14:47:53 -!- ais523 has joined.
14:51:51 <ehird> 16:28:26 <ais523> many people have alias rm = rm -I
14:51:53 <ehird> [ehird:~] % rm -rf *
14:51:53 <ehird> zsh: sure you want to delete all the files in /Users/ehird [yn]?
14:52:28 <ais523> zsh has a builtin version of the alias?
14:52:38 <ehird> ais523: it just specialcases *, i think
14:52:56 <ehird> 16:34:38 <ais523> on a disk that large, you'll have to spend days fscking if you have a power cut
14:52:56 <ehird> 16:34:44 <ais523> if you don't have journaling
14:53:03 <ehird> fsck takes days anyway when ubuntu decides to run it because WHY NOT
14:53:25 <ehird> 16:35:09 <pikhq> Why I have losslessly compressed copies of Big Buck Bunny & Elephants Dream, I'm not quite sure.
14:53:26 <ais523> I love the arguments that the makers of one particular version of rm went to to try to prove it was UNIX-compliant
14:53:33 <ais523> they wanted to specially handle rm -rf /
14:53:36 <ehird> pikhq: here, have a 200gb uncompressed movie: http://www.sitasingstheblues.com/wiki/index.php?title=SitaSites
14:53:59 <ais523> in the end, they successfully argued that they were allowed to treat it differently because rm is allowed to treat . and .. differently, and rm -rf / necessarily deletes .
14:54:31 <ehird> 16:35:50 <ais523> oh, yes, there are ext2 drivers for Windows, but not ext3 drivers
14:54:37 <ais523> (I think that argument might be slightly bogus; if you have two nested chroots, it's possible to be in a directory outside /)
14:54:45 <ehird> http://www.fs-driver.org/ handles ext3
14:54:47 <ais523> ehird: they have full ext3 drivers now, which work with journaling?
14:54:53 <ehird> Linux Ext3 volumes can also be accessed. To do that, please read the FAQ section.
14:55:03 <ehird> i guess it writes w/ only ext2 features
14:55:16 <ehird> but the only use for those drivers is reading, really
14:55:24 <ais523> <FAQ> "The Ext2 file system driver of the Ext2 IFS software will refuse mounting an Ext3 file system which contains data in its journal, just like older Linux kernels which have no Ext3 support. In this way data loss and damaging the file system is avoided when the journal is subsequently replayed. So you can access only those Ext3 volumes with the Ext2 IFS software which have been cleanly dismounted beforehand."
14:55:38 <ais523> it can only read an ext3 drive which happens to be instantaneously legal ext2
14:55:42 <ais523> i.e. has an empty journal
14:55:52 <Deewiant> I.e. it's essentially not ext3. :-P
14:55:53 <ehird> is ext4's on-disk format stable?
14:55:58 <fizzie> There is a RM_STAR_SILENT option you can set for zsh to disable that speciality.
14:56:04 <ehird> since it's in the kernel
14:56:18 <ehird> Date resolutionNanosecond
14:56:24 <ehird> now THAT'S excessive!
14:56:57 <ais523> imagine a compile using make on a really fast computer
14:57:05 <fizzie> Btrfs is in the kernel and doesn't have a stable disk format. At least I think the kernel-config help said so. It was very much "I'm so experimental I will eat all your data *munch* *munch*" warningsy.
14:57:12 <ehird> ais523: define really fast
14:57:26 <ais523> fast enough that microsecond timestamps aren't good enough to see if something's been compiled or not
14:57:28 <Deewiant> ehird: Fast enough that nanoseconds matter.
14:57:52 <ais523> I can imagine compile steps happening in less than a microsecond
14:58:03 <ais523> some of the make rules in the C-INTERCAL build are just symlinking
14:58:13 <ehird> ais523: with a faster compiler than gcc you could prolly do it for a c project
14:58:16 <ehird> just a very dumb compiler
14:58:25 <ais523> and if you have an in-memory filesystem, I can imagine the creation of a symlink in less than a microsecond
14:58:28 <ehird> could run more or less instantly
14:58:46 <ehird> also, just use /tmp with tmpfs, nice and simple
14:58:49 <ehird> oh, you'd need a fast make too
14:58:59 <ehird> but I think you could do it, on really high end machines, today
14:59:28 <Deewiant> Compiling two files in less than a microsecond? I doubt it.
14:59:52 <ehird> <ais523> the ext2 standard has a flag saying "don't try to mount as ext2"
15:00:07 <Deewiant> I guess that's enabled for ext4.
15:00:14 <ehird> <meta http-equiv=content-type content=text/plain>
15:00:26 <Deewiant> [ 3.529311] EXT2-fs: sda3: couldn't mount because of unsupported optional features (40).
15:00:29 <ehird> 16:38:22 <AnMaster> how can anyone be insane enough to use that yet for anything important...
15:00:30 <ais523> #!/bin/cat at the top of a perl program
15:00:55 <ehird> This proposal was accepted, and on June 28, 2006 Theodore Ts'o, the ext3 maintainer, announced the new plan of development for ext4.[3] A preliminary development snapshot of ext4 was included in version 2.6.19 of the Linux kernel. On Oct 11, 2008, the patches that mark ext4 as stable code were merged in the Linux 2.6.28 source code repositories,[4] denoting the end of the development phase and recommending ext4 adoption. Kernel 2.6.28, containing the ext
15:00:57 <ehird> 4 filesystem, was finally released on December 25, 2008.[5]
15:01:04 <fizzie> Software-suspend used to have Very Big Warnings back when it was experimental, too. Nowadays there's just a rather sedate warning about not touching the filesystems between suspend and resume.
15:01:06 <ais523> well, it's not surprising you couldn't mount it as ext3, if you have at least one extent on the system, and you probably do
15:01:31 <Deewiant> I was just pondering that "don't try to mount as ext2" flag.
15:01:37 <ehird> fizzie: does linux do hybrid suspend/hibernate?
15:01:41 <ais523> put there for forwards compatibility, obviously
15:01:48 <ehird> we turn all the fans off, go into suspendy mode, and gradually write it all to disk
15:01:53 <ehird> then kill ourselves
15:01:56 <ehird> os x does that, it works nicely
15:02:05 <ais523> some people like it, especially for a desktop
15:02:06 <ehird> fast suspend/unsuspend with less power usage
15:02:08 <fizzie> The "s2both" does pretty much that.
15:02:10 <Deewiant> ais523: Yes, but my dmesg indicates that it's either not set for ext4 or that the ext2 driver looks at something else before that flag.
15:02:20 <ehird> 16:39:09 <AnMaster> coppro, the online defrag tool is _still_ missing for ext4
15:02:21 <ehird> 16:39:20 <AnMaster> I'm going to stay to ext3 and xfs for the next few years.
15:02:26 <Deewiant> Since it didn't say "couldn't mount because the flag says not to"
15:02:33 <ais523> for a laptop you have the problem of lots of hard drive activity immediately after suspend
15:02:44 <ais523> which is not good if you want to carry the laptop around and suspended it because of that
15:02:48 <Deewiant> Or maybe that was the flag, the message just wasn't obvious about it.
15:03:01 <ais523> I was really scared when I carried my laptop across campus in the middle of a distro upgrade
15:03:18 <ais523> it was even fun when I REISUBed in the middle of the upgrade, due to a hardware problem
15:03:33 <ehird> I have no idea how i'll install ubuntu with my really goddamn complex partition setup
15:03:35 <fizzie> I use s2ram nowadays with my desktop, since the boot-up from real hibernation is a bit on the slow side; I guess I could use s2both instead, but I'm not so very worried about a power-failure causing horrible complications.
15:03:48 <ehird> Can you do partitioning in the livecd, and then just tell the installer "Um, just use this fstab"?
15:04:00 <ais523> I'm not sure, quite possibly
15:04:10 <ais523> you might need to use the alternate install disk, that lets you do all sorts of weird things
15:04:16 <ehird> not a separate disk
15:04:21 <ehird> just a different install option nowadays
15:04:27 <ais523> there are two CDs to install from, the main one and the alternate one
15:04:41 <ais523> the alternate ones contain the stuff for doing weird stuff at installation, but fewer packages, IRC
15:04:41 <ehird> the ubuntu livecd has an alt install option iirc
15:04:47 <ehird> I want full packages
15:04:53 <ais523> you get them from the Internet later
15:04:55 <ais523> they just aren't on the CD
15:05:14 <Ilari> Does XFS still have that truly braindead kill-file-contents-on-crash misfeature? Does it get move-over-replacement right even w.r.t. system crashes?
