00:07:41 <Friendship> <elliott> http://esolangs.org/wiki/Talk:Basic_Input/Output_Commander WHY WOULD YOU ARCHIVE THE PAGE LITERALLY DAYS AFTER IT WAS CREATED // why would you archive the page ... at all.
00:09:47 <elliott> Friendship: Well, archiving talk pages when they grow long is reasonable. Archiving them days after they were created, and are one section long, is not.
00:09:55 <elliott> (+ not using the template designed for the purpose)
00:12:44 <Friendship> I'll add a comment "Please don't archive when the talk page isn't long", then a day later it'll vanish, replaced with Archive 2
00:14:14 <elliott> Friendship: Actually, go ahead, then it won't look like I'm picking on the guy when I delete the archiev :P
00:15:27 -!- CHeReP has quit (Quit: Leaving).
00:18:31 -!- derdon has quit (Remote host closed the connection).
00:18:59 -!- cswords has joined.
00:22:28 -!- cswords_ has quit (Ping timeout: 276 seconds).
00:22:31 <kallisti> elliott: what display manager are you using?
00:25:27 <elliott> Occasionally I've used xdm, which can be whipped up into looking reasonable rather than like ass in a few minutes.
00:28:56 <kallisti> I think slim looks pretty nice once properly configured.
00:30:04 <kallisti> ...didn't we already have this conversation and determine that it's maintained?
00:31:18 <kallisti> unless the website is out of date as well, it still has a maintainer.
00:31:34 <elliott> Plenty of packages "have maintainers" but are unmaintained.
00:31:52 <elliott> OK, last release was 2012-02. However the release before that was 2010-07 and is what Arch has (and I bet Debian too).
00:31:59 <elliott> Perhaps it's been revived.
00:32:08 <elliott> But that still means it likely has a couple of years bitrot in it.
00:32:16 <kallisti> also, I configured my prntscr key to automatically place the image in my dropbox public folder, and then copy the public URL to primary selection.
00:32:25 <kallisti> it's basically really obnoxious to anyone else but me.
00:32:37 <kallisti> because I could, if I wanted, just spam second-by-second screenshots.
00:32:52 <kallisti> yes but the public url is instant
00:32:57 <kallisti> you do have to wait for it to upload.
00:33:04 <kallisti> otherwise the link is just 404
00:34:49 <kallisti> also my xmobar is completely riced out.
00:35:05 <kallisti> I'm more ricer than those youtubers with compiz.
00:35:36 <kallisti> (okay, maybe not that extreme... that's pretty ricery)
00:35:59 <kallisti> but then... xmobar was already working and laziness happened.
00:36:10 <elliott> dzen2 is actually not that good, but I'm bound by pact of honour to shachaf to not use xmobar, and by pact of fizzie to use what he uses, i.e. dzen2.
00:36:26 <kallisti> I mean, what actual features does it have.
00:36:55 <elliott> Probably nothing over dzen. xmonad has integration for both.
00:37:41 <kallisti> gaze at my obnoxiously dark theme http://dl.dropbox.com/u/16495819/2012-03-04_19-36-26_1366x768.png
00:38:16 <elliott> Your text rendering settings are all wrong, :(
00:38:45 <elliott> You have medium/full hinting when you want slight.
00:39:00 <elliott> <!DOCTYPE fontconfig SYSTEM "fonts.dtd">
00:39:00 <elliott> <edit name="hintstyle" mode="assign">
00:39:08 <elliott> Put that in /etc/fonts/local.conf.
00:39:18 <elliott> (Except IIRC sometimes Chrome has trouble listening to those. Pah.)
00:39:36 <elliott> Also why the fuck do you use a line cursor rather than a block in your terminal?
00:40:21 <elliott> Why aren't you using Chrome?
00:40:30 <kallisti> also because google privacy policy
00:40:56 <kallisti> and I like line cursor because that's what I'm used to.
00:41:08 * shachaf didn't realize that pact existed.
00:41:18 <shachaf> Use xmobar if you like. I don't care.
00:41:31 <elliott> kallisti: "Google privacy policy" is rather irrelevant to Chrome.
00:41:45 <elliott> I mean, Mozilla aren't all sunshine and flowers, either.
00:41:46 <Sgeo_> elliott, does this page display properly for you in Chrome? It doesn't for me but does in Firefox.
00:41:46 -!- oerjan has joined.
00:41:47 <Sgeo_> http://www.flyingmachinestudios.com/programming/chunky-bacon-lisp/
00:42:07 <kallisti> elliott: also I was having issues with html5 video in youtube for some reason.
00:42:12 <kallisti> firefox doesn't have that problem.
00:42:37 <shachaf> elliott: I'm not using Chrome.
00:42:48 <shachaf> I hope there's no pact about that that I wasn't aware of.
00:43:06 <kallisti> elliott: why do I want slight hinting
00:43:09 <kallisti> I don't know much about hinting
00:43:16 <elliott> Sgeo_: Yes, Chromium 17.0.963.56 displays that page the same as Firefox 10.0.2 here.
00:43:33 <elliott> kallisti: Stronger hinting ruins the line shapes of the fonts and makes them look thin.
00:43:48 <Sgeo_> elliott, no words jumbled together and going off margins?
00:44:00 <elliott> kallisti: Slight hinting is "blurrier", but only for the short amount of time it takes you to get used to it, and is easier to read after that (on the majority of displays). It's the default for Ubuntu.
00:44:09 <elliott> Sgeo_: I'm on the Chrome developer build, though.
00:44:19 <Sgeo_> elliott, I seem to be using the same exact build
00:44:26 <kallisti> elliott: I don't have a local.conf. should I create it?
00:44:34 <Sgeo_> Developer Build 121963
00:44:47 <Sgeo_> Wait, Ubuntu 11.10? Wat?
00:44:55 <Sgeo_> I thought I was on 11.04, hmm
00:45:08 <elliott> kallisti: Yeah. You'll need to restart your session for it to take effect, too.
00:45:14 <elliott> kallisti: (You might also want to enable subpixel rendering, while you're at it.)
00:45:33 <Sgeo_> Could it possibly be LXDE's fault?
00:45:44 <elliott> $ cd /etc/fonts/conf.d && sudo ln -s ../conf.avail/10-sub-pixel-rgb.conf .
00:45:58 <Sgeo_> I should figure out how to screenshots on here
00:46:39 <Sgeo_> kallisti, Lubuntu 11.10
00:47:23 <lambdabot> elliott said 8h 25m 10s ago: gosc is the compiler ais523 talks about a lot from idealised ALGOL (with lambdas) to VHDL
00:48:03 <Sgeo_> http://imgur.com/Qz89S
00:48:10 <Sgeo_> Why does that page look like that?
00:48:44 <elliott> Try decreasing + re-increasing font size.
00:49:27 <Sgeo_> scrot was already installed. And apparently running the last few times I tried to take screenshots and thought I failed
00:49:59 <kallisti> have you had any trouble with ACPI?
00:50:05 <kallisti> my laptop doesn't suspend when I close the lid.
00:50:44 <elliott> This laptop doesn't ever do that, I don't think.
00:50:53 <elliott> I never use suspend on here though.
00:51:15 <kallisti> GTK apps look like shit still.
00:51:27 <Sgeo_> Hey, maybe I can actually re-read Boatmurdered now
00:52:04 <Friendship> What's the past tense of creep (verb)?
00:52:30 <Friendship> Was wondering why "creeped" had red squigglies.
00:52:51 <oerjan> i think creeped has a different meaning, as in creeped out
00:53:11 -!- iambored has changed nick to PiRSquared.
00:53:24 <oerjan> iambored at the jamboree
00:53:26 <elliott> He creeped out of the photo shoot, utterly crept out.
00:53:26 <kallisti> elliott: as long as I'm running firefox and terminal apps everything looks dandy.
00:53:35 <kallisti> but if I open up windowed emacs..
00:53:40 <elliott> kallisti: emacs has colour schemes too.
00:53:43 <elliott> GTK won't help much there.
00:53:50 <elliott> Install the elisp goodies package.
00:53:55 <kallisti> I've just been using emacs -nw ...
00:54:07 <elliott> Friendship: I TAKE IT YOU'RE NOT GOING TO MAKE THAT COMMENT
00:54:32 <kallisti> elliott: this has been a good venture. I've learned a few things about linux.
00:54:37 <kallisti> also, I've learned that screen is awesome.
00:55:02 <kallisti> I wish there was also a "screen" for X apps.
00:55:11 <kallisti> there's probably a way to do it.
00:55:34 <kallisti> I like it solely for the reason that it preserves console programs as I repeatedly restart my X session.
00:55:35 <elliott> kallisti: If you just use screen for the detaching (as implied by your X apps comment), you might consider dtach.
00:55:44 <elliott> But it doesn't do anything more than that.
00:55:48 <elliott> Also I'm not sure it's in Debian.
00:56:05 <kallisti> the ctrl+a thing actually gets in the way
00:56:14 <kallisti> and yes dtatch is in the repos
00:56:16 <elliott> Well, it has to have some key to detach :)
00:56:21 <elliott> screen's and dtach's are both configurable, I think.
00:56:27 <elliott> Anyway, yes there's detaching for X apps, but it's kind of ugly I think
00:56:43 <elliott> X gives you one nicety that nobody wants (network transparency) and withholds all the others.
00:57:08 <kallisti> I just... haven't. I think I have configure fatigue.
00:58:22 <Sgeo_> Maybe I can re-read Boatmurdered now.
00:58:31 <Sgeo_> Hmm, would an epub of Boatmurdered be a bad idea?
00:58:34 <elliott> <Sgeo_> Hey, maybe I can actually re-read Boatmurdered now
00:58:35 <elliott> <Sgeo_> Maybe I can re-read Boatmurdered now.
00:58:42 <elliott> Maybe Sgeo_ can re-read Boatmurdered now.
00:58:43 <kallisti> wtf was I even doing I forgot...
00:58:50 <elliott> kallisti: Reading conduit documentation.
00:58:51 <monqy> maybe sgeo- can re_read boatmurdered now
00:58:58 <Sgeo_> I think I'm losing my memory.
00:59:05 <olsner> maybe Sgeo_ can be boatmurdered now
00:59:08 <Sgeo_> Hmm, I think I'm having some memory issues.
00:59:15 <Sgeo_> Hmm, I think I'm forgetting stuff.
00:59:19 <Friendship> Maybe Sgeo_ can transmute into a boat now.
00:59:26 <kallisti> elliott: no that was a long time ago.
00:59:36 <kallisti> I was... likely about to configure something.
00:59:44 <kallisti> as that's what I've been doing for the past 4-5 days or so.
01:00:35 <kallisti> oh, right. I was going to restart slim.
01:00:46 <elliott> No, just quit yr session to get the hinting.
01:00:54 <elliott> OK, I guess that might technically restart slim.
01:01:14 <elliott> So how long did it take you to get sick of Xfce?
01:01:15 <kallisti> I can't tell if it's different.
01:01:25 <kallisti> http://dl.dropbox.com/u/16495819/2012-03-04_20-00-50_1366x768.png
01:01:38 <kallisti> it looks a bit better I think.
01:01:44 <kallisti> ...oh I need to change my irssi font.
01:01:48 <kallisti> or... is that terminators font?
01:01:52 <Sgeo_> How much is color needed to read Boatmurdered?
01:01:55 <Sgeo_> I want an epub version
01:02:06 <elliott> kallisti: What does it look like in Firefox?
01:02:15 <kallisti> so irssi uses whatever the terminal uses right?
01:02:41 <elliott> Shut up with stupid terminal questions and screenshot Firefox.
01:02:41 <kallisti> because... I'm pretty sure I went straight from Ubuntu with GNOME 2 to Debian with xmonad.
01:03:02 <elliott> No, you installed Xfce against my advice.
01:03:09 <Sgeo_> "They literally put the work into the public domain by putting it on a public forum for everyones use."
01:03:13 <Sgeo_> http://www.bay12forums.com/smf/index.php?topic=100035.0
01:03:14 <kallisti> no you're thinking of someone else.
01:03:20 <Sgeo_> Someone please hand me a face to palm.
01:03:36 <elliott> kallisti: This isn't amusing.
01:03:41 <kallisti> http://dl.dropbox.com/u/16495819/2012-03-04_20-03-02_1366x768.png
01:03:49 <kallisti> elliott: I have not once touched Xfce
01:03:57 <Sgeo_> "I guess the others assumed they would be greatful that other people are spending their free time to help spread the influence and accessability of their work, and not asking for a wage or any payment whatsoever for doing so.
01:03:57 <Sgeo_> They should be greatful.
01:03:57 <Sgeo_> Asking permission is like saying "Please oh please can I spend my time putting your work into a format so even more people can read it".
01:03:57 <Sgeo_> Fuck that, hes doing them a favor, though I'm sure he will ask out of pure courtesy.
01:03:58 <kallisti> a brief conversation about it, as I was considering new DEs
01:04:13 <elliott> kallisti: Firefox's toolbar uses the new settings. (You can compare with your previous screenshot.)
01:04:19 <elliott> Whether the text does is another matter.
01:04:30 <elliott> Your terminal continues to use the older sessions; I think I know why.
01:04:54 <kallisti> oh right that wouldn't make sense.
01:04:54 <elliott> Put this in ~/.Xresources:
01:05:14 <elliott> in your .xinitrc/.xsession/whatever.
01:05:18 <elliott> Then restart your session again.
01:05:28 <elliott> Actually, scratch that lcdfilter line.
01:05:51 <Sgeo_> Gah this person is a moron
01:05:52 <kallisti> uh, I'm guessing those random [[ ]] things are part of the config?
01:06:05 <elliott> kallisti: No, that's how I quote long sessions of text.
01:06:17 <Sgeo_> elliott, I know you don't believe that copyright is a good thing, but do you at least recognize that currently, the law is what it is?
01:06:26 <Sgeo_> I assume you do because you're sane.
01:06:45 <kallisti> does it make hinting better on LCDs?
01:06:47 <elliott> Sgeo_: What do you mean "what it is"?
01:07:05 <elliott> kallisti: The LCD filter is complicated font rendering stuff you wouldn't understand(tm). lcddefault is actually... the default.
01:07:12 <elliott> So you can remove that line.
01:07:16 <elliott> (Yes, I changed my mind again.)
01:07:16 <Sgeo_> That it exists. That it can be illegal to redistribute stuff, even if it's not immoral.
01:07:30 <elliott> No, I'm actually delusional.
01:07:34 <kallisti> elliott: I'm assuming it's a filter thigny that makes the stuff better on the liquid crystals?
01:07:36 <elliott> Everything is legal apart from killing pigs.
01:08:06 <elliott> kallisti: The rgba setting is what turns on subpixel (= LCD) rendering. The lcdfilter is just a parameter.
01:08:23 <elliott> See http://www.freetype.org/freetype2/docs/reference/ft2-lcd_filtering.html if you must know the details.
01:08:52 <Sgeo_> elliott, read the thread I linked?
01:09:13 <kallisti> elliott: recommendations on console font? I'm about to switch to Inconsolata (which is the font in my xmobar)
01:09:19 <elliott> Sgeo_: I read about three posts. Is the rest relevant or interesting?
01:09:34 <elliott> kallisti: I use the default. Inconsolata isn't very good with Freetype IMO.
01:09:52 <elliott> kallisti: Get hinting + subpixel working before changing.
01:09:58 <Sgeo_> elliott, it starts getting dumb about 5 posts in.
01:10:00 <elliott> It'll look different enough that you might change your mind.
01:11:45 <elliott> Screenshot? (For important science reasons. Also if you included Firefox with web page text that would be helpful.)
01:12:03 <elliott> Sgeo_: "I seriously doubt the legality of your statements." What an idiot, hasn't he heard of the FIRST AMENDMENT?
01:13:03 <Sgeo_> That I assume is just poor phrasing on his part.
01:13:17 <Sgeo_> I assume he meant to say legal correctness
01:13:30 <kallisti> http://dl.dropbox.com/u/16495819/2012-03-04_20-12-50_1366x768.png
01:14:06 <kallisti> you didn't really specify what kind of website
01:14:21 <elliott> Yes, that looks correct. Including the objective matter of which website to choose.
01:14:45 <elliott> Hmm, some colour fringing on one or two pieces of text on the site, but I assume it's due to the size and background; looks correct to me.
01:14:45 <kallisti> I'm noticing that inconsolata actually looks kind of weird now.
01:14:55 <elliott> kallisti: Actually xmobar has looked like that the entire time.
01:14:59 <elliott> It seems to be ignoring your settings in general.
01:15:12 <elliott> http://dl.dropbox.com/u/16495819/2012-03-04_19-36-26_1366x768.png
01:15:13 <elliott> http://dl.dropbox.com/u/16495819/2012-03-04_20-12-50_1366x768.png
01:15:29 <kallisti> ...this convenient screen shot may be a bad idea
01:15:30 <elliott> It does have quite awful colour fringing. I think that's due to the size it's at.
01:15:34 <kallisti> as elliott is going to compile a record of my screens
01:15:38 <kallisti> http://dl.dropbox.com/u/16495819/2012-03-04_20-12-50_1366x768.png
01:15:46 <kallisti> so I have this habit of middle clicking links
01:15:48 <elliott> THANKS HADN'T SEEN THAT ONE.
01:16:03 <elliott> kallisti: You can set rxvt up to open those in a browser.
01:16:06 <elliott> (What terminal are you using?)
01:16:09 <elliott> (If it's not urxvt, it's wrong.)
01:16:17 <kallisti> terminator has fancy link stuff
01:17:10 -!- pikhq has quit (Ping timeout: 245 seconds).
01:17:14 <kallisti> maybe I have to add something to xmobar config?
01:17:15 -!- pikhq_ has joined.
01:17:31 <kallisti> http://sprunge.us/EYXN?haskell
01:17:50 <elliott> I don't really know anything about xmobar configuration.
01:17:59 <kallisti> it uses some obscene pseudo-Haskell record syntax
01:17:59 <elliott> I'd just change the font to Mono. :p
01:18:21 <elliott> As in, I believe it does the same trick as xmonad.
01:18:22 <kallisti> I mean it's not an actual Haskell program obviously.
01:18:37 <elliott> It's a Haskell expression.
01:18:41 <kallisti> and it gives shitty parsec-like errors
01:18:43 <monqy> made to look like haskell
01:18:55 <elliott> dzen2's configuration is... okay, not any better, but less tacky.
01:19:04 <elliott> kallisti: Why not just use xmonad's xmobar module?
01:19:23 <elliott> Wait, it doesn't have one.
01:19:26 <kallisti> what is this i am bad at computer
01:19:47 <kallisti> I'll ask #xmonad about the font stuffs.
01:20:10 <elliott> xmonad is unrelated to xmobar.
01:20:15 <elliott> Don't spam some random other channel.
01:20:25 <kallisti> #xmonad helps people with all sorts of things
01:20:29 <kallisti> including things completely unrelated to xmonad
01:20:35 <kallisti> like how to get a display manager working.
01:20:35 <elliott> Anyway, it looks like it's actually using the same settings as your real settings now.
01:20:45 <elliott> It's just colour fringing badly because of the font and size.
01:21:03 <kallisti> Courier perhaps? I like courier....
01:21:24 <elliott> kallisti: Just put "Mono" there and it'll use the same as your terminal.
01:21:30 <elliott> Don't use Courier. You probably don't even have Courier.
01:21:45 <kallisti> I installed something that looked like it would have courier.
01:21:49 <elliott> (Except it should be xft:Mono:size=9.5.)
01:21:57 <elliott> Or, wait, is one in pt and the other px.
01:22:02 <elliott> Gah, it's too late for this crap.
01:22:08 <elliott> kallisti: Courier New, probably.
01:22:52 <elliott> The naming of ridiculous fonts can now end, as I've taken it to the fixed point.
01:23:01 <kallisti> dude goudy old style isn't ridiculous.
01:24:08 <kallisti> that will complement my dark theme (and soul)
01:24:45 <kallisti> I think I'll write my next paper in Zapfino
01:25:54 <elliott> Sgeo_: I like the resolution, where the guy admits he's wrong on an argument that hinges on SA's irrelevant paywall.
01:26:13 <elliott> A good twist ending to a fine comedy.
01:27:02 <elliott> 72 point Impact is the typeface of a generation.
01:27:20 -!- zzo38 has joined.
01:27:27 <zzo38> Do you write neatly?
01:28:51 <kallisti> elliott: is there a way to configure xscreensaver to make it look less shitty?
01:29:05 <elliott> But just don't use a screensaver.
01:29:15 <kallisti> the about window that pops up on startup gives me flashbacks of Windows 95
01:29:40 <elliott> It's jwz, man. The flashback you're looking for is Netscape 3.
01:30:01 <kallisti> or I guess I should say: is there a better screensaver / lock thing
01:30:19 <elliott> I don't think anyone uses anything other than xscreensaver for screen saving. For locking, sure.
01:30:28 <olsner> I've never seen xscreensaver pop up a window on startup
01:30:30 <elliott> There's xlock, or slock (http://tools.suckless.org/slock).
01:30:47 <olsner> (if you've seen it, you must be doing it wrong)
01:31:21 <kallisti> I see no options to "remove obnoxious about screen"
01:33:20 -!- cheater_ has quit (Ping timeout: 255 seconds).
01:34:00 <elliott> pikhq_: So how goes musl C++?
01:37:53 <pikhq_> elliott: 0.9.0 should support C++, current git basically works.
01:38:28 <pikhq_> Current work is convincing LLVM devs to not try to use std{in,out,err} as lvalues.
01:39:11 <kallisti> elliott: let's say in the future I want to completely move this setup to a new Debian install...
01:39:21 <kallisti> would it be insane to copy root into the new install?
01:39:41 <elliott> If you copied /, you'd be copying... the entire installation.
01:39:45 <elliott> Thus making it not a "new Debian install" at all.
01:39:56 <elliott> Why would you want a new Debian install?
01:39:58 <kallisti> like, say I have a new computer.
01:40:04 <kallisti> and I want all of this stuff exactly the same
01:40:05 <pikhq_> (C spec requires that those be macros that evaluate to a value of type FILE*, musl defines them as FILE* const variables and #define stdin (stdin))
01:40:14 <elliott> kallisti: Oh. That would be unwise.
01:40:39 <kallisti> elliott: so... copy every config file over by hand on the new system? sounds fun.
01:40:45 <elliott> kallisti: You can copy over non-dotfiles in ~, and important dotfiles (xmonad.hs, whatever xmobar's is called, .Xresources, .xsession, emacs config, etc.) individually.
01:40:51 <Friendship> <pikhq_> Current work is convincing LLVM devs to not try to use std{in,out,err} as lvalues. // wut
01:40:58 <elliott> kallisti: Don't copy everything, just important stuff.
01:41:12 <elliott> kallisti: You can keep dotfiles you care about in a $VCS repository and put it on GitHub or whatever.
01:41:15 <pikhq_> Friendship: Yes, they're morons and think that they can use them as lvalues.
01:41:20 <elliott> Then it's easier to use multiple machines.
01:41:29 <pikhq_> Which kinda-sorta works on glibc.
01:41:56 <kallisti> I'm just likely to forget what all I've configured.
01:42:20 <pikhq_> Testing hasn't been *extensive* of musl's C++ support as yet, but it sure seems to work.
01:42:25 <elliott> kallisti: You can put parts of /etc in a repo too.
01:42:27 <pikhq_> cmake has been built, for instance.
01:42:35 <elliott> kallisti: But most of them you'd want to maintain as essentially diffs.
01:42:58 <elliott> Things like local.conf could be copied wholesale, but modifying existing configs = you don't want to overwrite newer versions with that.
01:43:14 <kallisti> elliott: also I discovered mpd.
01:43:23 <kallisti> I don't know if you've ever used it.
01:43:35 <elliott> I think I preferred xmms2 architecture-wise, but I forget why.
01:43:36 <pikhq_> elliott: Not been tested just because those require a lot of library building, and nobody's bothered hardcore distroing quite yet.
01:43:40 <kallisti> the nice thing is the network support, which I may utilize later once I have a functioning server.
01:43:42 <elliott> But I think I gave up on it for having less language support.
01:44:25 <elliott> kallisti: I doubt your connection can stream your files.
01:44:32 <elliott> Unless they're 128 kbps MP3s or something.
01:44:34 <pikhq_> Perhaps a more fruitful test would be Firefox.
01:44:39 <elliott> kallisti: So you'll need to double-encode.
01:44:45 <kallisti> elliott: I meant with like a VPS or a dedicated server or whatever.
01:44:47 <pikhq_> (which is mostly C++, but has no non-C++ dependencies, as far as I'm aware)
01:44:51 <kallisti> elliott: in the distant future.
01:44:56 <elliott> kallisti: Yes, I know exactly what you meant.
01:45:03 <elliott> You would have to connect to that and stream audio to your home computer.
01:45:15 <elliott> This would be unviable unless you have either a very good connection, or are willing to double-encode them.
01:45:24 <elliott> It's designed for LAN use.
01:45:44 <elliott> (Also mpd has no native support for streaming the audio itself to the network.)
01:45:50 <elliott> (You'd have to run icecast separately.)
01:45:57 <elliott> It supports network _control_, that's all.
01:46:47 <kallisti> hm, I thought you could configure clients to network stream.
01:47:27 <elliott> You could point the server at an icecast backend, I think it supports that.
01:47:29 -!- cheater_ has joined.
01:47:33 <elliott> But you'd have to set up icecast yourself.
01:47:49 <elliott> And I think transcoding would be required unless absolutely everything you have is an MP3.
01:48:08 <kallisti> I converted everything to MP3 a long time ago to conserve disk space on my laptop.
01:48:19 <kallisti> in hindsight probably should have gone with ogg.
01:49:18 <elliott> You transcoded everything to MP3?
01:49:25 <elliott> I explicitly told you you were a fucking moron beforehand.
01:49:31 <kallisti> with lossless codecs my music collection is basically too big for my 300GB HDD.
01:49:44 <elliott> You transcoded non-lossless things too, I believe.
01:50:05 <elliott> Anyway, I was under the impression that you owned a 1 terabyte external disk.
01:50:17 <kallisti> and so it's less of a problem.
01:50:34 <kallisti> DRASTIC TIMES CALL FOR DRASTIC MEASURES, ELLIOTT
01:51:04 -!- Tiktalik has changed nick to Sicktalik.
01:51:08 <kallisti> otherwise my music library would have creeped to a standstill.
01:52:17 <kallisti> actually I have a lot of poor quality audio files that I need to replace.
01:52:23 -!- pikhq has joined.
01:52:29 -!- pikhq_ has quit (Ping timeout: 260 seconds).
01:52:30 <kallisti> (by poor I mean less than 300 kbps of course.)
01:53:18 <kallisti> (actually that's a bit extreme. less than 200)
01:53:45 <elliott> kallisti: A constant bitrate 200 kbps MP3 has a very good chance of being worse than an MP3 that averages 160.
01:54:00 <elliott> Because it indicates it's either old, or the encoder has no idea how to use LAME effectively.
01:54:07 <elliott> The VBR presets are a reason.
01:54:20 <elliott> The encoding community is full of a bunch of superstition (e.g. hating joint stereo because they don't understand it).
01:54:29 <kallisti> I have some files that are like.... 96 kbps
01:55:01 <kallisti> the <300 thing was a joke, the <200 thing was slightly less of a joke
01:55:28 <kallisti> but is it objectively worse to have a constant bitrate higher than 160?
01:55:32 <kallisti> like... that doesn't make sense to me.
01:56:06 <kallisti> assuming that the VBR presets were done correctly.
01:56:18 <elliott> No, not inherently, it is merely an indicator.
01:56:34 <elliott> Also, consider when you have a constant bitrate of 161, but a VBR preset averaging 160 would have gone to 220 for a certain amount of time.
01:56:53 <kallisti> yeah I didn't factor in that it was VBR.
01:57:17 <kallisti> which actually means something like "better than this number" but to what degree I don't really know.
01:57:43 <elliott> I don't know what you mean by "better than this number".
01:58:33 <kallisti> well I was basing this decision to replace some of my audio files on the little "bitrate" number that rhythmbox shows.
01:59:01 <elliott> It probably just divides the length of the song by the size of the audio data or such.
01:59:16 <kallisti> well, that would kind of be the average right?
01:59:16 <elliott> ...also known as the average, are you thick?
02:00:24 <kallisti> I'm not sure how to articulate what I meant
02:00:34 <kallisti> basically I don't know how to reason about the quality of a VBR based on its average bitrate.
02:01:12 <kallisti> but, you could think of a VBR with some average bitrate x, as being "slightly better" than something with a constant bitrate of x.... right?
02:01:26 <elliott> But anything with a sufficiently low average that isn't minimalist music or acapella something will likely be crap.
02:01:28 <kallisti> since they both have the same average.
02:01:42 <elliott> I'd just trash crap as and when I play it and hear that it is crap.
02:02:08 <kallisti> ...actually I don't have a good stereo system so I'm likely not going to hear much of a difference.
02:02:13 <kallisti> I'll let you know when I get my epic sound system
02:02:38 <elliott> 128 kbps on LAME since years ago is transparent to most people.
02:02:50 <elliott> 160 kbps should be transparent to just about everyone.
02:03:00 <elliott> Distinguishing 192 kbps from the real thing requires incredibly pathological samples.
02:03:17 <elliott> If something's 128 kbps it's probably done with really old LAME though, so there's that.
02:08:02 <shachaf> ~1 bps is transparent for me.
02:08:13 <pikhq> And if you think the *audio* encoding community is bad, wait till you see the video encoding community!
02:08:58 <pikhq> I've seen people use --crf 13 to encode and then use the resulting average bitrate from *that* as the target bitrate for a 2-pass encode.
02:10:29 <Friendship> I absolutely need a constant bitrate ... but I don't care what.
02:10:39 -!- nys has quit (Quit: quit).
02:11:25 <pikhq> elliott: x264 has 3 types of encodes, 1 pass, 2 pass, and CRF. The CRF type is a "constant quality" type of thing, similar to LAME's presets.
02:11:53 <elliott> So they use a preset and then throw away all the additional intelligence from the preset???
02:12:08 <elliott> To smooth out the imperfections, right?!?!
02:12:24 <pikhq> --crf 23 is pretty good, --crf 19 will look transparent for most everything, and --crf 13 gets you 1GiB 720p encodes of 20 minute episodes.
02:12:51 <Friendship> --crf 1 is actually larger than a directory full of uncompressed .tiffs.
02:13:04 <elliott> . o O ( Why does Friendship know about video encoding? )
02:13:16 <ion> I take it CRF uses two passes?
02:13:18 <pikhq> Friendship: Actually, no, that hits x264's lossless encoding mode.
02:13:19 <Friendship> You realize I'm spouting bullshit, right?
02:13:35 <elliott> Friendship: I thought you had actually measured it.
02:13:58 <Friendship> My knowledge of video encoding is exactly up to what's necessary to occasionally upload something to YouTube.
02:14:08 <pikhq> It gets worse. In x264, CRF and 2 pass mode are almost identical. CRF just plugs in a constant where 2 pass mode computes a value from the 1st pass.