15:05:29 <ais523> ah, no, it's the same packages
15:05:34 <ais523> it has a text-based installer rather than a graphical one
15:05:38 <ais523> which can handle more weird cases
15:05:53 <ais523> (the reason it's text-based is in case the installer can't get the graphics going without manual intervention, it seems)
15:05:56 <ehird> ais523: it's just that my current planned setup is that /home is a unionfs where .anything goes to /aux/ehird/dotfiles and everything else is on the main HD, except /home/ehird/local is binded to /aux/ehird/local
15:06:02 <ehird> not the simplest setup
15:06:08 <fizzie> Debian-installer lets one configure horribly complicated raid/LVM setups; but I've only installed Ubuntu long ago with a very standard "here, use this partition" fashion, so I don't know about the partitioning.
15:06:14 <ehird> maybe /aux should be /var/home
15:06:18 <ehird> in analogy to /var/tmp
15:06:36 <ais523> if you're only messing inside /home, then couldn't you install, then mess with the fstab by hand afterwards
15:06:43 <fizzie> Unionfs is not in the mainline kernel, so configuring that "correctly" during the installation sounds tricky. But certainly you can mess it up later.
15:06:44 <ais523> then delete all the old dotfiles?
15:06:50 <ais523> wiping the dotfiles isn't a problem on a clean install
15:07:45 <ehird> ais523: well, I suppose. I have other odd setup thingies though.
15:07:52 <ehird> Like /tmp as tempfs.
15:07:59 * ais523 laughs at Wubi's advertising: 1. Select a new password. 2. Click install. 3. There is no three.
15:07:59 <ehird> I'd rather get it right first time than clean it up later
15:08:15 <ehird> fizzie: ubuntu has unionfs
15:08:17 <fizzie> And /tmp is the one place you can safely clean up even post-installation. :p
15:08:19 <ais523> I think Wubi's good for its target audience
15:08:22 <ehird> but on the subject of Wubi
15:08:27 <ehird> every time I turn someone to ubuntu
15:08:29 <fizzie> I guess the live-cd would, at the very least, yes.
15:08:35 <ehird> I have to tell them to ignore the Ubuntu livecd windows-autostart
15:08:38 <ehird> which advertises wubi
15:08:44 <ehird> as "install without partitioning" or sth
15:08:55 <ehird> then I get to explain why they shouldn't use it, etc etc etc
15:09:02 <ehird> and it's a bloody load of fuss for an unuseful feature
15:09:03 <fizzie> They really should get union-mounting done properly; there's apparently been talk on doing it on the VFS level (like mount --bind is done), which (as far as I could figure out) was the reason aufs2 was very steadfastly rejected for kernel-inclusion.
15:10:00 <ais523> the feature = almost impossible to get wrong even if you don't know what you're doing
15:10:06 <fizzie> I think good old Slackware, in ages past, had a "install into a file on a FAT filesystem, which is then loop-mounted as ext2" option.
15:10:23 <ais523> creates a subdirectory under C:\ that's the Ubuntu filesystem
15:10:23 <ehird> fizzie: installs ubuntu by making the filesystem a folder in the ntfs windows partition
15:10:30 <ehird> it's slow and ties you to windows.
15:10:34 <ais523> and tells Windows it's a Windows application
15:10:39 <ehird> ais523: so is the livecd installer, as far as impossible to get wrong goes
15:10:50 <ehird> my instructions to the technically illiterate are "click next and read the text"
15:10:52 <ais523> ehird: not so, you can accidentally wipe the Windows partition like that
15:10:55 <fizzie> Does it use separate files under NTFS, then?
15:11:06 <ais523> you end up with an NTFS Ubuntu tree, I think
15:11:10 <ehird> ais523: oh, in the most recent case it was a new laptop
15:11:14 <ehird> so Vista was unwanted
15:11:20 <ehird> so the wiping was not a problem
15:11:23 <ais523> oh, over here people generally dual-boot
15:11:43 <ais523> I've actually got relatively good at securing Vista
15:11:45 <fizzie> At least Slackware had a good excuse for using a loop-mounted ext2 image; can't see that thing working very well on a FAT filesystem.
15:11:54 <ais523> as in, worse than people who know what they're doing, but better than people who don't
15:11:57 <ehird> I think ubuntu fills all this guy's needs, prolly
15:12:10 <ehird> ais523: what do people use the windows partition for?
15:12:27 <ais523> also, it's used as a sandbox
15:12:34 <ais523> in case the Linux side catches a virus
15:12:40 <ais523> only the Linux side connects to the Internet
15:12:46 <ais523> the Windows doesn't, because it's less secure
15:12:51 <ais523> and all the important stuff's stored there
15:13:35 <fizzie> Out of curiosity... how does it do device files on an NTFS file system? (Although I guess it wouldn't really need to, much, with udev-handled tmpfs /dev. Still.)
15:13:44 <ehird> ais523: so, no shitty university stuff that needs windows?
15:13:54 <ehird> that's good, since he got it for going to uni with
15:14:50 <fizzie> We have around here a couple of horribly horrible administrative webblications that are strictly IE-only; but since most of the workstations around here are Linux systems, there's a Windows Terminal Server box you can "rdesktop" in to use those.
15:15:31 <ais523> ehird: there isn't exactly
15:15:43 <ais523> there's WebCT, which is truly awful, but it runs just as badly in Windows as it does in Linux
15:15:50 <ehird> ais523: apart from the windows-based policy things? like antiviru
15:15:50 <ais523> being a rather badly written webapp
15:16:01 <ais523> ehird: they didn't say "windows-based", they just said antivirus
15:16:08 <ehird> ais523: windows-based as in without thought of other OSes
15:16:15 <ehird> all sane people will summarily ignore the clause, ofc
15:16:25 <ais523> oh, I meet both the letter and the spirit of the rules
15:30:31 <ehird> "Frequent television portrayals of Christians as absurd make it more difficult for believers to defend themselves, a national journalist has said. "
15:30:31 <ehird> http://www.christian.org.uk/news/20090420/tv-sends-message-that-christians-are-nutters/
15:30:34 <ehird> Bahahahahahahahaha
15:31:46 <ehird> hahaha, another article on there: "BBC portrays pro-lifers as murderous terrorists"
15:32:03 -!- oerjan has joined.
15:34:02 <ehird> Have you ever thought Tetris(R) was evil because it wouldn’t send you that straight “I” brick you needed in order to clear four rows at the same time? Well Tetris(R) probably isn’t evil, but Bastet certainly is. >:-) Bastet stands for “bastard tetris”, and is a simple ncurses-based Tetris(R) clone for Linux. Unlike normal Tetris(R), however, Bastet does not choose your next brick at random. Instead, Bastet uses a special algorithm designed to
15:34:05 <ehird> choose the worst brick possible. As you can imagine, playing Bastet can be a very frustrating experience! — http://fph.altervista.org/prog/bastet.shtml
15:34:25 <ehird> Ye olde, like your face.
15:35:05 <ehird> http://iseverythingonlineyet.com/
15:35:15 <ais523> does it say yes or no?
15:35:29 <ehird> ais523: it says more than either.
15:35:39 <ehird> it has information, see.
15:35:44 <ehird> sites used to do that
15:36:00 <ehird> it works fine in w3m.
15:36:16 <ais523> does that work in w3m, I wonder?
15:36:16 <ehird> because my browser doesn't render them
15:36:25 <Deewiant> I looked at the source, though.
15:36:26 <ais523> ehird: isn't that standards-incompliant?
15:36:36 <ais523> people are going to have to resort to javascript blink to be portable, now!
15:36:44 <ehird> it's not incompliant
15:36:50 <ehird> browsers can render elements however they want in html4
15:37:05 <ais523> so they render blink as not blinking
15:37:22 <ais523> someone should make a browser which renders everything except blink as blinking
15:37:25 <ehird> ais523: how can you blink text-to-speech?
15:37:47 <Deewiant> So you can do whatever you want with it.
15:37:54 <oerjan> ehird: i DON'T see WHY that SHOULD not BE possible
15:38:03 <ais523> ehird: I've actually blinked speech before now
15:38:10 <ais523> by putting a hand over my ear and moving it back and forth quickly
15:38:11 <Deewiant> However, CSS has 'text-decoration: blink'
15:38:19 <ais523> in a hubbub of lots of people talking, you hear random syllables
15:38:19 <Deewiant> Which is actually valid CSS 2.1, not sure about 3
15:38:24 <ais523> they don't make any sense, but it's kind-of soothing
15:38:35 <Deewiant> But the UACG say that you're allowed to ignore it.
15:38:43 <ehird> "Go read the comments, as usual they are the best part of Slashdot." ← HI 2005, HOW YOU DOING?
15:39:15 <ais523> the comments are the best part, though
15:39:18 <ehird> AnMaster: you're 2005? figures
15:39:20 <ais523> because reddit gets the articles first
15:39:24 <ais523> and the summaries are awful
15:39:34 <ehird> 15:38 ehird: "Go read the comments, as usual they are the best part of Slashdot." ← HI 2005, HOW YOU DOING?
15:39:34 <ehird> 15:39 AnMaster: hi
15:39:59 <AnMaster> I just got back, rather tired.
16:04:37 -!- Slereah has joined.
16:13:47 <oklopol> for bastet, can't you change the keys
16:15:31 -!- Slereah_ has quit (Read error: 110 (Connection timed out)).
16:15:50 <oklopol> blah i'll just use some other computer
16:17:36 <ais523> ehird: how did your Enigma 1.1ing go?
16:17:42 <ehird> ais523: didn't fix it.
16:17:47 <ehird> why isn't there a binary ;_;
16:17:53 <ais523> because it's still alpha
16:17:57 <ais523> and they don't build nightlies
16:18:19 <ais523> I went and translated a puzzle from Agora-Enigma into computer-game-Enigma
16:20:44 <ehird> 0% [Waiting for headers]
16:20:48 <ehird> ais523: fix this shit.
16:28:05 <Deewiant> I wonder which version I've played
16:28:47 <Deewiant> 0.37 at least is a pain-in-the-ass version
16:28:51 <ehird> [[A wise man once said : "in order to understand recursion, you must first understand recrusion". ]]
16:28:55 <Deewiant> Where it's hard to get even one line
16:29:00 <ehird> And to understand recrusion, ...
16:29:07 <ais523> ehird: I've seen that quote in lots of places before
16:29:11 <ehird> ais523: look carefully
16:29:36 <ehird> ...which ruins the joke, but can be used to make it into a joke about mutual recursion
16:30:41 <oklokol> anyway seems level is just speed
16:31:14 <oklokol> yes, but it's a bit annoying you can't slide things under others after they hit ground, on level 9, by just holding the key
16:31:56 <ais523> is this a tetris clone?
16:32:51 <ehird> ais523: it uses an algorithm
16:32:55 <ehird> to pick the worst piece possible
16:33:01 <ehird> thus, bastard-tetris
16:33:15 <Deewiant> 0.43 makes me feel like I'm just getting really unlucky with the RNG
16:34:00 <Deewiant> Occasionally it gives me exactly the piece I want
16:34:26 <oklokol> 43 is supposedly even harder
16:35:31 <Deewiant> (Playing the harder version, of course)
16:36:34 <Deewiant> Hmmh, messed up a bit, ended with 9 lines
16:37:03 <Deewiant> Anyway, with 0.41 I typically get my first line when I've filled half the screen or so
16:37:28 <Deewiant> With 0.43 I can consistently get a line before I've got even 10 blocks
16:39:14 <oklokol> but i was really unlucky! :D
16:39:31 <Deewiant> No, you just weren't clever ;-P
16:39:38 <Deewiant> I think the only random thing is the first piece
16:39:52 <Deewiant> Or no, of course if multiple pieces have the same probability
16:40:07 <Deewiant> Unlike normal Tetris(R), however, Bastet does not choose your next brick at random. Instead, it uses a special algorithm designed to choose the worst brick possible.
16:40:22 <ehird> what version is xorg 1.6?
16:40:50 <Deewiant> oklokol: So... what exactly should I read
16:40:58 -!- KingOfKarlsruhe has joined.
16:41:08 <ehird> I have 1.3, 1.4, 1.4.99, 1.5, 6.7, 6.8, 6.8.99, 6.9, 7.0, 7.1, 7.1.99, 7.2, 7.3, 7.4 as options
16:41:22 <ehird> so which would that be for xorg 1.6? I'm puzzled by 7.x, this xorg is recent?
16:42:11 <ais523> I just did Test of Dexterity
16:42:32 <oklokol> read linux source and umm 0.41
16:43:19 <Deewiant> Should just be 100,100,100,100...
16:43:41 <oklokol> but it sure as hell doesn't seem that way, i just got like 7 2x2's in a row, anything else would've done the trick.
16:44:13 <oklokol> also it's hard to lose the tetris reflexes, "i'll just put this here and wait for a 4x1"
16:44:35 <Deewiant> You have to intentionally mess up to confuse it, I find
16:45:30 <ais523> isn't 7 2x2s enough ot make a row?
16:45:44 <Deewiant> Well, if you put them sideways it is
16:45:51 <Deewiant> Of course it wouldn't have given you the last one if you had :-P
16:45:53 <oklokol> ais523: yes, unless your keeping the hole open just in case...
16:48:26 <ehird> oklokol YOUR/YOU'RE ERROR
16:48:45 <oerjan> GOLD WITH PINK SPOTS ALERT
16:51:20 <Deewiant> I haven't played that much :-P
16:52:00 <Deewiant> I've played ~10 games altogether also
16:52:23 <oklokol> then it's probably more about luck, most of my attempts were zeroes, and i started with a 6.
16:52:25 <Deewiant> I can't remember my old records but I think they didn't go up to 10, might have exceeded 5
16:54:38 <ehird> Deewiant: fuck you now X won't start :|
16:55:02 <oklokol> why are configurations in source, i'm not gonna compile this to try a deterministic game.
16:55:37 <Deewiant> ehird: extra/xorg-server 1.6.1-1; extra/xorg-server-utils 7.4-4
16:55:47 <oklokol> well. computer probably always wins, in tetris.
16:56:11 <oklokol> (thinking of it as a 2 player game, computer being the one choosing pieces)
16:56:18 <ehird> Deewiant: how can I finds out my servers versions
16:56:21 <ais523> how well would an AI do at playing bastet?
16:56:32 <oklokol> ais523: that's essentially what i'm wondering.
16:56:59 <ais523> make an evolutionary Befunge-93 program to test it out
16:57:05 <ehird> Deewiant: that doesn't tell me the 7.butt thing
16:57:14 <Deewiant> ehird: What is the 7.butt thing anyway
16:57:29 <Deewiant> It's not the server version, it's something else
16:57:31 <oklokol> oerjan: what'a tetris's complexity?
16:57:35 <ehird> Deewiant: "xorg-server-utils" apparently.
16:57:58 <Deewiant> ehird: I think it's just a sort of package version number for all their crap
16:58:01 <Deewiant> ehird: http://www.x.org/wiki/Releases/7.4
16:58:17 <ehird> I don't know how to find out what version I have!
16:58:29 <Deewiant> ehird: There is no such version.
16:58:40 <Deewiant> I.e. no single program has such a version.
16:58:44 <Deewiant> Or that's what it seems like to me.
16:58:56 <Deewiant> Looking at http://www.x.org/wiki/Releases/7.4.
16:58:56 <ehird> I dont' care about program
16:59:04 <ehird> I just want to know what 7.butt I have, I guess apt knows
16:59:21 <Deewiant> If your stuff is from a package manager then yes, of course ask the package manager and not me
16:59:35 <ehird> Must be kernel issues. it mentioned ABI
16:59:40 <Deewiant> ehird: http://www.x.org/wiki/Releases/7.4 is 1.5.1.