02:14:23 * shachaf likes monotonic bit rates.
02:15:01 <pikhq> Oh, sorry, I should also specify. 1GiB 720p encodes of 20 minute *anime* episodes.
02:15:14 <shachaf> pikhq: How can the video encoding community be worse than the audio encoding community?
02:15:28 <pikhq> (anime encodes stupidly well with x264)
02:15:28 <elliott> Anime has shallower plots than American media, so it compresses better.
02:15:40 <shachaf> I have no idea what audio is. Video makes sense, at least.
02:15:49 <pikhq> elliott: Animation in general compresses well, due to the reduced movement.
02:15:56 <elliott> No, it's definitely the plot.
02:16:19 <elliott> (Surely it's the simplicity of the image detail compared to RL, too?)
02:16:36 -!- augur has quit (Remote host closed the connection).
02:16:37 <pikhq> Friendship: Actually, if it was porn, it'd be encoded to standards appropriate to 2000.
02:17:16 <pikhq> elliott: Actually, if anything that's a bit harder.
02:17:41 <pikhq> Friendship: Not scene releases!
02:17:52 <pikhq> Friendship: They use Xvid!
02:18:04 <elliott> Friendship: Can you tell by the pixels?
02:18:22 -!- augur has joined.
02:18:23 <Friendship> elliott: And having seen a lot of porn in my day.
02:18:43 <pikhq> ... Put in RARs, split at 100MiB.
02:19:32 <pikhq> With a target *filesize* of 350MiB per episode.
02:20:11 <pikhq> (so you can back up your downloads on CDR)
02:40:39 <Sgeo_> Anyone know what language it is? The moron in #lisp who posted it thinks its Smalltalk, but that makes no sense.
02:40:40 <Sgeo_> http://paste.lisp.org/display/128148
02:43:38 <elliott> i'll answer if you paste an irc log :P
02:45:26 <Sgeo_> http://pastie.org/private/8nasfxrmidggvcyts3yna
02:45:39 <Sgeo_> Although I don't need the answer, my confidence that it is Ruby is high
02:45:48 <Sgeo_> <vortixo> it is smalltalk!
02:46:09 <elliott> it's almost-valid Ruby (needs a "proc" or "lambda" before the {); I think there's also a embedded scripting language that nicks the lambda syntax from Ruby in which it might be fully valid
02:46:35 <elliott> <maxm-> probably asked on #haskell and they trolled him to come here
02:47:13 <elliott> Sgeo_: Anyway, {} is array literal in many Smalltalks, I think.
02:47:19 <elliott> And || declares local variables, not arguments.
02:47:24 <elliott> I doubt his code is even syntactically valid.
02:47:51 <elliott> I wonder if it's the same Fare.
02:49:53 <Sgeo_> (Ok, I shouldn't make fun of that)
02:51:07 <Sgeo_> elliott, he is calling syntax "sintaxis"
02:51:15 <Sgeo_> (The conversation is ongoing)
02:54:17 <elliott> oerjan: did you notice that guy archiving a talk page after only one section and a few days...
02:54:17 <oerjan> top of the morning to you
02:54:34 <oerjan> yes. and he didn't even use your superb archiving system.
02:55:29 <elliott> which makes it harder for me to revert it
02:55:35 <elliott> and impossible for non-admins :P
02:55:58 <elliott> he also edited my comment before reverting that.
02:56:09 <elliott> i have half a mind to block him for reminding me of dagoth ur
03:00:23 <oerjan> well his language is better designed than _that_. just not very innovative or usable.
03:01:07 <elliott> yes, it's the non-language behaviour :P
03:02:17 <oerjan> i suppose he did add conditionals and looping after you asked
03:02:51 <elliott> yes. and helpfully striked out parts of my comment as he added them.
03:05:21 <Jafet> * elliott is now known as nerevarine
03:19:32 <Sgeo_> elliott, new paste from the person: http://paste.lisp.org/display/128149
03:20:43 <Sgeo_> <vortixo> it is a programming language called supercollider
03:20:43 <Sgeo_> <vortixo> it is based on smalltalk
03:21:32 <elliott> Sgeo_: i demand a complete log :'(
03:22:36 <Sgeo_> http://pastie.org/private/h1uxbqpnrur7yarkxbupa
03:23:57 -!- azaq23 has quit (Quit: Leaving.).
03:24:22 -!- azaq23 has joined.
03:30:55 <elliott> Ah, yes, it's the same Fare.
03:31:06 <kallisti> xscreensaver has pretty cool screensavers
03:31:11 <kallisti> I wish it had an electricsheep plugin or something.
03:31:48 <elliott> I believe you can run arbitrary programs as hacks, though I'm not sure.
03:32:29 <elliott> I mean use the electricsheep executable as a hack.
03:32:46 <elliott> xscreensaver screensavers are called hacks
03:32:54 <elliott> sudo apt-get install electricsheep gets you a program
03:33:03 <elliott> "It should configure itself to be your screensaver, but you can also run it from the command line just by typing "electricsheep". You can also use "electricsheep-preferences" to configure it."
03:33:09 <elliott> So it might even work with xscreensaver, going based on "configure itself"
03:33:41 <kallisti> considering I already have electricsheep
03:33:49 <kallisti> maybe it's in xscreensaver-demo and I missed it or something
03:34:48 <elliott> You could just combine electricsheep and a lock program.
03:35:05 <kallisti> ...well, I like xscreensaver screen savers too
03:36:08 <elliott> I think there are programs that rotate a series of other programs to be screensavers and lock; you could rotate a non-locking xscreensaver and electricsheep.
03:36:16 <elliott> Although electricsheep would run a lot more often than each individual hack, obviously.
03:36:55 <kallisti> also that would fix the problem that I don't like xscreensavers lock window. :>
03:37:58 <elliott> http://www.jwz.org/xscreensaver/toolkits.html
03:38:26 <kallisti> I suspected xscreensavers man page might be lacking in details.
03:38:43 <elliott> http://www.jwz.org/xscreensaver/faq.html
03:40:56 <kallisti> "Where does the text for those silly quotes come from" heh
03:41:44 <ion> electricsheep doesn’t work with xscreensaver despite coming with /usr/share/xscreensaver/config/electricsheep.xml?
03:42:45 <kallisti> I just can't find the option to turn it on.
03:43:42 <ion> I’ve only used it with gnome-screensaver, it’s been like 10 years since i used xscreensaver directly. I’m afraid i’m of no help.
03:44:04 <kallisti> I probably just need to configure electricsheep or something.
03:44:52 <kallisti> there's a "video driver" option that's currently blank.
03:44:56 <kallisti> I have no clue what it wants there.
03:45:59 <ion> It’s an optional parameter to mplayer, it can be left blank.
03:46:39 <elliott> kallisti: Are you sure it's not listed in xscreensaver-demo?
03:46:46 -!- TeruFSX has quit (Read error: Connection reset by peer).
03:47:10 <kallisti> (still getting used to "focus follows mouse")
03:47:52 <ion> Here’s a hack that downloads all the sheep (gigabytes of stuff) from archive.org somewhat quickly. It was a one-shot program, so it has my home directory hardcoded. https://gist.github.com/1369649
03:47:54 <kallisti> assuming it's named something sensible like "Electric Sheep". yes. let me scroll through the whole list.
03:48:14 <elliott> kallisti: You can use focus-follows-click with xmonad.
03:48:21 <kallisti> elliott: not sure if I want to.
03:48:32 <kallisti> I just accidentally type things in other windows due to not being accustomed to it.
03:48:50 <ion> elliott: HXT made me do it. :-(
03:48:56 -!- TeruFSX has joined.
03:49:29 <ion> Also, dunno whether it has suffered from bitrot during the past few months.
03:50:32 <elliott> Hmm, for xscreensaver all i had to do was to add this line to my ~/.xscreensaver, right below "programs:"
03:50:32 <elliott> to make it appear in xscreensaver-demo
03:50:48 <elliott> <ion> electricsheep doesn’t work with xscreensaver despite coming with /usr/share/xscreensaver/config/electricsheep.xml?
03:50:52 <elliott> these xml configs might be a gnome thing or something.
03:50:58 <elliott> Since it has its own configumarator.
03:51:03 <ion> Huh, interesting
03:51:11 <elliott> i can't see jwz using xml :)
03:51:21 <elliott> in preference to mork perhaps :P
03:51:54 <ion> I wouldn’t put it past them to shove their stuff in /usr/share/xscreensaver. :-P
03:51:59 <kallisti> what the heck kind of line seperator is \n\
03:52:32 <kallisti> all the program entries have it.
03:52:51 <ion> \bar" == "foobar"
03:53:07 <ion> Oh, not Haskell?
03:54:23 <ion> I faintly remember the xscreensaver config file having a weird syntax.
03:54:51 <kallisti> yes all the program entries end with \n\
03:59:58 -!- cswords_ has joined.
04:00:38 <oerjan> kallisti: well that is a natural way of doing it in some languages
04:01:07 <oerjan> \n for the newline and \ for line continuation
04:01:37 * oerjan cannot really believe kallisti doesn't know that.
04:03:28 -!- cswords has quit (Ping timeout: 276 seconds).
04:05:48 <kallisti> sure, I know about those two things. but they don't make any sense put together in this context.
04:07:20 <elliott> well if all settings are one line...
04:07:28 <elliott> would be how you do a three-line setting
04:08:10 <kallisti> man electric sheep is so cool.
04:08:13 <kallisti> I could stare at it for hours.
04:22:37 <Sgeo_> http://paste.lisp.org/display/128150 I feel like this is ugly somehow
04:26:17 <kallisti> probably because of all the lisp.
04:26:47 <kallisti> but there are different ways to look for beauty in code.
04:27:05 <kallisti> a lot of Lisp code is beautiful for its structure.
04:27:24 <kallisti> and not quite so much because of the arrangement of syntax.
04:27:43 <kallisti> ...though the huge expansion on the arithmetic stuff looks kind of weird. you probably don't need that much indentation there.
04:28:44 <kallisti> like the (- (* b b) (* 4 a c)) could be on one line
04:29:02 <kallisti> also, no exponentiation operator?
04:30:13 <Sgeo_> I don't know it offhand and was lazy
04:30:49 <kallisti> assuming this is common lisp I think it's expt
04:32:09 <kallisti> but yeah anyone in the habit of reading lisp isn't going to need that much indentation on the arithmetic
04:34:38 <kallisti> makes these occasional audible clicks
04:34:53 <kallisti> it seems louder than it should be
04:34:56 <kallisti> since usually it's pretty silent
04:35:01 <kallisti> any idea what that's all about?
04:35:04 <oerjan> kallisti: hey mine too. i think.
04:35:48 <oerjan> two clicks just since i said that.
04:36:10 <oerjan> i probably should make another backup soon, it's been a few years.
04:36:27 <kallisti> I make a backup like every few weeks....
04:36:52 <kallisti> I think I'll make that a thing.
04:37:03 <oerjan> (anything interesting ends up in my web directory on nvg servers anyway.)
04:37:13 <kallisti> yeah I only have this laptop atm.
04:37:26 <kallisti> has all of my work source code.
04:37:35 <kallisti> I guess I could use GitHub or something.
04:37:41 <elliott> kallisti: yes, that's worrying
04:39:10 <kallisti> and, being the perfect human being I am, there's no way I could misconfigure my Debian installation and horribly destroy everything on the file system.
04:43:41 <kallisti> elliott: someone was trying to tell me that I could take root from my virtual box installation of Debian, and replace everything on my Ubuntu partition with it.
04:43:53 <kallisti> ....and I was pretty sure that wouldn't work.
04:44:07 <kallisti> (re: the thing about "copying root" earlier)
04:44:32 <kallisti> I mean, it could work, but there's so many ways it could go wrong.
04:45:16 -!- pikhq_ has joined.
04:47:42 <elliott> kallisti: yeah it's just not a good idea
04:48:00 -!- pikhq has quit (Ping timeout: 252 seconds).
04:49:36 -!- MoALTz has quit (Ping timeout: 240 seconds).
04:50:03 -!- MoALTz has joined.
04:50:04 <kallisti> hm, does virtualbox use special drivers?
04:50:07 <kallisti> or does it emulate other drivers?
04:52:07 -!- PiRSquared has changed nick to USERHOST.
04:52:41 * Sgeo_ would like to use XenClient on his computer.
04:54:10 <oerjan> it had been more years than i thought...
04:54:21 <Sgeo_> Instead of Common Lisp?
04:55:02 <Sgeo_> I took a Perl class once.
04:55:08 <Sgeo_> I don't think I want to touch it again.
04:55:29 <kallisti> I don't even know how you would go about teaching someone perl.
04:56:02 <kallisti> they probably skip to OO quickly or something
04:56:11 <elliott> oerjan: how many years was it TELL ME
04:56:40 <kallisti> Sgeo_: but no perl's good man. I promise.
04:56:52 <kallisti> official kallisti approval stamp.
05:01:06 <elliott> oerjan: you last backed up in 5 CE? wow.
05:01:46 <elliott> if you said 5, I would have said the exact same thing
06:00:33 -!- MDude has changed nick to MSleep.
07:16:12 <kallisti> electricsheep is dumping .avis into my home directory
07:16:17 <kallisti> there's some in .electricsheep as well.
07:16:46 <kallisti> by default firefox's download setting for zip files is to unzip them automatically
07:19:10 <kallisti> elliott: uh, what is the topic from.
07:22:04 <monqy> it is a good topic
07:25:10 <zzo38> Just some questions; what things you disliked and/or would be designed differently by you, in Haskell?
07:25:24 <zzo38> (And other programming languages too)
07:26:15 <monqy> haskell not being perfect really frustrates me
07:26:29 <monqy> if i designed haskell differently, it would be less not perfect.
07:26:31 <Sgeo_> Is running forever really undefined behavior? Is running out of memory? You could make a language that does a defined-by-the-language thing upon hitting a memory limit. Also note that given a memory limit, something with more memory could determine whether it infinite loops or not.
07:27:03 <zzo38> monqy: Well, what, in your opinion, would be closer to perfect?
07:32:12 <pikhq_> Sgeo_: In what context are you referring to "undefined behavior"?
07:32:29 <pikhq_> Sgeo_: And please tell me that's a quote from someone else, or else I will hit you
07:32:47 <Sgeo_> pikhq_, response to topic?
07:33:59 <Sgeo_> Erm, as in my response
07:34:41 <pikhq_> Consider yourself hit.
07:40:02 <zzo38> OK, but not right now.
07:42:08 <elliott> (Deletion log); 07:41 . . Ehird (Talk | contribs | block) deleted "Talk:Basic Input/Output Commander/Archive 1"
07:42:08 <elliott> (diff | hist) . . Talk:Basic Input/Output Commander; 07:41 . . (+843) . . Ehird (Talk | contribs | block) [rollback]
07:45:14 <elliott> oerjan: i have this awful feeling i'm going to be considering the stux incident completely reasonable within months.
07:45:30 <oerjan> what was the stux incident again
07:46:16 <elliott> http://esolangs.org/wiki/User_talk:Stux#Incident
07:46:31 <elliott> (the "junk pages" were the year categories, now reinstated.)
07:48:15 <zzo38> Do you agree with me that blackjack is a single player game? (Although can also be played with multi players)
07:56:15 <elliott> oerjan: why doesn't GHC do polymorphic unpacking :(
07:56:42 <oerjan> because polymorphic functions don't get passed actual types
07:57:20 <oerjan> i _suppose_ it might be put in the GC data...