17:01:40 <GregorR> The Xorg releases are versioned, but each release is just a bundle of packages. The packages each have their own versioning scheme, so it's entirely likely that a "7.x" version a package manager reports to you is nigh-on meaningless.
17:02:03 <ehird> GregorR: Well, this installer likes to know.
17:02:18 <GregorR> Deewiant: Human being who installs stuff :P
17:02:25 <ehird> Parallels VM tools thing.
17:02:33 <ehird> So I cans get my compiz's on my VM's.
17:02:34 <GregorR> OH, an actual package installer.
17:02:43 <ehird> It also has 1.x versions.
17:02:46 <ehird> But only up to 1.5.
17:02:54 <ehird> Deewiant: It makes the windows marginally less ugly. I like shadows.
17:02:58 <ehird> And cuz I have 1.6
17:04:28 <ehird> I'll try the 1.5 thang tho.
17:05:07 <ehird> module ABI major version (4) doesn't match the server's version (5)
17:05:09 <ehird> module ABI major version (2) doesn't match the server's version (5)
17:05:13 <ehird> Deewiant: so 1.5 prolly won't work.
17:05:49 <GregorR> Just hexedit it and change the ABI version 8-D
17:05:59 <ais523> GregorR: that so wouldn't work
17:06:56 <ehird> Deewiant: How, I'm on a binary distro.
17:07:54 <Deewiant> Run it on every binary that matches the glob *[xX]* just in case
17:08:17 <ehird> Deewiant: it stores the abi version as a string?
17:08:22 <ehird> using the decimal digits?
17:08:38 <ehird> in case they overflow int or sth?
17:08:51 <Deewiant> All ints are stored as text, duh?
17:09:28 <Deewiant> How else could the computer read them?
17:14:52 -!- KingOfKarlsruhe has quit (Remote closed the connection).
17:16:32 <oerjan> oklokol: i don't remember tetris' complexity but iirc i have read that you cannot win against a stream of equal zigzag pieces
17:17:50 <Slereah> Yes, but that is quite unlikely
17:18:25 <ais523> you could put them each standing upright
17:18:30 <ais523> that gives you a row in the centre out of three
17:18:38 <ais523> then the next row standing upright nets you two lines
17:18:41 <Slereah> ais523 : That leaves you holes
17:18:43 <ais523> and likewise for all futue
17:18:56 <ais523> that leaves holes in the bottom row, but you keep on getting lines 2 and 3 ad infinitum
17:18:58 <Slereah> Eventually they'll accumulates
17:18:59 <ais523> and it never grows any higher
17:19:28 <oklokol> ais523: what if there's an odd amount of columns?
17:19:35 <Deewiant> Quick, somebody code a version of tetris that only gives a Z and test it out!
17:19:40 <ais523> ah, it probably doesn't work then
17:19:42 <Deewiant> Use VB to make a GUI interface
17:19:54 <ais523> if anything, I'd modify quinquis
17:20:02 <ais523> even though I'd have to find a 16-bit DOS compiler from somewhere
17:20:11 <ais523> quinquis is a tetris clone using pentominoes I wrote years ago
17:20:21 <ehird> ais523: You have to use VB to make a GUI interface. How else will you track an IP?
17:20:32 * ais523 deliberately misses the point
17:20:40 <ehird> That's not a GUI interface!
17:20:55 <Deewiant> Traceroute has no GUI, therefore it cannot trace an IP.
17:21:18 * ais523 System | Administration | Network Tools | Traceroute
17:22:17 <ehird> ais523: is it written in visual basic?
17:22:39 <ais523> I don't know, do you want me to apt-get source it and look?
17:22:46 <oerjan> ais523: hmph, you are right, it must have been more complicated
17:23:04 <ais523> in theory it could be written in Mono
17:23:11 <ais523> except I think I don't have Mono installed here atm
17:23:33 <ehird> gnome ships with it
17:23:36 <ehird> well, depends on it
17:23:49 <ais523> ah, yes, I have tomboy, therefore at least bits of mono
17:23:58 <ais523> I know that the Jaunty update uninstalled most of it though as an unused dependency
17:24:05 <ais523> so I probably only have bits of the core, rather than all the libraries
17:24:34 <oklokol> oerjan: you checked both odd and even?
17:25:08 <oerjan> ais523: ah, http://www.geom.uiuc.edu/java/tetris/explanation.html
17:25:10 <oklokol> for even it would clearly work, consider a 2-column game
17:25:22 <oerjan> oklokol: i only cared about normal size
17:25:41 <oerjan> and that link should work
17:25:46 <oklokol> and you'll realize it works for all sizes 2n.
17:25:53 <ehird> "However, it might be theoretically possible to create a perfect computer Tetris player which could successfully play Tetris until its plug was pulled. The arguments presented on this page prove that that is impossible."
17:25:56 <ais523> oh, alternating Z and S
17:25:58 <ehird> It might be possible. It's not possible.
17:26:16 <ais523> my proof that all-Z isn't lethal brakes down if they alternate
17:26:19 <Deewiant> Alternating is impossible, of course.
17:26:50 <Deewiant> Or hmm, I guess that works for even columns also?
17:26:55 <ais523> besides, I've won Tetris before
17:27:11 <oklokol> oerjan: you checked odd too?
17:27:20 <ais523> the gameboy version had a victory condition
17:27:50 <ais523> I got gameboy tetris free with a bank account
17:27:55 <oerjan> oklokol: it's my proof
17:28:00 <oklokol> i have one that plays a cool victory tune when you reach 10000 points, takes about 2 hours, i remember getting to 3h or something
17:28:19 <oklokol> 19:22… oerjan: ais523: hmph, you are right, it must have been more complicated
17:30:26 -!- oerjan has quit ("Later").
17:31:56 <oklokol> a simple yes or no would've sufficed.
17:32:18 <oklokol> i'm gonna guess "i don't remember, i was too drunk"
17:34:59 -!- pikhq has joined.
17:40:12 * Deewiant has played http://www.youtuberepeat.com/watch/?v=zXeCEzaNLKM for over 10 minutes now
17:40:16 <ais523> that's pretty nasty of you ,take it back!
17:40:34 <Deewiant> Okay, I do, this music is pretty cool actually, hehe
17:41:53 <ehird> ais523: there are far worse insults.
17:42:00 <ehird> 'damn you' is... pretty tame.
17:42:07 <ais523> not really, it has nasty religious overtones
17:42:20 <ehird> Eh, religious people say damn you all the time.
17:42:23 <ehird> There's plenty of history to anything.
17:42:33 <ais523> it's expressing a preference for the worst possible thing (by definition) to happen to someone
17:42:45 <ais523> if you're unreligious, you believe that that's impossible, but you still know what the insult means
17:42:48 <Deewiant> So "fuck you" should be taken as a compliment?
17:43:04 <ais523> I never really understood why "fuck you" was an insult
17:51:47 <ehird> "The Right Honourable the Lords Spiritual and Temporal of the United Kingdom of Great Britain and Northern Ireland in Parliament assembled "
17:52:02 <ehird> it's like something alesteir crowley [sp] would write.
18:03:57 <oklokol> ais523: because masturbation is pathetic?
18:04:54 <oklokol> also umm, this code here, it can set the value of a pointer in two ways, one is to use main's argument, other is to make an array in an if block, and assign the pointer to it.
18:05:19 <oklokol> isn't that a trivial C error?
18:08:15 -!- bsmntbombdood has quit (Read error: 110 (Connection timed out)).
18:08:43 <ais523> how is the array made?
18:08:47 <ais523> automatic storage duration?
18:08:53 <ais523> if so, you can't use the pointer outside the block
18:12:59 <AnMaster> ais523, C question. Is return stack->entries[--stack->top]; well defined?
18:13:10 <AnMaster> as far as I can tell it does what I intended.
18:13:38 <ais523> there's no guarantee as to whether the first stack will return the decremented or original version
18:14:05 <oklokol> AnMaster: isn't C like your primary language?
18:14:34 * oklokol is annoyed at lecturers insisting on C if they can't get it right
18:15:32 <Deewiant> ais523: That decrements stack->top and accesses stack->entries so no problems?