07:57:37 <oerjan> (the necessary size information)
07:57:45 <elliott> oerjan: well all it has to do is duplicate a type with unpacked polymorphic fields whenever it's used monomorphically
07:58:10 <elliott> oerjan: or heck, just turn it into a type family, and add an instance (with the same fields and all, just different constructor namse) whenever you give it a parameter it hasn't seen before
07:58:52 <oerjan> erm the point is you need there to be an actual polymorphic implementation in some cases...
07:59:20 <elliott> but this would work most of the time
08:01:41 <oerjan> you might want to ask an actual ghc implementor.
08:04:07 <elliott> oerjan: well, augustss has said that it's known how to do it, they just didn't bother.
08:06:56 <elliott> oerjan: i wouldn't be surprised if you could do it in all cases in haskell 98
08:07:29 <kallisti> elliott: I feel special. Today I taught an #xmonad regular that join (fmap f m) = f >>= m
08:08:35 <oerjan> elliott: you _can_ do newtype Fix t a = Fix (t (Fix t a)) in H98. i think that would be a problem...
08:09:40 <oerjan> hm that a looks unused. anyway.
08:10:27 <oerjan> (if t unpacks its type argument.)
08:11:24 <elliott> oerjan: point. well, you can go simpler than that
08:11:40 <elliott> otoh the simpler could recognise that Void is 0 bytes and solve the equation ;)
08:12:06 <elliott> newtypes are _already_ unpacked.
08:12:07 <oerjan> elliott: yes, my point is you could make a generic Fix and pass it a t which is independently defined to be unpacked.
08:14:30 -!- oerjan has quit (Quit: CU).
08:14:43 <kallisti> elliott: so there's some work going on to make the xmobar config into a compiled Haskell program rather than a parsed format.
08:14:55 <kallisti> and to in general improve the features.
08:15:23 <kallisti> for one it would be nice to work with dbus for things like a system tray and notifications.
08:15:31 <kallisti> instead of using trayer seperately.
08:16:02 <elliott> what the fuck is in your tray and why do you use it
08:16:56 <elliott> just don't add one for fuck's sake
08:17:36 <kallisti> dropbox would be good. nm-applet is nice.
08:17:56 <kallisti> ...uh, I'm sure there are others?
08:18:36 -!- atehwa has joined.
08:18:46 <elliott> Why do you need a dropbox icon? And nm-applet is eternal pain and living hell.
08:18:55 <elliott> Well, NetworkManager in general is.
08:19:14 <Jafet> Get a desk job if you enjoy using trays
08:19:25 <kallisti> strong opinions on system trays here, I see.
08:19:43 -!- derdon has joined.
08:20:11 <Jafet> I use the Windows tray every few minutes
08:20:24 <Jafet> Because there are many Windows programs that make the tray useful
08:20:33 <kallisti> also indicators in general are a useful UI feature.
08:20:42 <Jafet> Gnome, on the other hand...
08:20:54 <kallisti> for example, why should I have a volume bar constantly on when I can have an indicator appear only when I'm changing the volume?
08:21:15 <kallisti> of course I can go without either (and currently am), but the visual feedback is nice.
08:22:05 <kallisti> as long as we're talking about necessary here...
08:22:06 <elliott> kallisti: You have the fucking current temperature on your bar!
08:22:12 <elliott> Replace that with the volume.
08:22:32 <kallisti> as long as we're talking about necessary here...
08:22:37 <kallisti> nothing on xmobar is necessary
08:22:42 <kallisti> all of the information can be derived from elsewhere
08:22:59 <elliott> Oh, no, just memory, CPU, network, and a really awful battery indicator.
08:23:23 <elliott> Anyway, the notification daemons all suck.
08:24:00 <elliott> Jafet: I find Gnome 2's tray much more useful than Windows'.
08:24:07 <elliott> Probably because Gnome 2 puts almost nothing in it.
08:24:34 <kallisti> elliott: ideally the battery indicator would only show AC when there's AC power
08:24:52 <kallisti> but the current xmobar doesn't have that in its default battery monitor. so I need to write the script to do it myself.
08:25:11 <elliott> If you're going to go to those lengths, just switch to dzen2 and use fizzie's thing: https://github.com/fis/dot-xmonad
08:25:14 <kallisti> it should only show time left when there's no AC power.
08:25:54 <elliott> Time left estimations are almost wrong
08:25:58 <elliott> Just display a percentage.
08:26:15 <kallisti> I think because my battery is broken
08:26:17 <kallisti> but I don't know why for sure.
08:31:26 <Jafet> elliott: the fact that you could remove it entirely also helped
08:39:48 <kallisti> elliott: I've been told that dzen is bad at right-aligning?
08:44:12 <zzo38> Is Fermat's Last Theorem provable in TNT?
08:45:50 <pikhq_> elliott: I don't think I would blame Windows for the tray being so cluttered.
08:46:02 <pikhq_> Out of the box, it's got hardly anything in it.
08:46:22 <pikhq_> Unfortunately, every single damned Windows dev seems to think their program is *special*.
08:46:56 <olsner> for some meaning of special that's probably true
08:47:14 <pikhq_> ... Aaand Windows comes out-of-the-box relatively sparse, so you install a lot of programs in the course of making it usable.
08:47:19 <zzo38> On my computer the only icon in Windows system tray is the volume icon
08:47:31 <pikhq_> zzo38: I am not even slightly surprised.
08:48:57 <elliott> <kallisti> elliott: I've been told that dzen is bad at right-aligning?
08:49:05 <elliott> fizzie figured out how! But the standard thing is just to run two of 'em.
08:49:34 <elliott> At least as of $YEARS ago.
09:05:52 -!- derdon has quit (Remote host closed the connection).
09:07:29 <fizzie> The right-aligning also doesn't exist in the released versions, just in the SVN.
09:08:05 <fizzie> Or at least that was the case.
09:10:22 -!- Frooxius has quit (Ping timeout: 252 seconds).
09:13:32 -!- monqy has quit (Quit: hello).
09:21:43 <zzo38> What comonads can you make on the IO monad Kleisli category?
09:24:13 -!- zzo38 has quit (Remote host closed the connection).
09:31:27 <kallisti> elliott: I have a bunch of texlive fonts
09:33:59 <kallisti> ...nevermind I think I found out how to do it
09:34:06 <kallisti> the tedious part will be finding all of the font directories.
09:34:58 <kallisti> ... I installed texlive-fonts-extra
09:35:42 <kallisti> wow... textlive-full is a 1.1 GB download
09:41:58 <elliott> kallisti: They're unlikely to be in the right formats.
09:42:07 <elliott> Computer Modern, f.e., is Metafont.
09:42:40 <shachaf> I never metafont I didn't like.
09:43:05 <kallisti> elliott: https://wiki.archlinux.org/index.php/TeX_Live#Fonts
09:43:41 <elliott> They're mostly crap. Why do you want them in fontconfig?
09:43:56 <kallisti> because I have a shortage of fonts.
09:51:17 <kallisti> elliott: so by putting .conf files in /usr/share/fonts
09:51:27 <kallisti> fontconfig can automatically slurp all of those up?
09:51:31 <kallisti> or do I need to tell it I've added a new file.
09:51:43 <oklopol> can you really even define "emacs"
09:54:12 <oklopol> i should probably start doing this in #philosophy
09:54:41 <kallisti> so fc-cache doesn't seem to like my fancy new config file.
09:54:48 <kallisti> or isn't caring about it or something.
09:54:55 <oklopol> deep is when someone says something that's very deep
09:55:26 <oklopol> i learned how to define at urbandictionary.com
09:55:37 <fizzie> kallisti: I would guesstimate that config files go to /etc/fonts/conf.d/ or something, unless of course your system is different. (This is an Ubuntu workstation.)
09:56:03 <elliott> fizzie: Please, that directory actually does something.
09:56:15 <elliott> If he keeps putting them in there, he'll be kept from fucking things up.
09:56:24 <elliott> Especially since there's no chance he'll get the load order right in conf.d.
09:56:25 <kallisti> this is what I get for getting help from the Arch wiki on a Debian machine.
09:56:32 <elliott> Or put them in conf.avail before symlinking to conf.d.
09:56:37 <elliott> Arch has /etc/fonts/conf.d too.
09:57:04 <fizzie> elliott: Who am I to forbid people from e.g. getting the "chemarrow" font in fontconfig, for some hot "chemical-reaction arrow on LibreOffice" action.
09:57:31 <kallisti> they say to make /etc/fonts/texlive.conf
09:57:41 <elliott> Whatever they say, it's a supremely dumb idea.
09:58:02 <elliott> For example, the Type1-or-was-it-TTF version of Computer Modern is something like 30 fonts, I believe.
09:58:15 <elliott> They are not designed to be used as main system fonts.
09:58:19 <elliott> Install some ttf- packages instead.
09:59:47 <fizzie> https://bbs.archlinux.org/viewtopic.php?id=108670 "Personally, I quickly regretted the decision to make all the TeX fonts available to fontconfig, --"
10:00:42 <mroman> there's volume control in xmonad?
10:00:52 <elliott> kallisti doesn't listen to advice, he just does the stupid thing and acts indignant when it goes wrong and you point out you told him so.
10:01:05 <shachaf> elliott: HLEP WHY IS MY LAPTOP OVERHEATING
10:01:54 <elliott> fizzie: Heh, so it completely breaks XeLaTeX.
10:02:02 <kallisti> "merely because the huge number of them (at least for a full TeXlive install) made using any font selection dialogue overwhelming."
10:02:07 <kallisti> these are the words of lesser men.
10:02:43 <elliott> And who need to be able to use LaTeX.
10:03:09 <elliott> Anyway, we've entered stage #3 of your terrible ideas, "sarcastically dismissing all objections"; roll on stage #4.
10:04:08 <kallisti> dude, texlive-fonts-extra only took like 50 minutes to install. What's the problem?
10:06:16 -!- Vorpal has joined.
10:06:25 <kallisti> image having a different typeface for every character....
10:07:39 <fizzie> I have a rather horrible font selection dialogs too, because of $ dpkg-query -l 'ttf-*' | grep ^i | wc -l
10:07:48 <fizzie> I just kept pressing + there.
10:08:14 <kallisti> there doesn't seem to be that many ttf-* packages on Debian...
10:08:37 <fizzie> Now there's a dozen fonts that only differ in what sort of "funky" symbol they put in the middle of all 'o's; peace signs, four-leaf colvers, and so on.
10:08:59 <kallisti> I just want some more old style fonts. is that too much to ask?
10:09:50 <fizzie> That was like the best idea.
10:09:54 <Vorpal> Are the pedals in cars mirrored in countries where you drive on the left side compared to countries where you drive on the right side?
10:10:21 <Jafet> Little-known fact: left-side driving countries have more left-handed people
10:10:29 <Vorpal> Jafet, kind of doubt that.
10:10:39 <Vorpal> but I guess it is possible..
10:10:59 <fizzie> Vorpal: As far as I know the arrangement tends to be the same, just the seat with the controls changes.
10:11:00 <Jafet> They're not, silly
10:11:09 <kallisti> fizzie: where are you getting all of these font packages?
10:11:23 <fizzie> kallisti: It's an Ubuntu system, you see.
10:11:43 <kallisti> why did I install Debian again?
10:11:47 <fizzie> They need to have all the "used by people in flyers of their 'band'" fonts available in order to be competitive.
10:12:11 <kallisti> I want... reasonable normal fonts. my font selection is scarce.
10:12:32 <Vorpal> anyway, the reason I was thinking about this, is that if you for some reason can't drive the car but need to move it forward (and the ground is fairly flat) you can (if the clutch pedal is outermost) just put one foot in the car to press the clutch and move it by hand. Doesn't need a separate person doing the pushing, and is safer than putting the car in neutral if the ground isn't flat enough.
10:12:41 <fizzie> 90-plus percent of the 1795 are the crap.
10:12:52 <Jafet> What's a clutch pedal
10:12:56 <fizzie> And by "crap" I mean real crap.
10:13:12 <kallisti> you see, there's this thing in manual transmissions called the clutch
10:13:15 <Jafet> Is it a pedal on cars that often have to be pushed
10:13:21 <Vorpal> my point is, it would be hard to do this if you didn't have the clutch pedal nearest to the door
10:13:22 <kallisti> that connects the transmission with the engine
10:13:35 <Jafet> Or a pedal in countries without bricks
10:14:04 <kallisti> and the clutch pedal, disconnects this connection briefly, so that you can remove the coupling between engine RPMs and transmission RPMs as you abruptly change the transmission RPMs
10:14:11 <elliott> fizzie: Can see your worst font?
10:14:29 <Vorpal> Jafet, I presume you don't have a driving license? :P
10:14:56 <fizzie> Vorpal: No, e just carries a brick around at all times in order to depress the clutch.
10:15:43 <Jafet> Yes, I use cars that stop working all the time
10:15:55 <fizzie> elliott: Sowwy, but I'm at work, so I don't have them here. Anyway, the one that was called "Hots" that was made of revealingly dressed girls making different shapes that did not even correspond to the letters was probably quite bad.
10:16:02 <Jafet> The brick is also essential for hurling at other cars
10:16:18 <Vorpal> fizzie, ... why did you have that font?
10:16:45 <fizzie> Vorpal: I installed most things that started "ttf-" as a lark.
10:16:54 <Jafet> fizzie has downloaded a large font collection
10:16:56 <Vorpal> fizzie, which distro had this font?
10:17:07 <kallisti> fizzie, like myself, understands that it's important to have all the fonts.
10:17:36 <fizzie> $ dpkg-query -S `locate hots.ttf`
10:17:36 <fizzie> ttf-larabie-uncommon: /usr/share/fonts/truetype/ttf-larabie-uncommon/hots.ttf
10:17:38 <elliott> fizzie: That's the bestest font.
10:17:49 <fizzie> "Less common freeware TrueType fonts from Ray Larabie. This package contains fonts which are beautiful for special decorations and headlines."
10:17:57 <elliott> fizzie: You should set your IRC client to use hots.ttf.
10:17:59 <fizzie> "Special decorations" indeed.
10:18:06 <Vorpal> three ttf fonts are enough: one sans serif, one serif, and one monospace. Then you need one bitmap font as well.
10:18:13 <elliott> Total shoulder-glancing protection.
10:18:14 <Vorpal> then the fallbacks and such will work
10:18:21 <elliott> You don't need bitmap font.
10:18:34 <fizzie> elliott: You can see it in action at http://www.fonts4free.net/hots-font.html
10:18:45 <Vorpal> elliott, sure you do, for your terminal. ttf fonts don't work in the vt
10:18:48 <kallisti> Vorpal: no all serif fonts are created equally.
10:19:04 <Vorpal> kallisti, if you took me seriously you missed the whole point
10:19:35 <elliott> fizzie: http://www.fonts4free.net/custom.php?owntext=%3Celliott%3E%20*fonts&ttf=hots.ttf
10:19:38 <Jafet> It could be used for diagramming cheerleading moves
10:20:15 <fizzie> elliott: Oh, they have the letters in their shirts.
10:20:17 * elliott wonders what classes as "revealing" in Finland.
10:20:27 <elliott> I was wondering what that was.
10:20:28 <fizzie> elliott: I couldn't see that in the font selection dialog, it's pretty small.
10:21:16 <kallisti> elliott: oh no, I actually do have Courier as well as Courier New
10:21:50 <fizzie> Anyway, http://www.larabiefonts.com/ recommends the Ultimate Font Download, it has 10000 fonts.
10:21:59 <fizzie> "All the free Larabie Fonts are included and a lot more. It’s the only big font download pack I recommend."
10:22:58 <elliott> I wonder if all creators of big font download packs say the same.
10:23:11 <elliott> "This one is good -- but, you know, the other ones are too."