18:15:56 <ais523> Deewiant: is that (--stack)->top or --(stack->top)?
18:16:02 <ais523> I think it's the first, but I'm not sure
18:16:06 <ais523> unary over binary, isn't it?
18:16:08 <Deewiant> ais523: I'm quite sure it's the latter
18:16:21 <oklokol> actually yeah probably for those . and -> things it's like that
18:16:35 <Deewiant> prefix decrement has lower precedence than ->
18:16:48 <Deewiant> According to http://www.difranco.net/cop2220/op-prec.htm at least
18:16:52 <oklokol> -> and . of course have the highest
18:17:10 <Deewiant> Postfix -- has the same precedence as ->, though
18:17:22 <Deewiant> But it's left-to-right so I guess it doesn't matter
18:19:59 -!- ais523 has quit (Remote closed the connection).
18:21:42 <AnMaster> <ais523> there's no guarantee as to whether the first stack will return the decremented or original version <-- um? Why not
18:22:34 <Deewiant> Short answer: it is well defined.
18:22:38 <AnMaster> so -- operates on the value of top then. Good.
18:23:26 <oklokol> even shorter would be "yes"
18:24:05 <AnMaster> Deewiant, since when is a space shorter than 1?
18:24:16 <oklokol> i tried not to say anything at all, but i'm not sure you would've noticed it.
18:24:23 <AnMaster> and in what language is a space == true
18:24:37 <Deewiant> AnMaster: Space is 0x20, 1 is 0x31
18:24:49 <Deewiant> And don't you know that silence is acceptance?
18:25:11 <AnMaster> Deewiant, *shorter* not "lower value"
18:26:29 <oklokol> you know the part containing the ones is shorter.
18:28:22 <oklokol> also i could close the repeating tetris music at some point.
18:37:45 -!- bsmntbombdood_ has joined.
18:38:04 <Deewiant> AnMaster: As unary it's shorter
18:39:01 <AnMaster> Deewiant, depends on field width.
18:39:13 <AnMaster> well I guess you mean unpadded
18:39:39 <Deewiant> Well sure, I can pad everything to infinity and then everything's the same length. :-P
18:39:43 <AnMaster> Deewiant, still I never heard of "silence is acceptance" before
18:40:03 <Deewiant> Meh, no such saying in Swedish either?
18:40:15 <ehird> Deewiant: so are all finns rapists?
18:40:42 <Deewiant> ehird: Since when are rape victims silent?
18:41:10 <ehird> Deewiant: clearly you have never raped a zombie.
18:49:36 <ehird> http://www.zareason.com/shop/product.php?productid=16189&cat=0&page=1 ← this is cute but ugh I hate those keys that are all rough
18:49:57 -!- Hiato has joined.
18:50:03 <ehird> can you buy standalone laptop keyboards?
18:50:06 <ehird> i love those thing
18:51:02 <lament> apple keyboard is pretty much like a laptop keyboard
18:51:14 <ehird> yeah, that's why I love them
18:51:14 <lament> http://www.instructables.com/files/deriv/FTQ/89XQ/F8QANFHG/FTQ89XQF8QANFHG.MEDIUM.jpg
18:51:35 <ehird> but I'd rather not pay apple quite a bit of dough for a kb when I'm moving away from macs.
18:51:49 <fizzie> "silence implies consent" seems to be the corresponding phrase; from latin, qui tacet consentiret; also some sort of legalistic thing, it seems.
18:56:57 <ehird> http://www.youtube.com/watch?v=6CPmd9hmNAI Fucking hell this is loud
18:59:45 <fizzie> This "Logitech UltraX Flat" keyboard I use is rather laptoppy; but this is the old model, the new one got buggered keys.
19:00:08 <ehird> http://matias.ca/tactilepro2/ ← apple extended keyboard in neu USBy form, looks nice
19:00:22 <ehird> wonder how much noise
19:00:23 <fizzie> And anyway they had a "Flat" model and a "FlatX" and some sort of "Ultra" thing and it was all so confusing; there was also a no-known-brand clone which looked pretty much identical.
19:01:11 <ehird> fizzie: http://www.youtube.com/watch?v=88E2xpkdxFk&feature=related, second one (flatx) doesn't look too tactile
19:02:48 <Deewiant> ehird: Er, you have a keyboard model you like but it's from the wrong company so you'll get a different one? Wtf?
19:02:57 <ehird> That's not what I said, Deewiant.
19:03:03 <ehird> You could try reading :P
19:03:06 <Deewiant> 09-04-22 20:51:35 ( ehird) but I'd rather not pay apple quite a bit of dough for a kb when I'm moving away from macs.
19:04:31 <fizzie> ehird: Yes, well, this one is neither "Ultra Flat", nor "Ultra X Premium"; it's "UltraX Flat", which *looks* like the later one (though not *exactly* like), but the keys are... you know, less softy. But they don't make this thing any more; I tried the new version in a shop, and it feels very different.
19:05:01 <ehird> I'd go for the das keyboard, were it... y'know... not as bloody loud.
19:05:02 <fizzie> "later one" there meaning the later one shown in the video.
19:05:09 <fizzie> I'm not sure which one it was, in fact.
19:05:20 <Deewiant> The DAS isn't even particularly loud, I don't think.
19:05:20 <fizzie> No audio made it a bit... abstract.
19:05:51 <ehird> Deewiant: Not about to buy one to see if that's true, quite expensive thing; reviews say it's = model m volume.
19:06:38 <Deewiant> The DAS has blue cherries, the model M is buckling spring. I'd expect an M to be noticeably louder.
19:06:46 <ehird> "wow u didnt even hit the spacebar nt 1 time u face shit" (comment on video of japanese person typing)
19:07:08 <fizzie> Keyboards are so boring, anyway; 2009 should mean some sort of direct neural connection already.
19:07:38 <ehird> Deewiant: It's Das, btw.
19:07:55 <ehird> "Das Keyboard sports best-in-class German-engineered gold-plated mechanical key switches"
19:07:59 <ehird> MONSTER CABLE FOR YOUR KEYBOARD
19:08:33 <Deewiant> Or not really, it's more the connector type.
19:08:34 <ehird> Deewiant: You're not srsly saying you think highly of Monster Cable?
19:08:49 <ehird> Yes. A scamster company.
19:08:56 <ehird> Charge $100+ upwards for a $5 cable.
19:09:01 <ehird> Claim it's better quality. Results say: no.
19:09:06 <ehird> Also, sue people for using the name "Monster".
19:09:09 <ehird> In any field whatsoever.
19:09:15 <ehird> Because they invented the word monster, see.
19:09:18 <fizzie> I'd buy a new keyboard like this if Logitech still made 'em; this particular one seems to, after some few years of service, have generated a problem that the space bar doesn't always register if I click on the extreme end of the key.
19:09:41 <ehird> do people actually do rollover when typing?
19:09:45 <ehird> I only press one key at a time, in general.
19:09:49 <ehird> albeit in rapid succession
19:10:37 <fizzie> Yes, well, Monster Cable sells "faster" HDMI cables too.
19:10:40 <Deewiant> I sometimes do accidental rollover which causes one of the keys to not register.
19:11:08 <fizzie> "The Speed You Need From Your High Definition Home Theater Cables". Well, maybe it's just flowery text there.
19:13:19 <ehird> anyone have a space cadet?
19:13:53 <fizzie> Selling gazillion-dollar cables for all-digital connections is a bit suspicious to me, anyway. Oh, and some of the cables are unidirectional; they won't work "right" if you hook them up the wrong way around, even though the connectors are identical on both ends.
19:14:30 <ehird> Deewiant: http://www.youtube.com/watch?v=cnY36odkP6o&feature=related this sounds the loud to me
19:14:44 <fizzie> "Well, turn down the volume."
19:15:11 <Deewiant> That's what cherry blues sound like. :-P
19:16:07 <Deewiant> See the video at the bottom of http://hothardware.com/cs/blogs/mrtg/archive/2009/03/08/mechanical-key-switch-keyboards-demystified.aspx for some variation.
19:17:20 <ehird> It's a bit high pitched, is my thinking.
19:17:27 <ehird> My model m and apple keyboard are both deep sounds.