10:23:27 <fizzie> $ dpkg-query -S `locate $(fc-match 'Hippy Participants' | cut -d: -f1)`
10:23:28 <fizzie> ttf-larabie-uncommon: /usr/share/fonts/truetype/ttf-larabie-uncommon/hippp___.ttf
10:23:37 <fizzie> It's the same guy's fault.
10:23:43 <kallisti> I'm guessing I don't want "ttf-opensymbol"
10:23:48 <fizzie> I suspect that's where most of the stupidity is.
10:24:08 <fizzie> Anyway, there's all kinds of "binary" fonts, and one composed of just sequences of cigarettes called "Holy Smokes" or something, and so on...
10:24:29 <elliott> fizzie: Have you considered: Removing them?
10:24:44 <fizzie> Unsurprisingly, Holy Smokes comes from ttf-larabie-uncommon.
10:24:59 <kallisti> removing them? that's counterintuitive
10:25:04 <kallisti> that would involve having less fonts.
10:25:26 <fizzie> /usr/share/fonts/truetype/ttf-larabie-uncommon/skeletor.ttf
10:25:32 <fizzie> I'm sure that's one of the best things.
10:25:40 <fizzie> Or the "stupefac.ttf".
10:25:48 <fizzie> Or the "worthles.ttf".
10:26:40 <kallisti> oh wait I think I found a good old style font.
10:27:12 <Vorpal> fizzie, are all those 10000 fonts novelty ones? Or are there "proper" ones as well? :P
10:27:29 <fizzie> Vorpal: I don't know, but it's the Ultimate Font Download.
10:27:45 <fizzie> ii ttf-larabie-deco 1:20011216-1.1 Decorative fonts from www.larabiefonts.com
10:27:49 <fizzie> ii ttf-larabie-straight 1:20011216-1.1 Straight fonts from www.larabiefonts.com
10:27:52 <fizzie> ii ttf-larabie-uncommon 1:20011216-1.1 Special decorative fonts from www.larabiefonts.com
10:27:55 <fizzie> I suppose the "straight" fonts are "proper".
10:28:37 <fizzie> 161 fonts (well, .ttf files) in straight, 161 in deco, 149 in uncommon.
10:28:45 <Vorpal> Not sure what they mean by straight there...
10:29:45 <kallisti> http://www.ultimatefontdownload.com/index.htm?hop=typodermic
10:30:04 <Vorpal> how do you search in installed packages on ubuntu? As in just getting the package names matching something and that are installed
10:30:48 <fizzie> Vorpal: Anyway, you can see probably most of the Larabie fonts at http://www.myfonts.com/foundry/Larabie/ --
10:31:08 <elliott> Vorpal: dpkg --list | grep ttf-
10:31:13 <fizzie> Vorpal: And "dpkg-query -l '*foo*' | grep ^i" is what I've done, though I'm sure there's a better way.
10:31:47 <fizzie> "dpkg --list" seems to equal "dpkg-query -l"
10:32:13 <Vorpal> dpkg-query -l 'ttf-*' | grep ^i | awk '{print $2}'
10:32:29 <Vorpal> hm that is a lot of packages.
10:32:52 -!- azaq23 has quit (Quit: Leaving.).
10:32:53 <elliott> http://stackoverflow.com/questions/1645615/what-is-the-use-of-brainfck
10:33:14 <elliott> kallisti: Perhaps more if you enabled the other repos.
10:33:19 <Vorpal> why do I have fonts for languages I can't read installed
10:33:24 <Vorpal> might as well save some disk space
10:33:34 <elliott> Vorpal: So Wikipedia pages aren't littered with boxes?
10:33:35 <kallisti> that's how I have wireless and such
10:34:05 <Vorpal> elliott, uh? en.wikipedia.org and sv.wikipedia.org are the only relevant ones anyway?
10:34:27 <fizzie> aptitude search '~i ttf-' would have been an alternative.
10:34:31 <Vorpal> I wonder what IRC would look like in CM btw...
10:34:38 <kallisti> elliott: also the use of brainfuck is to increment and decrement bytes in an unbounded tape.
10:35:05 <elliott> Vorpal: To suggest that foreign scripts never appear on Wikipedias is pretty highly ignorant.
10:35:24 <Vorpal> elliott, well of course it does, but to what degree depends on which articles
10:35:36 <elliott> Not to mention the various recontextualised uses of them -- e.g. see that Kannada letter in the look of disapproval, and so on.
10:36:01 <elliott> There are easier ways to save disk space than to decrease the literacy of your system.
10:36:01 <Vorpal> anyway doesn't dejavu have most of them
10:36:22 <Vorpal> I don't need fonts dedicated to those letters if I have them in dejavu.
10:36:59 <fizzie> elliott: Oh yes, http://www.myfonts.com/fonts/larabie/ennobled-pet/ was pretty bad too. But maybe not quite "Hots" bad. And, uh, http://www.myfonts.com/fonts/larabie/oil-crisis/
10:37:40 <elliott> fizzie: Oil Crisis: best IRC font?
10:38:02 <Vorpal> what are the points of those fonts? Wouldn't just using a vector graphics clipart be better? The effect would be pretty much the same
10:38:20 <elliott> fizzie: I think you should /kick Vorpal for blasphemy.
10:38:44 <kallisti> it's stylistic scripts for people who can't create custom graphics.
10:38:51 <kallisti> they're all quite specific in their uses.
10:39:04 <Vorpal> kallisti, clipart is for people who can't create custom graphics too!
10:39:19 <fizzie> Then Sappy Mugs (just people and fish with silly faces), Riot Act (made out of monsters, I *think*), Lucky Ape (slot machine slots), ...
10:39:54 * elliott hovers his delete-finger over [[Grasp]] -- uh, [[User:Fizzie]].
10:39:57 <kallisti> should I install Mathematica fonts?
10:40:38 <kallisti> I think the answer is obviously YES
10:40:41 <fizzie> I seem to recall them being utterly utterly useless without, you know, Mathematica.
10:40:51 <Vorpal> kallisti, I have them, because I have mathematica
10:41:05 <elliott> http://esolangs.org/wiki/Main_Page?uselang=qqx -- this is the hardcore.
10:41:08 <kallisti> maybe I've been living in a cave.
10:41:16 <Vorpal> kallisti, I said I have it, not that I use it (very much)
10:41:52 <Vorpal> hm I wonder how much of the stuff in the mathematica fonts lacks a standarlised code point in unicode.
10:42:02 <Vorpal> there are quite a lot of math symbols there.
10:42:34 <fizzie> I don't think they have the corresponding Unicode code point information even in the cases where it would exist.
10:42:40 -!- kallisti_ has joined.
10:43:56 -!- kallisti_ has quit (Client Quit).
10:44:01 <Vorpal> unicode even have code points for stuff like 𝕀
10:44:16 <Vorpal> (the letter "I" in blackboard bold)
10:44:29 <kallisti> I lost my irssi window, but it was still attached according to screen...
10:45:06 <kallisti> also it's ttf-mathematica so it should be usable elsewhere
10:45:12 <kallisti> but it looks like a shitty font.
10:46:01 <elliott> I believe it just assigns symbols to random codepoints.
10:46:18 <kallisti> it's a nice semi-calligraphic kind of script.
10:46:30 <kallisti> ...you know because I totally write formal invitations.
10:47:04 <elliott> Anyway, if you want an OS with a lot of nice serifs, you're just going to have to use OS X.
10:51:50 <kallisti> there are free versions of Garamond out there, but I doubt they're as high quality.
10:52:20 -!- hagb4rd has quit (Ping timeout: 276 seconds).
10:52:53 <elliott> kallisti: You'll have to convert the dfont files, and since OS X fonts have no hinting (since OS X doesn't hint), they'll look like crap.
10:52:57 <elliott> Especially since freetype sucks.
10:53:04 <elliott> The free Garamonds are okay.
10:53:23 <fizzie> Install all the free Garamonds and take their average to get an even better approximation of the real thing.
10:53:25 <elliott> A new free Garamond came out recently, I think.
10:53:43 <elliott> fizzie: Clicked my Avería link, eh?
10:54:40 <kallisti> elliott: wait what? why doesn't it hint?
10:56:35 <elliott> Hinting distorts font shapes in return for crispness. OS X's font renderer is good enough, and their displays high-DPI enough, that they don't need it.
11:02:10 -!- kmc has quit (Ping timeout: 245 seconds).
11:07:44 <Vorpal> elliott, how does .dfont differ from .ttf?
11:07:51 <Vorpal> I mean, why would apple not use ttf
11:08:03 <Vorpal> (they co-developed ttf after all, iirc)
11:08:23 <elliott> "Datafork TrueType is a font wrapper used on Apple Macintosh computers running Mac OS X. It is a TrueType suitcase with the resource map in the data fork, rather than the resource fork as had been the case in Mac OS 9. It uses the file extension .dfont."
11:08:25 <elliott> Oh, but: "In Mac OS X 10.6, released August 28, 2009, the dfont format is being (gradually) replaced with the TTC format (TrueType) [1]."
11:08:32 <elliott> I don't know if freetype can do .otf.
11:08:53 <elliott> http://en.wikipedia.org/wiki/Truetype#Suitcase
11:08:56 <elliott> is the diff between dfont and ttf
11:08:57 <Vorpal> I thought they abandoned resource fork mostly
11:09:15 <Vorpal> but it seems it is still supported on os x
11:09:31 <Vorpal> why though, intel macs can't even run classic
11:09:32 <fizzie> .sfont has the font curves in the salad fork.
11:10:04 <elliott> Vorpal: <elliott> "Datafork TrueType is a font wrapper used on Apple Macintosh computers running Mac OS X. It is a TrueType suitcase with the resource map in the data fork, rather than the resource fork as had been the case in Mac OS 9. It uses the file extension .dfont."
11:10:08 <elliott> i.e., it uses the data fork on OS X.
11:10:21 <Vorpal> elliott, yes indeed, but read what I said
11:10:21 <elliott> Anyway, OS X is still embarrassed by its filesystem.
11:10:32 <elliott> They wanted to go for ZFS.
11:10:36 <elliott> But then Oracle happened and they gave that up.
11:10:51 <Vorpal> didn't they support UFS2?
11:11:24 <elliott> I suspect they'll assign some engineers to make a new "next-gen" (like btrfs, ZFS, etc.) filesystem sometime.
11:11:28 <elliott> They certainly have the expertise and the need.
11:12:12 -!- pikhq has joined.
11:12:15 -!- pikhq_ has quit (Ping timeout: 260 seconds).
11:12:34 <Vorpal> elliott, does OS X just do unix permissions or is there any support for more advanced ACL?
11:12:56 <elliott> I think they're OS X-specific.
11:13:00 <Vorpal> layered on top of HFS+...
11:13:12 <fizzie> There was some sort of UFS thing, I remember reading about it when installing yaboot. But it might be abandoned nowadays.
11:13:29 <elliott> Vorpal: They still develop HFS+.
11:14:01 <Vorpal> is oracle going for zfs or btrfs in the future? Or both?
11:14:07 <elliott> kallisti: Can you register a wiki account so I can ban you for that?
11:14:11 <Vorpal> Maintaining both doesn't seem sensible
11:14:21 <kallisti> elliott: you can think of it as vaciously banned
11:14:32 <elliott> Pronounced "ze buffer FS".
11:14:49 <Vorpal> anyway how old is zfs by now?
11:14:58 <Vorpal> I don't think you could call it "next-gen" any more
11:15:24 <elliott> It's next-gen when everyone is using ext4/whatever the BSDs are on these days.
11:15:41 <elliott> OK, "current gen" might be roughly XFS.
11:16:22 <Vorpal> I think FreeBSD is still on UFS2 with that crazy soft-update stuff instead of journaling?
11:16:36 <kallisti> zfcfs -- Zermelo-Fraenkel set file system.
11:17:07 <elliott> "The page you have tried to view [...] is currently available to LWN subscribers only."
11:17:11 <elliott> I bet fizzie is a LWN subscriber.
11:17:50 <kallisti> I'm about to open GIMP without any of the default xmonad config stuff that makes it sane to use
11:18:28 <elliott> I believe GIMP sets floating hints on its windows.
11:19:44 <kallisti> oh right I'm using the thing with the acronym
11:21:03 <kallisti> I had to configure it in xmonad...
11:21:12 <kallisti> http://hackage.haskell.org/packages/archive/xmonad-contrib/0.9.1/doc/html/XMonad-Hooks-EwmhDesktops.html
11:22:06 <kallisti> so .xpm is a bitmap format right? any way I can set a background that isn't a bitmap?
11:22:30 <elliott> kallisti: just use xsetroot in xsession
11:22:55 <elliott> Oh, you want a full-colour image thing.
11:23:02 <elliott> Install an image viewer and run it in the root window in your xsession.
11:28:26 <kallisti> so now instead of Debian's default spaceship stuff I have BEAUTIFUL PLANTS PREGNANT WITH RAINDROPS.
11:30:18 <kallisti> I wonder how easy it is to randomly change backgrounds occasionally.
11:30:25 <Vorpal> kallisti, what, that doesn't make sense. You surely mean pregnant with seeds or such?
11:30:50 <kallisti> I... okay it may not make sense.
11:30:53 <kallisti> but they have raindrops on them
11:31:03 <fizzie> And 'pregnant' is a very widely-applicable word.
11:31:04 <fizzie> 3. fraught, pregnant -- (filled with or attended with; "words fraught with meaning"; "an incident fraught with danger"; "a silence pregnant with suspense")
11:33:19 <kallisti> should I consider using a file browser?
11:33:23 <Vorpal> what, windows 7 is freaking me out. There are two folders with the exact same name in my "Documents library"
11:33:29 <Vorpal> how does that even work
11:33:53 -!- hagb4rd has joined.
11:34:07 <Vorpal> oh it is a merged view of "shared" documents and "my documents"
11:34:19 <Vorpal> but why not merge those directories too
11:34:39 <Vorpal> I wonder if I can get just plain old my documents in the start menu then
11:34:45 <fizzie> That's not what a library does. I mean, they don't merge the pages of books with similar titles.
11:35:16 <Vorpal> fizzie, on the other hand, when I create a new directory in that "library", how does it decide where it should go
11:35:18 <kallisti> actually this makes it very easy to randomly shuffle background images.
11:35:48 <fizzie> Vorpal: I don't know about that, but there's a "Locations" list that you can edit.
11:35:48 <elliott> <kallisti> should I consider using a file browser?
11:35:56 <elliott> A terminal works just fine. If you must, use ROX Filer.
11:36:20 <kallisti> the main reason I would want to use one is things like images
11:36:28 <kallisti> but I could just learn how to use an image viewer effectively
11:36:30 <elliott> geeqie is good at showing a bunch of images.
11:36:34 <kallisti> gqview is kind of bad though....
11:36:35 <elliott> Deewiant recommended it to me.
11:36:40 <elliott> geeqie is a fork of gqview, I think.
11:37:03 <kallisti> when I open qwview the title say "geeqie"
11:37:04 <fizzie> Vorpal: One of them is supposedly the default.
11:37:10 <fizzie> Vorpal: "Change the default save location. The default save location determines where an item is stored when it's copied, moved, or saved to the library. For more information, see Customize a library."
11:37:26 <Vorpal> anyone know an image viewer/searcher that can search/filter on EXIF data?
11:37:27 <fizzie> It's in the Locations list.
11:37:37 <kallisti> elliott: I may just use feh. it seems pretty good.
11:37:59 <elliott> kallisti: feh doesn't really do directories much, afaik.
11:38:05 <Vorpal> also, it needs to support raw images.
11:38:14 <Vorpal> I would really like a tool like that
11:38:32 <kallisti> elliott: if you pass it a directory
11:38:36 -!- oklofok has joined.
11:38:37 <kallisti> also you can choose sorting options
11:38:52 <Vorpal> kallisti, does it support sorting on EXIF tags?