19:18:13 <Deewiant> Yeah, the Das is higher-pitched than other cherry blue keyboards, or so I've heard.
19:18:47 <ehird> anyone used a scissor-switch kb
19:18:54 <ehird> supposedly the ones laptops use.
19:19:24 <Deewiant> Meh, they're membrane keyboards so they suck anyway. :-P
19:19:52 <ehird> I dunno, I like the feel of laptop kbs. You don't need to put much pressure on and that makes my type faster.
19:20:03 <ehird> fizzie: is your X thing scizswitch
19:20:51 <fizzie> How should I know? It certainly didn't say on the box. :p
19:21:21 <ehird> A special case of computer keyboard membrane switch is the scissors switch. The keys are attached to the keyboard via two plastic pieces that interlock in a "scissor"-like fashion, and snap to the keyboard and the key. It still uses rubber domes, but a special plastic 'scissors' mechanism links the keycap to a plunger that depresses the rubber dome with a much shorter travel than the typical rubber dome keyboard
19:21:24 <fizzie> I don't remember what it looked like, inside, either. Except that one of the longer keys had some sort of metallic-wire-trickery that was very tricky to get back in place when I removed it.
19:21:28 <ehird> I think my current one is that
19:21:36 * ehird pops off a key to see
19:22:42 <ehird> The keycap is separate; it has a hollow tube in the middle. This snaps into a thing on the actual keyboard which has another hollow tube with a rubber membrane at the bottom, again a hollow tube, although smaller and shorter.
19:22:57 <ehird> Deewiant: what does that sound like to you?
19:23:58 <ehird> Deewiant: Yes. Seems likely.
19:24:26 -!- tombom has quit ("Peace and Protection 4.22.2").
19:25:16 <ehird> http://matias.ca/tactilepro2/resources/images/TP2_hands_diagram_1.jpg "This keyboard for people with three or more hands"
19:25:39 <Deewiant> Hmm, those keyboards look identical
19:25:54 <fizzie> ehird: Yes, there seems to be two plastic pieces in a scissors-like fashion under the keycaps of this thing.
19:26:07 <ehird> Your mom is two plast— shot
19:26:09 <fizzie> Well, I only examined a sample of N=1 keys, but I don't really think they'd alter.
19:26:29 <Deewiant> Maybe it's the only one like that!
19:26:49 <ehird> it's to fool customers
19:26:55 <ehird> they told people to pop off keys
19:27:02 <ehird> and the 3 most common ones they made scissor'd
19:27:09 <ehird> to fool customers, the rest is just bogstandard membrane
19:28:06 <ehird> is the das keyboard smooth-key?
19:28:07 <fizzie> For the record, it was the number key 1. But as I said, they no longer make this thing; I'd bet the current version is not scissory at all, since the feedbacky sort-of-clicky-but-not-quite feeling is gone, it's just a soft resistance.
19:28:32 <fizzie> (Well, the current version of their product that still looks similar. I'm not quite sure how the all-black also-rather-slim version acts.)
19:28:33 <ehird> Deewiant: a lot of keyboards have plastic keycaps where if you look closely there's lots of little bobble
19:28:43 <ehird> and much prefer smooth plastic keycaps
19:28:57 <Deewiant> I'm annoyed by the bobbles on the home row keys and such
19:32:12 <ehird> My model m sounds nicer in my recording than in the real thing
19:32:34 -!- kar8nga has joined.
19:33:30 <ehird> http://www.youtube.com/watch?v=b5CeNunbHto&feature=related ← this sounds godly
19:33:39 <ehird> could do with more keys. and keycaps.
19:35:47 <fizzie> Whenever I press a key, something else on my board says something approximated by "bräng", thanks to the vibration. (I guess the onomatopoeicity of that is possibly a bit Finnish-only.)
19:36:11 <fizzie> Or is any table a board?
19:39:44 <ehird> And the Cherry MX Brown is tactile, but not clicky
19:40:44 <ehird> abs m1 is apparently tactile !click
19:40:59 <ehird> windows key has some sort of smooth bullshit on though
19:41:37 <ehird> http://www.youtube.com/watch?v=oue4KQkmBEM er sounds very clicky
19:42:59 <fizzie> Nowadays the Windows key should be in that embossed circle thing nowadays; I don't have any keyboards like that here.
19:43:28 <ehird> http://www.viddler.com/explore/HotHardware/videos/69/ second one sounds hot
19:43:31 <fizzie> Incidentally, do they still use the "correct" layout of ins-home-pageup-del-end-pagedown keys in addition to the weird vertical thing?
19:43:45 <ehird> it's clicky but without the cliiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiick
19:44:14 <fizzie> "We put the ICKY back in CLICKY" would be an awesome slogan for a keyboard manufacturer.
19:44:59 <ehird> call it kICKY or something
19:45:56 <ehird> the 7g has unsmooth keys
19:45:59 <ehird> look: http://www.steelseries.com/images/18/1561-1866.jpg
19:46:04 <ehird> also wtf@that layout
19:47:55 <GregorR> Conversation fragment from outside my office: "Yeah, but what if L7 goes up to sixty?" "What do you mean by sixty?"
19:48:06 * GregorR wonders how "sixty" can ever be ambiguous.
19:48:49 <GregorR> Hmmmmmmmmm, 'spossible. I assumed they were talking about some algorithm since this is the CS lounge :P
19:50:09 <GregorR> Is it actually possible to buy domain names from squatters, or is it purely mythical :P
19:51:19 <fizzie> I've been getting lots of targeted spam about gehennom.com becoming soon available for purchasation, to the webmaster@gehennom.org address.
19:51:42 <GregorR> I want plof.{something} :P
19:51:52 <GregorR> But they're all squatted because it's a four-letter word.
19:51:57 <ehird> Come up with a cctld yo.
19:53:04 <fizzie> As in the famous insult, "You're such a ploffo!"
19:53:59 <fizzie> Or "plof.it"; then you can paraphrase the famous slogan and have something in the just.plof.it address.
19:54:43 <fizzie> Some of those ccTLDs are rather strict about requiring someone/something in the country itself.
19:55:02 <ehird> I wanted velociraptu.re
19:55:09 <ehird> but you need to be in REtard land.
19:56:16 <fizzie> For .fi you also need to be someone living in Finland, or a company/organization/foundation registered in the corresponding Finnish registry.
19:57:28 <fizzie> fu.fi is registered to "All You Need Finland Oy" ("Oy" being the company-suffix).
19:58:04 <fizzie> The front page looks a bit bizarre. Maybe some sort of religious cult?
19:58:32 -!- Slereah_ has joined.
20:01:18 <fizzie> Funny, you only get to use (in addition to the "normal" characters) the characters å, ä, ö and the Sami language specials (but only those officially spoken in Finland) in IDNs in the .fi registry. That's rather a limitation.
20:05:03 <ehird> why aren't ther any smooth keyboards ;_;
20:05:34 <ehird> smooth criminalboards
20:13:32 -!- Slereah has quit (Read error: 110 (Connection timed out)).
20:17:42 -!- olsner has joined.
20:22:07 <ehird> the steelseries 7g is single width
20:22:15 <ehird> kill it with fire!!
20:27:40 <ehird> i just want a decent non-{tactile,clicky} mechanical keyboard
20:27:43 <ehird> preferably with smooth keys.
20:34:04 <fizzie> Just buy one of those Optimus Maximus keyboards, and be the envy of your neighbourhood. Even though as a keyboard I don't think it's especially good.
20:34:29 <ehird> no way am I paying $1k for a shitty keyboard.
20:36:35 <fizzie> Heh, right, it was that steelseries 7G which had gold-plated connectors. :p
20:36:51 <ehird> fizzie: what's your kb again
20:37:18 <pikhq> ehird, you want a clicky keyboard. Buckling springs FTW.
20:37:20 <fizzie> Logitech UltraX Flat, if I recall correctly; but it's been discontinued; they have something vaguely similar, but not at all.
20:37:33 <ehird> pikhq: Fuck that. I hate their sound and they're not fast to type on.
20:37:46 <ehird> The only vaguely similar thing I'm considering is the das keyboard.