11:38:53 <kallisti> and... probably a bunch of other stuff.
11:39:12 <Vorpal> (and filtering possibly)
11:39:52 <elliott> I've learned much languages, but now I want to choose one, but the language that I most liked was Haskell, it is like a interpreted language, but is a compiled. Then I want to know the pros and cons of this powerfull language(just to make the correct choice).
11:39:54 <Vorpal> anyway, if you happen to come across a competent image library organizer software, do tell me.
11:40:10 <kallisti> Allowed sort types are: name, filename, width, height, pixels, size, format.
11:40:14 <Vorpal> elliott, ouch the (lack of) grammar
11:40:24 <Vorpal> kallisti, nah, not enough for me
11:40:51 <kallisti> my version is apparently compiled without EXIF support
11:41:04 -!- oklopol has quit (Ping timeout: 255 seconds).
11:41:22 <Vorpal> kallisti, I want to be able to filter and sort on shutter settings and such. (This is very much useful when trying to check through HDR photography)
11:42:02 <Vorpal> kallisti, it doesn't really allow be to browse though
11:42:13 <kallisti> eh just dump everything to feh or something.
11:42:16 -!- kmc has joined.
11:42:38 <Vorpal> kallisti, dump everything of what?
11:42:40 <fizzie> Vorpal: Shotwell is Gnome's new standard image organizer, I suppose you've given it a whirl?
11:42:42 <kallisti> or <insert image viewer that can take a list of files as arguments and allow you to cycle through them in that order>
11:42:50 <Vorpal> fizzie, is it from gnome 3? then no
11:42:58 -!- oklofok has quit (Ping timeout: 252 seconds).
11:43:14 <Vorpal> kallisti, I would like to see thumbnails kind of thing, I believe eog can do that though. Oh well.
11:43:18 <fizzie> Vorpal: It's what they replaced F-Spot with, don't know about Gnome version numbers.
11:43:21 <kallisti> yeah feh can take multiple files, directories, or URLs, apparently.
11:43:22 <Vorpal> but it doesn't show me the EXIF data on the side iirc
11:43:29 <kallisti> and you can specify a zoom size....
11:43:57 <Vorpal> fizzie, well, f-spot was a joke for my purposes
11:44:14 <kallisti> --draw-exif (only if compiled with exif=1) display some EXIF information in the bottom left corner, similar to using --info with exiv2 / exifgrep .
11:44:27 <fizzie> Vorpal: Given that it's a Gnome tool, and the default Gnome tool at that, it's probably more or less oversimplified.
11:44:35 <kallisti> I think feh is your program of choice.
11:44:41 <Vorpal> kallisti, exiv2 bugs out on my files, exiftool works fine though
11:44:46 <Vorpal> at least last I checked it bugged out
11:44:53 <kallisti> Vorpal: that was docs from feh
11:44:58 <Vorpal> didn't properly parse the vendor specific extension
11:45:19 <Vorpal> I think it byteswapped them
11:46:09 <elliott> http://yorba.org/shotwell/
11:46:20 <elliott> It's not GNOME-originated.
11:47:25 <elliott> fizzie: It has replaced F-Spot as the standard image tool for several GNOME-based Linux distributions, including Fedora in version 13[1] and Ubuntu in its 10.10 Maverick Meerkat release.[2]
11:47:44 <Vorpal> I wonder if there is any sort of browsing support in rawtherapee... I should check
11:47:48 <fizzie> Oh, I guess it could be just that distribution-makers replaced F-Spot with it.
11:47:50 <Vorpal> that is quite a nice tool
11:48:17 <kallisti> yeah I think I'll just use feh, since I'm basically migrating entirely to console apps...
11:48:56 <kallisti> elliott: yeah dude, it's the future.
11:49:08 <fizzie> http://redmine.yorba.org/projects/shotwell/wiki/ShotwellFeatureComparison <- bestly formatted comparison table.
11:49:33 <fizzie> I suggest you migrate to aaview.
11:49:42 <Vorpal> hm it does. I need to install it on this laptop, or check later on if it is suitable on my desktop (when I rebooted to linux)
11:49:43 <fizzie> Or cacaview if you want to be "fancy".
11:52:13 <Vorpal> fizzie, err is it just me, or is that page just broken? With some random chars on it "–" for example
11:53:07 <kallisti> elliott: it's a program you use from a command-line with an entirely keyboard-based interface. my bad.
11:53:35 <Vorpal> fizzie, I can't read that "table" at all
11:53:40 <Vorpal> because it isn't a table
11:53:44 <kallisti> imagine that it's not a table.
11:54:14 <elliott> kallisti: You don't "use it from a command-line".
11:54:17 <fizzie> Vorpal: Some of the earlier versions do show the first table, but not the rest.
11:54:20 <elliott> It can be started from a command-line; so can Firefox.
11:54:46 <kallisti> elliott: with feh I'm pretty sure it's a requirement
11:54:56 <kallisti> because it needs options and parameters
11:55:11 <elliott> What do you think happens when you double-click on an HTML file in a file manager?
11:55:35 <kallisti> yes, okay, other programs can launch feh.
11:55:43 <kallisti> I'm not really seeing your point.
11:56:29 <Vorpal> fizzie, I'm looking at rawtherapee's file browser atm. Well trying to. The Ubuntu 10.04 download doesn't run on Ubuntu 10.04 due to missing library (wut)
11:56:37 <Vorpal> I used the program for other stuff previously
11:56:44 <Vorpal> (mostly as a more advanced ufraw)
11:57:01 <elliott> http://esolangs.org/wiki/Minibiatch i like the part where this page has no content and the page it links to purporting to contain the content states "I have personally removed this because Oerjan fucked with it."
11:57:14 <elliott> i like the part where does anybody object to me obliterating that thing
11:57:43 <kallisti> I object to the part where elliott likes the parts
11:58:53 <fizzie> Vorpal: Well, anyway; Shotwell is the only specifically-an-image-organizer app I know of (except silly F-Spot and KDE's digiKam); it does have (some of) the sensible features, like an attempt at raw support and the whole non-destructive edits thing, but presumably the image editing functions are pretty streamlined. (Still, it should be able to do metadata-driven searching right.)
11:59:11 <elliott> More like E-Spot am I right???
12:00:25 <fizzie> elliott: You are more like a DEE-spot. You know. A despot.
12:00:43 <elliott> WELL it was nice knowing [[User:Fizzie]] too bad he's BANNED FOREVER.
12:00:45 <Vorpal> fizzie, so how hard is it to do a search like "exposure time > 1/60 && exposure time < 1/80"
12:01:08 <Vorpal> well, I mixed up those < > but apart from that
12:02:35 <kallisti> wow this is infinitely better than GNOME 2...
12:02:53 <kallisti> I HAVE COMPLETE CONTROL (within the limits of available software, time, and effort)
12:03:21 <elliott> Also [[User:wikipedia/The-Prophet-Wizard-of-the-Crayon-Cake]].
12:03:26 <elliott> Yes, I blocked a fucking cloak hostname, fuck you.
12:03:30 <Vorpal> fizzie, I doubt shotwell supports anything but 8 bits per channel though? Thumbnails of images with 16-bits per channel would be nice.
12:03:57 <kallisti> am I banned from internet now?
12:06:03 <fizzie> Vorpal: How should I know, I don't even have it installed? Anyway, it's got a search editor where you can list conditions; I don't know what sort of conditions it supports. And a higher-depth workflow with GEGL seems to be on the roadmap, but probably not in yet.
12:06:55 <elliott> fizzie: You mentioned something; you must be an expert in it. Vorpal rules.
12:07:02 <kallisti> elliott: I'm sorry you can't realize your true potential. :(
12:07:39 <Vorpal> elliott, no, that isn't the reason. The reason is that fizzie is an expert on everything, and if he isn't, it is able to become one within minutes. This is not a general rule. It just applies to fizzie.
12:08:22 <elliott> Counterpoint: Deewiant is an expert on everything without having to learn it.
12:08:42 <Vorpal> elliott, hm, possibly, haven't noticed that though
12:09:07 <elliott> That's because you're Swedish and stupid.
12:10:07 <kallisti> Vorpal: perl is probably your program of choice for making the thing you want happen.
12:10:49 <kallisti> it uses an advanced configuration language that allows you to configure the exact results you want it to give you.
12:10:51 <Vorpal> kallisti, well yes, but it is perl. Anyway it looks like rawtherapee supports most of the features I want. Just not all.
12:11:08 <Vorpal> plus as a bonus it provides /really/ good editing
12:11:47 <Vorpal> (well, for pre-processing)
12:12:28 <fizzie> There's a raper, and the rapee.
12:12:57 -!- kallisti has changed nick to theraper.
12:13:23 <theraper> theraper. n. someone who gives therapy
12:13:31 -!- elliott has quit (Quit: Leaving).
12:14:16 -!- theraper has changed nick to kallisti.
12:14:48 <kallisti> so now that elliott is gone I can act sensible again.
12:15:17 <kallisti> I think I'll start by going to sleep.
12:16:01 <fizzie> Time zones are a Zionist conspiracy.
12:16:22 <kallisti> yes they were perpetuated by the white man to enslave the children of the earth.
12:16:54 <kallisti> also, Muslims are taking over Europe.
12:17:41 <kallisti> I wish I could dream electricsheep
12:17:45 <fizzie> Yes, they have made the time zones so that they are in the future and can control the past.
13:06:58 -!- elliott has joined.
14:01:21 -!- Taneb has joined.
14:01:41 <Taneb> That's almost my username!
14:02:21 <Taneb> Does the halting problem apply to Push-down automata?
14:14:14 -!- Taneb has changed nick to Ngevd.
14:14:18 <Ngevd> NOW WHAT WILL YOU DO?
14:15:01 -!- Ngevd has changed nick to ettioll.
14:15:47 -!- ettioll has changed nick to Taneb.
14:15:54 -!- Frooxius has joined.
14:17:32 <elliott> Taneb: can you change your name to tanb towl
14:17:36 <fizzie> Ngetanevdb: I'm not sure what "apply" means; I mean, certainly you can wonder about whether PDAs halt, but the question is decidable, so maybe it's not so much of a problem? (Proof outline at http://www.cse.ohio-state.edu/~gurari/theory-bk/theory-bk-threese6.html -- can't find our theory course's notes.)
14:18:15 <elliott> You call something a problem if it can inspire wonder. That's how it works.
14:18:45 <Taneb> elliott, yes, but I'm not going to
14:19:16 <Taneb> fizzie, I was wondering about its decidability
14:20:05 <elliott> fizzie: can you op me please, thank you
14:21:39 <elliott> i had a dream where i smashed a glass window to get out
14:22:43 <elliott> fizzie: hi, realise you're busy, but, it's been a few seconds since my request, what is the progress report on me getting opped?
14:23:39 <fizzie> I've "tabled" it. I hear that's a kinda thing that happens. I don't know what it means.
14:24:19 <elliott> okay. can you table it into happening?
14:24:23 <Jafet> fizzie is probably trying to figure out what the right command is. I think the relevant mode is b.
14:24:40 <elliott> sorry, i think it's o instead
14:24:42 <elliott> are you using a dvorak keyboard?
14:24:46 <elliott> i think b and o might be close on that.
14:24:56 <elliott> it's not on qwerty. interesting to meet someone who uses strange keyboard layouts!
14:25:04 <elliott> we will be friends in the future. fizzie am i an op yet?
14:25:20 <Jafet> Sorry, you're right. I meant to type q.
14:25:52 <elliott> oh i think you may not be so good at typing the letters, friend!!!!
14:26:01 <elliott> but not with all those letters like that, just o.
14:26:07 <elliott> fizzie: i think: /mode +o elliott should do it. thanks!
14:27:04 <elliott> you could also /mode +o Jafet. i think he is a good person.
14:27:40 <Jafet> I'm not so sure that command is correct. I tried it and it didn't work.
14:28:29 <elliott> oh!!!!! oh no. oh no. oh no.
14:28:33 <elliott> ask fizzie for help, i think. he is a good man
14:28:38 <elliott> he will get to you as i am sure he will get to me
14:28:42 <elliott> he gets to us all, one day.
14:30:58 <Taneb> Yay my graphics card doesn't want me to play Amnesia: the Dark Descent
14:31:32 <elliott> it might be satan. ask fizzie for help.
14:57:23 -!- ais523 has joined.
15:02:47 <ais523> how's (insert project here) getting on?
15:03:17 <ais523> and have you been working on anything recently?
15:03:39 <ais523> anyway, fun fact about Linux: UNIX sockets don't respect chroots, despite looking like filenames
15:03:43 <ais523> and, hmm, last week or so?
15:03:58 <coppro> what do you mean they don't respect chroots?
15:04:34 <ais523> if you have a socket /tmp/.X11-unix/X0, then you can't have another socket called /tmp/.X11-unix/X0 open at the same time
15:04:42 <ais523> even if they happen to be different tmps due to the processes having different root directories
15:05:03 <ais523> I was trying to figure out htf a process inside weboflies could detect X outside it, apparently that was how
15:05:28 <elliott> ais523: start the X as 99?
15:05:47 <ais523> elliott: nah, better, there's a kernel flag, implemented quite recently, to separate network namespaces
15:05:51 <elliott> or rewrite filenames in your -- right
15:05:54 <ais523> in fact, I think it makes socketcall work by itself
15:06:14 <ais523> you try to make a TCP connection and get all sorts of beautiful errors because there's no networking-related initialisation at all
15:06:39 <ais523> do weboflies nc localhost 9999, and you get an error message back that it couldn't determine the IP address that localhost referred to
15:06:46 <elliott> anyway, lately-as-in-past-week-or-so: not really much :(
15:06:50 <ais523> because there isn't an /etc/hosts inside weboflies, and it has no other sort of DNS
15:07:34 <elliott> did you see the editor who archived a talk page with only one, unfinished conversation, only a few days old?
15:08:07 <ais523> + pty output: telnet: could not resolve google.com/telnet: Servname not supported for ai_socktype\x0d\x0a
15:08:17 <ais523> elliott: no, not yet, haven't looked at Esolang since I came online
15:08:22 <ais523> but that's somewhat amusing
15:08:24 <elliott> ais523: aw, I spoiled the surprise
15:08:42 <elliott> it's the Basic Input/Output Commander guy
15:08:46 <ais523> ah, you revesed the archiving
15:08:49 <fizzie> I was thinking about mentioning the rather unsightly thing about Unix sockets and the filesystem.
15:08:51 <elliott> they also edited my comment a few times before /reverting/ themselves before archiving it...
15:08:58 <fizzie> Linux has that "abstract" Unix socket namespace, too.
15:09:09 <elliott> only slightly more competent at making languages
15:09:29 <ais523> we need a Category:Almost as Shameful
15:10:10 <ais523> hmm, I think we should ban low-level stuff that works along the same lines as CPUs worked decades ago
15:10:41 <ais523> because it's perpetuating the myth that computers actually work like that, and leading to worldwide inefficiency as a result
15:10:42 <elliott> ban crap? are you nuts? nobody would visit any more
15:10:49 <ais523> oh, I meant everywhere, not Esolang
15:11:14 <ais523> this would also imply banning x86, but I don't see the problem :)
15:11:19 <elliott> ais523: I do my part by fuming inside whenever anyone hails assembly as "right on the bare metal"
15:11:31 <elliott> (when they say C is that, I just laugh instead)
15:11:37 <elliott> (when they say C++ is that, I close the tab)
15:11:49 <Taneb> elliott, Lisp is like that
15:12:24 <elliott> well, hey, the lisp machine OSes used lisp even for mega low level stuff
15:12:32 <elliott> they got as much claim as C
15:13:00 <Taneb> Wow, I didn't know that
15:13:03 <ais523> elliott: oh, did you hear about the Linode hack?