20:38:25 <ehird> fizzie: http://www.amazon.com/Logitech-967568-0403-X-Flat-Resistant-Keyboard/dp/B000I4UQZM
20:38:33 <fizzie> "Gaming keyboards" are funny. This one advertises itself has having "2 megabytes of built-in memory and an integrated TurboCore unit".
20:39:12 <ehird> http://www.amazon.com/Logitech-Illuminated-Ultrathin-Keyboard-Backlighting/dp/B001F51G16/ref=pd_cp_e_2?pf_rd_p=413863501&pf_rd_s=center-41&pf_rd_t=201&pf_rd_i=B000I4UQZM&pf_rd_m=ATVPDKIKX0DER&pf_rd_r=1BWD2KVWZWF5K9H5VEKD this looks quite the nice
20:40:59 <fizzie> I think that "Ultra-X Flat" in the first Amazon model is the new variant, with non-scissory just-sort-of-soft-touch keys. It has that black border above, which I seem to recall was about the only obvious physical difference. But I don't really know.
20:41:53 <fizzie> Er, first Amazon link, I mean.
20:42:08 <fizzie> I think I did try some other all-black thing Logitech keyboard, and it didn't feel any better. I don't think it had any illuminatus connections, though.
20:42:09 <ehird> "What's this? A laptop-style keyboard on the desk of a TR editor, a heavy-duty typist and ostensible keyboard purist? Surely there must be some mistake"
20:42:12 <ehird> This sounds promising.
20:42:31 <ehird> http://techreport.com/articles.x/16522 Looks nice enough at first glance.
20:43:21 <ehird> although I hate that wrist rest
20:44:48 <ehird> "What's more, the Aurora fits in perfectly with my dual black Dell monitors, black speakers, black Logitech wireless mouse, black headphones, and shiny black Antec Sonata case."
20:44:55 <ehird> "But alas, I cannot change my race."
20:45:21 <ehird> grr that windows key makes me want to kill something
20:45:28 <ehird> FUCK YOUR LOGO AND FUCK YOUR GLOS
20:45:44 <ehird> ahahahahahhhhhhhhhhhhhhhhhhhhhhhhhhhh
20:45:57 <fizzie> I hate keyboard-shopping since it would be more optimal to actually get to poke them things, and that's hard over the internets.
20:46:19 <ehird> but i've never heard of a keyboard shop
20:47:08 <fizzie> Keyboard shops are hidden away at dark alleyways, only for the true keyboard connoisseur.
20:48:18 -!- Slereah_ has set topic: E=$BDE!N(Bc2 | http://tunes.org/~nef/logs/esoteric/?C=N;O=D.
20:48:46 <fizzie> The tiny backspace is tiny, truly.
20:50:20 <ehird> a youtube keyboard review without any typing video
20:50:55 <ehird> this one has full sized backspace key in this review
20:50:58 <ehird> but not on the site
20:51:16 <fizzie> Might be layout-dependant; I've seen that sort of thing.
20:51:31 <fizzie> Or maybe they're just playing with you.
21:07:42 -!- BeholdMyGlory has left (?).
21:09:10 -!- Hiato has quit ("Leaving.").
21:09:44 <fizzie> Our local the-major-computer-parts-retailer-around-these-parts-I-guess sells this thing called "Überkeyboard" which openly advertises the laptop-likeness and scissor-switchiness. But it also completely lacks all key markings. And has a bit miniaturized layout when it comes to cursor keys. And is a cheap plasticky sort of thing, too.
21:16:28 -!- neldoreth has quit (Connection timed out).
21:25:01 -!- k has joined.
21:25:29 -!- k has changed nick to Guest49992.
21:34:50 -!- kar8nga has quit (Read error: 110 (Connection timed out)).
21:50:48 -!- [flaming] has joined.
21:51:06 <[flaming]> ++++++++++[>+++++++>++++++++++>+++>+<<<<-]>++.>+.+++++++..+++.>++.<<+++++++++++++++.>.+++.------.--------.>+.>.
21:52:17 <[flaming]> can i run it without downloading anything?
21:52:43 -!- [flaming] has left (?).
21:54:11 <pikhq> I think that's "Hello, World!\n", but I'm not sure.
21:54:50 <Deewiant> Probably. I mean, what else would you run into at random? :-P
21:55:57 <fizzie> That was rather impatient.
21:55:58 <pikhq> That outputs 14 bytes?
21:56:15 <fizzie> ^bf ++++++++++[>+++++++>++++++++++>+++>+<<<<-]>++.>+.+++++++..+++.>++.<<+++++++++++++++.>.+++.------.--------.>+.>.
21:56:26 <fizzie> You must be a psychic.
21:56:48 <fizzie> The last dot there is probably the newline.
21:57:01 <pikhq> Yeah. Definitly a newline.
22:01:49 -!- okloduk has joined.
22:03:32 <kerlo> You know, I just noticed that in today's xkcd, the railing in the sixth panel is apparently a cross section.
22:04:05 <kerlo> If it were a side view, it would probably curve back toward the house at some point.
22:06:05 -!- oklokol has quit (Read error: 60 (Operation timed out)).
22:11:25 -!- ehird has left (?).
22:11:27 -!- ehird has joined.
22:35:13 -!- oerjan has joined.
22:36:49 <oerjan> <oklokol> a simple yes or no would've sufficed.
22:37:55 <okloduk> well okay, i guess yours was better.
22:37:56 <oerjan> oklopol: erm, do you get highlighted on any oklo* ?
22:38:20 <okloduk> actually just found two orthogonal 4x4 latin squares, and came to celebrate
22:38:37 <Deewiant> You probably should, though, since your nicks vary so much
22:38:54 <okloduk> oerjan: when you pair them up elementwise, all pairs different
22:39:30 <oerjan> why does your client give _two_ version replies? :D
22:39:50 <okloduk> which one did you version?
22:40:12 <okloduk> well it's mirc, it tends to do weird stuff when you ask it stuff.
22:41:59 <oerjan> you better paste those squares, that doesn't make sense...
22:44:29 <oerjan> the wp article on latin squares uses the term, but doesn't define it...
22:44:56 <okloduk> you have two matrices, and you take each element (i,j) from both matrices, if the matrices are size n, you get n^2 ordered pairs, orthogonality = all pairs different
22:45:17 <okloduk> also they both have to be latin squares
22:45:24 <oerjan> oh right just one for each matrix
22:45:42 <oerjan> i was confused and thought you took a pair from each
22:47:41 <okloduk> anyway it's a trivial problem, i wasn't trying to imply i was proud of that or anything
22:47:58 <okloduk> just needed an excuse for having appeared just in time
22:48:41 <oerjan> oh well, i'll pretend to believe your excuse then. er, whoops.
22:49:21 <oerjan> what the heck is that topic
22:49:54 <fizzie> The topic is something that did not work.
22:50:06 <fizzie> 22:48:18 ... Slereah_ changed the topic of #esoteric to: E=Æ$BDE!N(Bc2 | http://tunes.org/~nef/logs/esoteric/?C=N;O=D
22:50:09 <fizzie> 22:48:28 < ~Slereah_> Hm, it no work
22:50:19 <fizzie> I am unsure as to how it did not work.
22:51:04 <oerjan> maybe it was a wild attempt at injection
22:51:25 <oerjan> clearly he forgot to balance parentheses
22:54:11 <okloduk> in soviet russia, parentheses balance themselves
22:54:51 * oerjan swats okloduk for a clear meme failure -----###
22:54:57 <okloduk> okay i've done 48 exercises, if i do 3 more, i'll get cool extra points in the exam
22:55:05 <okloduk> but i already got 30/30 from the first exam
22:55:15 <okloduk> have wasted time vs. waste time
22:55:32 <okloduk> any suggestions? or maybe requests for clarification.
22:56:25 <oerjan> in soviet russia, time wastes YOU
22:56:39 <oerjan> admittedly, that has been known to happen under capitalism as well
22:57:32 <oerjan> also, in soviet russia, extra points in the exam get YOU
22:57:46 <okloduk> i actually had a "good" soviet russia joke today, but i just couldn't do it.
22:57:56 <okloduk> but now i'm tired and i can say anything.
22:58:10 <okloduk> i mean SR jokes are somewhat old
22:59:06 <fizzie> Yes, strontium jokes have truly ceasiumed to be amusing.