15:13:27 <ais523> do you have opinions on it?
15:13:52 <elliott> ais523: the breach itself, in terms of what was done (just stealing credentials), is uneventful; the security policies it revealed are really bad
15:14:06 <elliott> but (a) Esolang has no really confidental data and (b) I have no reason to believe anywhere else is better
15:14:14 <fizzie> ais523: I suggested bitcoin micropayments (0.50 BTC/edit) for the esowiki, in order to make it a target for such activities; currently all the cool guys just ignore it, going after bitcoins.
15:14:30 <ais523> elliott: well, they also stole thousands of bitcoins, a bunch of Slashdot people think it was an inside job
15:14:40 <elliott> ais523: the breach itself, I meant as in "the attack"
15:14:43 <ais523> fizzie: 0.50 BTC is actually quite a lot
15:15:03 <ais523> it's between $2 and $3, right?
15:15:07 <elliott> ais523: "inside job" conspiracies sound right up Slashdot's alley
15:15:18 <ais523> seems a bit much for an edit
15:15:23 <ais523> elliott: subset-of-Slashdot? indeed
15:15:36 <elliott> $3 for an edit? it'd solve the BF derivatives
15:15:53 <ais523> I wouldn't help maintain the wiki if I had to pay $3 per edit
15:16:09 <elliott> ais523: oh, admins actually /get/ 0.50 BTC per edit
15:16:52 <elliott> ais523: other edits, obviously
15:17:06 <fizzie> Then do you get locked out from editing when the wallet is empty?
15:17:14 <fizzie> If you're an admin, I mean.
15:17:18 <elliott> No, it just doesn't pay you until someone else edits.
15:17:30 <elliott> This will lead to a boom in popularity as we desperately advertise it.
15:19:32 <elliott> ais523: anyway, the real story is that people would store such large amounts of money on budget-ish VPS providers
15:20:27 <fizzie> At least it wasn't anyone's full savings (AFAIK), just their day-to-day operations money.
15:20:29 <ais523> elliott: actually, they seem to have been acting quite sensibly, all the money stolen was a small proportion of the money that the person actually owned
15:20:51 <ais523> stuff that was being actively traded rather than in storage
15:21:12 <elliott> ais523: well, I hear figures around a month's worth of money and the like
15:21:16 <ais523> which leads me to wonder, how did Bitcoin exchanges get so big that around $200000 is only a tiny proportion of the total amount of money they're handling?
15:21:25 <elliott> anyway, it was mostly organisations, not people, as I understand it
15:21:30 <elliott> the biggest mining pool, the faucet, etc.
15:21:47 <elliott> at least ISTR Silk Road is the most popular BTC thing
15:22:01 <elliott> but I may be completely wrong :)
15:22:07 <ais523> yep, wrt organisations
15:22:49 <fizzie> ISTR that Bitcoinica was the one from whom the lottest money was stolen, and their post about it mentioned that due to X and Y their "hot wallet" was larger than other operators'.
15:23:30 <elliott> The problem with Bitcoin is that it's totally the most boringest of the cryptocurrencies.
15:24:30 <fizzie> Anyway, according to bitcoincharts Mt. Gox's 30-day volume for the bitcoin/USD exchange market is something like $14 million USD, so there's certainly some money happening there.
15:25:11 <elliott> oh, I think I heard there was some money laundering going through BTC, but I may be totally wrong again
15:25:47 <Taneb> I don't really get how Bitcoin works
15:26:06 <fizzie> http://bitcoincharts.com/charts/mtgoxUSD#tgSzm1g10zm2g25 -- the stablest currency ever.
15:26:14 <Taneb> Mind you, I don't really get how most things work
15:26:23 <elliott> Taneb: It doesn't, really.
15:26:33 <elliott> fizzie: Is that dip the MtGox hack?
15:27:06 <fizzie> It does have labels on the time axis; I can't recall when it happened.
15:27:33 <Taneb> fizzie, elliott, does mcmap work with MC 1.2.x?
15:28:56 <elliott> I doubt it will in the forseeable future, though fizzie would be the one to change that prediction.
15:30:08 <ais523> elliott: hmm, I'm suddenly reminded of the way that some commonly used programming language (I can't remember whether it was Perl or Python) treats invalid characters in its input, when re-encoding, by encoding them to invalid characters in its output
15:30:13 <fizzie> I would probably make that happen if I could motivate myself to MC somewhere.
15:30:23 <ais523> e.g. converting invalid UTF-8 to invalid UTF-16 in a reversible way
15:31:25 -!- sebbu has quit (Read error: Connection reset by peer).
15:31:39 <elliott> fizzie: You realise the entire map format has changed, right?
15:31:42 <elliott> Like, the world is twice as high now.
15:33:24 <ais523> "I saved that straight from the cache and the [code] tags should stop any formatting. I'll upload it when I have a chance, but I'm pretty sure it's identical."
15:33:27 <ais523> binary files do not work like that!
15:33:52 <fizzie> That doesn't sound like a terribly major change. Anyway, no, I don't "realise" anything; haven't looked at all.
15:34:04 <ais523> anyway, really crazy idea: is it possible to get a hashlife-like algorithm to work on a random initial state by lazifying the randomness?
15:34:51 <ais523> other anyway: I'm annoyed that X seems to segfault inside weboflies but not outside (both Xorg and Xvfb, which appear to be doing the same thing when they segfault)
15:35:03 <ais523> how do you request debug symbols for a package in Ubuntu?
15:35:13 <ais523> then all I'll have to do is get core dumps working inside weboflies…
15:35:17 -!- sebbu has joined.
15:42:20 <fizzie> Having the randomness "lazy" sounds problematical in the sense that you never know what sort of things would be coming out of those regions, without evaluating them, and their effects can I guess spread at c or something.
15:53:42 <Jafet> That is really crazy because hashlife would utterly suck at it
15:53:57 <Jafet> At least until it stops being random
15:54:25 <elliott> <ais523> "I saved that straight from the cache and the [code] tags should stop any formatting. I'll upload it when I have a chance, but I'm pretty sure it's identical."
15:54:29 <elliott> ais523: we quoted that yesterday, too
15:54:35 <elliott> <fizzie> That doesn't sound like a terribly major change. Anyway, no, I don't "realise" anything; haven't looked at all.
15:54:42 <elliott> fizzie: They changed the save format.
15:54:55 <elliott> And, as I understand it, every chunk packet, completely.
15:55:12 <elliott> <ais523> how do you request debug symbols for a package in Ubuntu?
15:55:50 <ais523> ah, there's an xserver-xorg-core-dbg
15:56:23 -!- Phantom_Hoover has joined.
15:58:40 -!- MSleep has changed nick to MDude.
15:59:49 -!- Taneb has quit (Quit: Oh no!).
15:59:53 <elliott> wow, how long has it been since I *really* programmed?
16:00:59 <coppro> https://plus.google.com/109925364564856140495/posts/g4LbUEFBXTw
16:01:05 <coppro> elliott: I know the feeling
16:01:08 <coppro> <3 operating systems class
16:01:52 <fizzie> Oh, so it's 16x16x16 chunks now in a 256-high column? Well, still.
16:02:02 <elliott> People are tweeting on Google+ now?
16:05:11 <coppro> that one is a dedicated page so meh
16:11:18 * Phantom_Hoover idly ponders the presence of some brocolli on his desk.
16:12:54 <Phantom_Hoover> http://www.fileformat.info/info/unicode/char/1f360/index.htm
16:13:16 <Phantom_Hoover> Finally, a character to express roasted sweet potato without wasting all that space!
16:22:25 <ais523> I can't even figure out why weboflies would make a program segfault
16:22:48 <ais523> my guess is that it's assuming the existence of a file that doesn't exist and not checking an error condition, but it doesn't happen right after a file-related syscall
16:24:05 <fizzie> ais523: How do you deal with shared-memory blocks?
16:24:35 <ais523> mmap-based shared memory? I simply ensure that the scheduler's deterministic
16:24:42 <ais523> on the basis that processes can't access the shared memory while they aren't running
16:24:59 <ais523> this will break if someone spinlocks on shared memory, but who does that?
16:25:07 <ais523> (also, it'll break by going into an infinite loop, rather than segfaulting)
16:26:44 <ais523> which is, err, nontrivial to explain
16:26:53 <ais523> (formerly known as the Secret Project, but it's less secret than it used to be)
16:26:54 <elliott> Phantom_Hoover: you know weboflies
16:27:01 <elliott> <ais523> I can't even figure out why weboflies would make a program segfault
16:27:16 <elliott> ais523: ooh, you should post your weboflies problems on SO, the reactions would be priceless
16:27:20 <ais523> elliott: you /do/ know what happens if you put gdb and weboflies together, right?
16:27:22 <Phantom_Hoover> What was it that it did? Something to do with spoofing something, no?
16:27:25 <elliott> "Have you tried using gdb?" "Well... I can't."
16:27:41 <Phantom_Hoover> Oh, or giving you an environment to do things which are normally Bad?
16:27:53 <elliott> 5 minutes later: closed as too localised 30 seconds ago by [...]
16:28:05 <elliott> Phantom_Hoover: it's for speedruns
16:28:10 <ais523> stackoverflow is meant to be a collaborative FAQ building site, right?
16:28:15 <ais523> Phantom_Hoover: it's a program regulariser
16:28:24 <ais523> it makes Linux programs run in a repeatable way
16:28:25 <fizzie> ais523: I mean shared memory objects, you know, the sysv IPC style. Those have their own namespace (the 'keys'), and so on.
16:28:37 <ais523> I don't think I've caught anything using them yet
16:28:44 <elliott> <ais523> stackoverflow is meant to be a collaborative FAQ building site, right?
16:28:51 <ais523> but I don't see any reason why they couldn't be regularised the same as everything else?
16:29:45 <fizzie> ais523: It's just that X has that widespread MIT-SHM extension, I was wondering.
16:30:29 <fizzie> (Also a failed attach of a shm segment might easily cause things to segfault.)
16:30:34 <ais523> besides, there's only one process when it goes wrong
16:31:05 <ais523> so I don't think IPC is responsible
16:31:19 <ais523> and if weboflies doesn't have a syscall implemented, it forwards it to the actual kernel
16:32:52 <elliott> hey ais523, can you buy me more RAM?
16:33:13 <ais523> the answer is probably going to be no anyway
16:33:19 <ais523> but that doesn't stop me being curious
16:33:25 <elliott> well, you see... I want it
16:34:18 <elliott> editing a terrible article for some reason
16:34:21 <ais523> that seems statistically likely, I wouldn't expect him to have died unexpectedly
16:35:33 <elliott> always expect the unexpected
16:37:22 <fizzie> ais523: Anyway, can you get core dumps out of weboflies'd processes? It sounds not impossible for those to be gdb'able, depending on how things go.
16:37:37 <ais523> fizzie: that's what I'm wondering at the moment
16:37:56 <ais523> part of the problem is that even if it is dumping, I have no way to tell
16:38:02 <ais523> as it's being dumped on a filesystem that doesnt' exist
16:38:17 <ais523> and I can't do something like sh -c'program; ls' due to a ptrace misfeature
16:38:22 <ais523> that strace has an insane workaround for
16:38:45 <ais523> that I'll probably have to implement some day
16:39:03 <ais523> that reminds me, why /didn't/ I download strace's source, rather than relying on strace strace to determine everything about how it works?
16:40:59 <ais523> 3-clause BSD, by the look of it
16:41:11 <ais523> so I could even use code from it if necessary
16:41:21 -!- Frooxius_ has joined.
16:43:19 -!- Frooxius has quit (Ping timeout: 265 seconds).
16:43:20 -!- Frooxius_ has changed nick to Frooxius.
16:43:54 <fizzie> ais523: What was the basic weboflies mechanism, anyway? ptrace with PTRACE_SYSCALL?
16:44:05 <ais523> exactly the same as strace
16:44:39 <ais523> btw, weboflies works inside strace (but not strace -f, nor does strace work inside weboflies)
16:44:46 <ais523> I wonder if /ltrace/ works inside weboflies?
16:44:47 -!- tzxn3 has joined.
16:45:59 <ais523> nope, there's a bunch of "program uses forbidden syscall 26" followed by ltrace complaining about ENOSYS errors
16:47:58 -!- USERHOST has changed nick to WHOWAS.
16:48:15 <fizzie> In that case, I suppose you get the sigsegv in your tracer, and could just PTRACE_GETREGS the eip out of it? (And/or PTRACE_GETSIGINFO + si_addr as well.) Those numbers + /proc/maps + the debugging symbols might well give some sort of an indication to the cause of the fault, even without a proper core dump.
16:52:54 <ais523> right, I can figure out the IP pretty easily
16:53:25 <elliott> The IP address on the Web o' Flies.
16:57:28 -!- itidus21 has joined.
17:01:14 -!- Taneb has joined.
17:02:06 <itidus21> It is a day to celebrate Taneb
17:10:07 <ais523> I declare 5 March as International Taneb Day
17:10:59 -!- Ngevd has joined.
17:11:01 <itidus21> or international celebrate things based on grammatical ambiguities way day
17:11:17 -!- Taneb has quit (Read error: Connection reset by peer).
17:11:38 -!- Ngevd has changed nick to Taneb.
17:11:51 <Taneb> * Disconnected (An established connection was aborted by the software in your host machine)." mean?
17:16:35 <ais523> the connection was reset, basically
17:22:07 <Taneb> I reckon this arc in Gunnerkrigg Court is going to be weird
17:23:28 -!- augur has quit (Remote host closed the connection).
17:27:21 -!- elliott has quit (Ping timeout: 260 seconds).
17:38:00 <Taneb> Did you ever start a replacement to Rosyarrow?
17:38:45 <Phantom_Hoover> I kind of wanted elliott to, since both of us have had a shot at dictating the basic layout of a fort.
17:39:03 <Taneb> elliott's slightly asleep
17:39:04 <Phantom_Hoover> i.e. a beautifully efficient powerhouse vs. a confusing, sprawling mess.
17:39:20 <Taneb> Well, my fortress design is improving
17:50:18 -!- cswords_ has quit (Ping timeout: 260 seconds).
17:50:32 -!- cswords has joined.
17:53:00 <itidus21> computer simulation of reality is so absurd...
17:53:01 -!- Zuu has quit (Ping timeout: 260 seconds).
17:53:17 <Sgeo_> What's this about call/cc possibly being removed from Scheme?
17:53:29 -!- augur has joined.
17:53:31 <Sgeo_> Or at least Oleg wants to remove it
17:53:31 <Sgeo_> http://lists.scheme-reports.org/pipermail/scheme-reports/2012-February/001824.html
17:53:52 <itidus21> even the best simulators are all rigid optimizations of a physics problem
17:54:44 <itidus21> such simulations burning up exponentially large levels of electricity compared to the actual thing being simulated
17:56:07 <itidus21> on the other hand.. the large hadron collider i can respect...
17:57:16 <itidus21> as far as broadcasting sporting events.. if people just had local stadiums they wouldnt need hdtv
17:58:38 <Taneb> What if they were fans of American Football, but lived in Helsinki?
18:00:12 <itidus21> that would be better than being an australian in irc at 5am ranting about the limitations of technology
18:00:29 <Taneb> What part of Australia?
18:00:50 <Taneb> I've got family there
18:01:19 <itidus21> (humbly) its certainly one of the best places on earth to live
18:02:12 <Taneb> On which note, I'm thinking of learning C++
18:02:26 <ais523> Sgeo_: it's too low-level an abstraction
18:03:55 -!- Taneb has quit (Quit: dinner).