22:59:37 <oerjan> in soviet russia, yoda grammatical order changes
22:59:43 <okloduk> there exercises have about 3 pages of shit for me to read
22:59:53 <okloduk> don't they realize it's fucking past midnight
23:00:44 -!- FireFly has quit ("Later").
23:01:19 <kerlo> SR isn't strontium, it's sulfide.
23:01:40 <kerlo> Sulfides in general. That R could be anything.
23:01:55 <fizzie> I guess so; I just used "wn sr -over" to get something to say.
23:02:34 <oerjan> i don't know if it is germanium that we should sulphur such puns
23:02:57 <okloduk> fizzie: that's what you get for only using material from the channel
23:03:01 <oerjan> although it might lead to some gold
23:03:07 <okloduk> oh look i have a spoon next to me :o
23:03:16 <okloduk> probably because i just ate ice cream.
23:03:20 <fizzie> Like the old saying goes: when you make assumptions, you make an ass out of you and... 'mptions.
23:05:51 <fizzie> Well, it's a saying. I don't really get those either.
23:05:52 <kerlo> In Soviet Russia, the ability to get better services due to natural talent is deemphasized YOU!
23:06:10 <kerlo> When you assume, you make an ass out of u and me.
23:06:49 <ehird> 23:05 kerlo: In Soviet Russia, the ability to get better services due to natural talent is deemphasized YOU! ← capitalism fail
23:06:50 <fizzie> No, I don't think that's it. I'm certain it was something about umptions.
23:07:37 <oerjan> kerlo: that soviet russia joke didn't contain a silver of funny. but i guess we could iron something out
23:08:28 <kerlo> Yeah, sounds like a good idea.
23:09:53 <oerjan> fizzie: i think you must have read a scrambled version
23:10:53 <fizzie> My elementary comment wasn't a good idea, I take it.
23:13:00 <oerjan> fizzie: it was a boron idea, and if you do anything such again i'll call a copper
23:14:26 <fizzie> oerjan: Oh? Well... I'll unbihexuminate your untrilinums if you try anything that ununseptic.
23:15:12 -!- Guest49992 has quit (Connection timed out).
23:16:52 <oerjan> i think this is wearing a bit tin
23:19:16 -!- coppro has joined.
23:20:20 <fizzie> Ah! You in fact called the copper, it seems! With some slight changes in spelling.
23:20:58 <oerjan> you'd think they would see through a pun
23:23:42 <lament> I was gold by this conversation. I'm glad to see the puns finally argon.
23:24:12 <oerjan> lament: you repeated an element, lose 3 points
23:25:08 <oerjan> okloduk: there is no reason for swearing, sodium it
23:25:36 <okloduk> could you please stop seaborgium
23:26:04 <oerjan> this is getting out of hand, i don't understand the puns any longer
23:27:27 <okloduk> well, antimony to you too.
23:27:42 <okloduk> of course, i'm not actually sure these are international names, though they do sound like it
23:27:47 <oerjan> there could be a meaning to that, but i xenon
23:29:14 * pikhq drops some sodium into the pool.
23:30:24 <okloduk> lament is having a huge radon
23:30:40 <oerjan> pikhq: i think doing that is a chrome
23:30:43 <fizzie> All this is ytterbiumly ludicrous. Is there some curium I could use to make it stop? Perhaps some sort of good samarium could change the topic? Though it would be a titanium task.
23:31:43 -!- pikhq has set topic: Esoteric, the land of puns | http://tunes.org/~nef/logs/esoteric/?C=N;O=D.
23:31:53 <oerjan> this could go on for a lantanium
23:31:54 -!- ehird has set topic: Esoteric (verb) the land of puns | http://tunes.org/~nef/logs/esoteric/?C=N;O=D.
23:32:51 <okloduk> technetiumically we could try to do something less bohrium.
23:34:26 <lament> your puns are too krypton, try to bismuth
23:35:35 <ehird> VELOCIROLLERCOASTERAPTORS
23:35:55 * okloduk realized he can just download the solution manual and get the points for free!
23:36:13 <fizzie> Money for nothing and the points for free.
23:36:46 <okloduk> "how weird that you'd do the same mistake as the solution manual!"
23:37:04 <okloduk> (actual quote, although maybe it was less exclamation marky.)
23:37:10 -!- Dewio has joined.
23:37:11 <fizzie> Just try the "great minds think alike" defense.
23:37:39 -!- Gracenotes has joined.
23:39:48 <ehird> Gracenotes: see #concatenative.
23:39:52 <Gracenotes> usually, you should kick those sorts of peoples..
23:40:31 -!- Sgeo[College] has joined.
23:40:34 <Sgeo[College]> void countVowels(string str, int & aCt, int & eCt, int & iCt, int & oCt, int & uCt)
23:40:35 <ehird> Gracenotes: he uses cgi:irc
23:41:46 <Sgeo[College]> Can that be quieted without quieting all cgi:irc users using that network?
23:41:49 <oerjan> Sgeo[College]: make that varargs, so you can add an optional yCt ;D
23:42:13 <ehird> Gracenotes: claims he can't sleep without knowing slava's employer; he called phone numbers and shit
23:42:13 <Sgeo[College]> Seriously, I can't believe the professor thinks this is acceptable
23:42:22 <coppro> Why is that not an array?
23:42:27 <ehird> either completely psychotic or a troll; if he's a troll, he's incredibly good and incredibly dedicated
23:42:43 <coppro> that would probably be why
23:42:52 <Sgeo[College]> My question is why don't we just do countLetters(char ch, string str) and call it multiple times
23:43:11 <coppro> I'd use that as the underlying implementaton
23:43:18 <coppro> if only to irritate him
23:44:00 <Gracenotes> well, you could quiet his nick and make the channel +z
23:44:11 <Gracenotes> anyway. >_< trolls, shouldn't devote too much attention.
23:44:39 <ehird> current prediction is that he'll die of sleep deprivation
23:46:21 <Gracenotes> or there's the kick-on-sight solution. which can escalate
23:47:42 <kerlo> Gracenotes: how far along is said day?
23:48:57 <kerlo> I was about to say that I could make a list of everything I've eaten today, but then I realized that some flour tortilla chips and BUGLES have passed under the radar.
23:49:36 -!- Dewi has quit (Read error: 101 (Network is unreachable)).
23:49:53 <ehird> anyone want 3gb of ddr3 ram :-P
23:51:00 <Gracenotes> actually, 1,243... screwed up grams and calories. this logging might take some getting used to :3
23:51:46 <Gracenotes> I woke up at 9:00 AM, ate breakfast (bagel, so 600 calories), took a nap, ate sushi (258 calories), then some cereal (392 calories)
23:52:32 <Gracenotes> right now it's 7:00 PM and I'm going to the gym in an hour... phew.
23:53:01 <oerjan> Gracenotes: technically, those are probably kilocalories anyway
23:53:16 <Gracenotes> yes, they are, my SHIFT KEY IS ERRATIC RIGHT NOW
23:53:33 <Gracenotes> (actually, I just forgot to capitalize the C >_>)
23:53:39 <ehird> Gracenotes: you need some more grace notes in your day
23:54:14 <Gracenotes> oh, good idea, I should go play the piano after gym ^_^
23:55:25 <oerjan> did i just sense a pun backfiring?
23:56:44 <kerlo> Let's see, about 370 Calories at breakfast, 630 at lunch, then a bunch at whatever you call that meal between lunch and dinner.
23:57:25 <Sgeo[College]> I have a feeling that if I said "linner" before ehird, ehird would make a sarcastic remark about how unclever that is
23:57:33 <ehird> Sgeo[College]: yep.
23:58:19 <kerlo> Anyway, that and a fair bit of snacking puts me somewhere near 2000, I imagine.
23:58:34 <oerjan> Sgeo[College]: he'd probably call you a dunch for it
23:58:42 <kerlo> And it's 7 PM. I wonder if I could get away with not eating for the rest of the day.
23:59:17 <Gracenotes> I tend to eat into the night... so I should take that into account.
23:59:40 <Gracenotes> I'm just starting to log everything today. So far so good... eh.