18:04:12 <Sgeo_> What is an appropriate abstraction for, say, using a callback-taking library in a linear way? Threading?
18:04:42 <ais523> probably just using functions, why would you need continuations for that?
18:05:36 <Sgeo_> Because I don't want to keep putting the rest of my code in lambdas, etc., and then how could I, say, loop over such a thing
18:05:44 <Sgeo_> Wow, I'm not making myself clear at all :/
18:06:59 <Vorpal> ais523, uh, so how will you do the stuff you could do with call/cc instead?
18:07:09 <ais523> using higher-level abstractions, obviously
18:08:24 <Vorpal> ais523, I seen backtracking implemented with call/cc in a very elegant way, not sure what other abstraction in scheme would be suitable. Sure you could add new ones but will you need multiple ones or will whatever replaces call/cc be as flexible?
18:09:00 <Vorpal> I don't think you want to add in backtracking as such into the language standard for example.
18:26:18 <fizzie> Vorpal: Non-determinism is explicitly mentioned by the document. Generators are mentioned as a reasonable abstraction; not much details, though. (Also, have seen.)
18:27:40 <Vorpal> And uh, generators wouldn't replace several of the use cases of call/cc that I can think of. Unless I missed something huge.
18:29:19 <fizzie> Well, coroutines and the nondeterminism thing.
18:30:17 <Vorpal> Using call/cc to implement backtracking is probably my favourite usage of call/cc, and I can't see how to do that with generators.
18:30:31 <Vorpal> Unless they mean something else with generators than I think they do
18:31:44 <fizzie> I am presuming the nondeterminism bit means something that does (amb 1 2 3) and backtracking would instead turn into a set of (recursive) generators that would yield the total list of valid solutions when you keep prodding at it. Or something like that, anyway.
18:32:58 <Vorpal> fizzie, well, that sounds fancy. I think I see what you are getting at though
18:34:31 <fizzie> Another Oleg document at http://okmij.org/ftp/continuations/generators.html might be relevant.
18:36:14 <fizzie> I guess it would all be obvious if one knew any Icon.
18:39:05 -!- oklopol has joined.
18:59:17 -!- oklopol has quit.
19:01:52 <Vorpal> I wonder which province the next TES game will be set in. Which ones haven't they done already (well, apart from in Arena that is)?
19:02:28 <Vorpal> I would love to see Elsweyr.
19:03:10 <Jafet> They've already done the whole continent, technically.
19:03:20 <Vorpal> Jafet, I said that above (Arena)
19:04:12 <Jafet> Or Black Marsh, but lizard people doesn't sell.
19:04:47 <Vorpal> hm so, Elsweyr, Valenwood, Black marsh and Summerset Isle haven't been done since Arena. Except possibly in spinnoff games. Most of Morrowind hasn't either but I doubt they will just make a game with everything apart from Vvardenfall (sp?)
19:05:32 <Vorpal> I have to see that Valenwood or Summerset Isle sounds most interesting to me.
19:05:58 <fizzie> You're just some sort of a specieist.
19:06:21 <fizzie> "No furries or lizards" and all that.
19:06:35 <Vorpal> Well, I would like to see some jungles, and looking at http://www.uesp.net/wiki/File:TamrielMap.jpg Valenwood looks good for that
19:07:01 -!- augur has quit (Remote host closed the connection).
19:07:14 <Vorpal> I would also like better water content, perhaps with working boats. Which would be great with Summerset Isles perhaps
19:07:23 <fizzie> How about a Tamriel/Akavir real-time strategy wargame thing?-)
19:07:36 <Vorpal> well, I'm not an RTS fan
19:07:43 <Vorpal> but sure, Akavir would be interesting
19:07:56 <Jafet> Real time twitching
19:08:07 <Vorpal> fizzie, anyway where did the nords come from in the lore, didn't they come from yet another place?
19:08:46 <Vorpal> "The last purported emigrant from Atmora recorded in history is Tiber Septim, who presumably traveled directly to Skyrim, where he spent his youth.[5]"
19:09:09 <Vorpal> so, it looks like no one knows what has been going on there for several hundred of years at least?
19:09:45 <Vorpal> but I doubt they would do another place similar to Skryrim next
19:09:56 <Vorpal> could see Atmora as an expansion pack to Skyrim though
19:09:59 <fizzie> Yokuda, where the game starts, then you drown, and that's it. Cheap to make, at least.
19:12:14 <Jafet> Oblivion is another candidate, but bethesda is never in their life going to top Planescape: Torment
19:12:35 <Vorpal> also you need to come up with a name for that game
19:12:39 <Vorpal> you can't call it Oblivion
19:13:00 <Vorpal> because they called what should have been "Cyrodiil" "Oblivion"
19:13:17 <fizzie> Summerset Isles do sound very sellable, though. You can do lots of fancy architecture, and they've got the Psijics, and so on.
19:13:21 <Jafet> You die whenever the game encounters a bug
19:13:38 <Vorpal> fizzie, aren't the Thalmor (sp?) from Summerset Isles too?
19:14:00 <Vorpal> oh btw, are there any references to the Psijics before Skyrim? I certainly can't remember hearing of them.
19:14:13 <Jafet> Galerion the Mystic
19:14:25 <Jafet> Why do I know this.
19:14:48 <Vorpal> Jafet, congrats you are a TES lore expert!
19:14:53 <Vorpal> (or you just checked uesp)
19:14:59 <fizzie> I have a feeling there were books.
19:15:04 <Jafet> Probably because I actually read every book I find
19:15:20 <fizzie> I've just read all the UESP copies. :p
19:15:27 <Vorpal> Jafet, I read like half of them
19:16:27 <Vorpal> hm, I can't think of any other video game franchise with a quite as detailed, deep and intricate lore as TES
19:16:52 <Vorpal> (FF would probably have done it by now except they aren't even remotely consistent between the games anyway.)
19:17:42 <Jafet> They very obviously make shit up as they go along, unlike some other franchises
19:18:04 <Jafet> Eg. there was never any mention of a dragon enslavement of mankind before Skyrim
19:18:13 <Jafet> And even now it is impossible to date
19:19:10 <Phantom_Hoover> <fizzie> Summerset Isles do sound very sellable, though. You can do lots of fancy architecture, and they've got the Psijics, and so on.
19:19:27 <Vorpal> Jafet, well sure, but the games seldom outright contradict each other. And when they do it is usually some small detail, like a sword ending up two-handed in one game when it was one-handed before.
19:19:42 <Jafet> Well, they don't quite have the Psijics any more
19:19:46 <Phantom_Hoover> You could do some pretty impressive stuff with the setting, too, like if it was in the middle of an Imperial counterattack against the Thalmor.
19:20:05 -!- myndzi\ has joined.
19:20:09 <Phantom_Hoover> That'd also deal with the problem of almost everyone being an Altmer and thus terrible.
19:20:24 <Vorpal> Phantom_Hoover, nice, as long as they improve the engine so it handles large scale battles
19:21:04 <Vorpal> Phantom_Hoover, or they could just do it as a set of special operation missions. Don't need a lot of people then :P
19:21:15 -!- myndzi has quit (Ping timeout: 252 seconds).
19:21:21 <Jafet> Bethesda is a huge company with loads of clout, they can do just about anything with Tamriel now and it will fly. They just need to actually put two more years of work into each game
19:21:41 <Vorpal> Jafet, adding more bugs?
19:22:15 <Vorpal> Jafet, oh yes there is one HUGE piece of retcon. Between Arena and Oblivion.
19:22:26 <Vorpal> but that is the only really huge one I can think of
19:22:48 -!- Frooxius_ has joined.
19:22:58 <Jafet> Vorpal: they have bugs because they schedule out those two years of work
19:23:03 <Vorpal> (I'm talking about the shape and size of Cyrodiil)
19:23:26 -!- Frooxius has quit (Ping timeout: 260 seconds).
19:23:29 -!- Frooxius_ has changed nick to Frooxius.
19:23:35 <Vorpal> I'm just waiting for the unofficial skyrim patch
19:24:17 <fizzie> I'm just waiting for Skyrim to become at least 50% off. :p
19:24:18 <Jafet> And that work doesn't just go to fixing bugs, it goes to actually completing all the quests
19:24:23 -!- WHOWAS has quit (Quit: bye).
19:24:35 <Jafet> So many storylines and plot arcs are left incomplete in the games
19:24:40 <Vorpal> Phantom_Hoover, uh? You know that Imperial Province? Cyrodiil...
19:24:51 <Vorpal> Phantom_Hoover, that grew a LOT in Oblivion
19:25:07 <Phantom_Hoover> Sure, if you take it to be *only* the area around the Imperial City.
19:25:11 <Jafet> Maybe people become smaller when they enter Cyrodiil
19:25:14 <Vorpal> Phantom_Hoover, on the map it is
19:25:49 <Vorpal> Phantom_Hoover, also all the towns (apart from Imperial city), like Bravil, Anvil and so on are missing
19:26:37 <Phantom_Hoover> <Jafet> And that work doesn't just go to fixing bugs, it goes to actually completing all the quests
19:26:48 <Phantom_Hoover> So you mean the factions would have actually *mattered*?
19:26:56 <Jafet> If they added things to Cyrodiil, it's not so much a retcon as a forecon
19:27:26 <Vorpal> <Jafet> So many storylines and plot arcs are left incomplete in the games <-- I feel like the Winterhold College questline is way too short compared to the Mage guild one in Oblivion
19:28:06 <Jafet> If the factions mattered at all, a player couldn't rise to the tops of all of them with the same character
19:28:09 <Vorpal> I mean, you should have to /work/ to become Argemage
19:28:21 <Jafet> Sadly, this design has become entrenched since Oblivion
19:28:32 <Vorpal> I mean, they shouldn't just let any new random person become it
19:29:08 <Phantom_Hoover> Vorpal, that the end of the College questline is particularly dumb is not a point of contention.
19:29:23 <Vorpal> Phantom_Hoover, indeed, I'm just venting some steam here.
19:29:54 <Jafet> Skyrim is generally bad because of the release date
19:29:58 <Vorpal> the Theives Guild quests are better. Far more work involved. Quite repetitive though, which is not a good thing.
19:30:13 <Vorpal> Jafet, it was released too early yeah
19:30:54 <Vorpal> well, they won't be able to nail the next game at 12/12/12, and after that, it is going to take ages before the next such chance
19:30:56 <Jafet> Well, that's what happens when the executives make the games.
19:31:13 <Vorpal> at worst they can do an expansion pack at 12/12/12
19:31:24 <Jafet> The most colourful game was Morrowind
19:31:25 <Vorpal> or a fallout game (meh)
19:31:46 <Vorpal> Jafet, didn't ever finish Morrowind, it is very crashy on my computer for some reason
19:31:59 <Vorpal> nah, would be Fallout 4
19:32:39 <Phantom_Hoover> http://spikedmath.com/480.html Intelligent commentary from the mathematics community.
19:32:50 <Vorpal> Jafet, I really liked Shivering Isles
19:34:08 <Phantom_Hoover> It'd be cool but no fun if they brought Solsteim back again.
19:35:01 <Vorpal> Phantom_Hoover, uh, you mean Solstheim?
19:35:20 <Phantom_Hoover> I got tired of checking the UESP before saying anything. So sue me.
19:35:40 <Vorpal> Phantom_Hoover, I was trying to search for it, as I never got that far into Morrowind :P
19:37:22 <Phantom_Hoover> I pirated it only to discover the copy I have's in Russian.
19:39:01 <Vorpal> Also I just died a bit inside. I found out they are adding a pre-rendered cutscene to the next update of Witcher 2 (due out on the 17th April).
19:39:31 <Vorpal> There is absolutely no need to, that game engine is amazing... And upscaling will of course be a problem
19:39:43 <Vorpal> http://www.gog.com/en/news/the_witcher_2_assassins_of_kings_enhanced_edition_announced
19:40:05 <Vorpal> 3.5 minutes of it even
19:41:22 <Vorpal> Phantom_Hoover, have you played that game btw?
19:41:32 <fizzie> Even a microsecond would be a blasphemy.
19:41:48 <Phantom_Hoover> fizzie, let's just hope nobody tells Vorpal about pictures.
19:41:49 -!- zzo38 has joined.
19:42:06 <Phantom_Hoover> Vorpal, no, on account of having heard it's kind of crap.
19:42:45 <Vorpal> Phantom_Hoover, who said that? And what was the reason for it?
19:43:29 <Phantom_Hoover> Overcomplicated interface, repetitive boring gameplay, bad plot...
19:44:11 <Vorpal> The combat is quite hard even on normal difficulty (some people like that, I like it). There is some LOD popin sometimes though. That is about the only bad things I can think of.
19:44:43 <Vorpal> Phantom_Hoover, and no the interface isn't overcomplicated. It works unlike that of Skyrim (which has a bloody awful interface on PC)
19:45:18 <Vorpal> it is the normal, inventory/map/level-up/journal-and-quests split for the menus, they are quite straight forward too.
19:46:35 <Vorpal> okay so there are a few minor menus, like the alchemy/crafting one. But that one is simple: Select recipe, click craft, select how many from the dialog that pops up.
19:46:56 * Phantom_Hoover wonders how the hell the term 'RPG' can cover both Skyrim and The Witcher 2.
19:47:11 <Vorpal> Phantom_Hoover, anyway, no the GUI isn't complicated. As for gameplay being boring. I would argue that Skyrim has fairly boring gameplay.
19:47:18 <Vorpal> compared to that of witcher 2
19:47:33 <Vorpal> the combat in Witcher 2 is really good. It could be boring if you play on easy.
19:47:39 <Phantom_Hoover> I mean, does role-playing mean following a set character on a largely predetermined path, or projecting yourself into the gameworld?
19:48:01 <Vorpal> and unlike Skyrim, the combat animations in witcher 2 are in the right place dammit
19:48:02 <Phantom_Hoover> They're both exact opposites and yet they're both treated as the same genre.
19:48:37 <Vorpal> Phantom_Hoover, well, in witcher 2 it follows a set character, but you can make major choices affecting the rest of the game in several places.
19:48:47 <Vorpal> There are two completely different chapter 2 for example
19:48:49 <Jafet> Not to be confused with roll playing
19:49:33 <Phantom_Hoover> Picking between 2 plots is hardly creating your own story.
19:49:56 <Vorpal> Phantom_Hoover, true, but how much story do you really create in Skyrim.
19:50:34 <Jafet> There isn't much story, just running from one NPC to the next
19:51:12 <Vorpal> Phantom_Hoover, You don't really change the game world that much. NPCs aren't going to treat you differently for beating the main quest apart from a few greetings here and there (unless they are related to the main quest, like the Blades or such)
19:51:26 <Phantom_Hoover> From a purely in-game perspective the Dragonborn is just some weirdo running around doing odd jobs.
19:51:35 <Vorpal> Phantom_Hoover, exactly
19:51:41 <Phantom_Hoover> It's up to you to tie that together into a coherent system of motivations and aims.
19:52:07 <Jafet> Not really. The game developers just weren't doing their job.
19:52:23 <Jafet> In other games, I don't need self-conscious effort to role-play.
19:52:32 <Vorpal> None of the NPCs really care about what you have done. You might be the hero of the realm and they will still great you like "hey, jerk who isn't a nord"
19:53:16 <Jafet> More to the point, the developers take great pain to have quests affect as little of the game world as possible
19:53:34 <Jafet> Because they don't want the casual teenage idiot to whine
19:53:48 <Vorpal> Phantom_Hoover, and I would say the plot is fairly good in Witcher 2. It is a plot of political intrigues really.
19:53:55 <Vorpal> and really good as such
19:54:18 <Vorpal> Phantom_Hoover, unless they are playing COD I'm not going to complain :P
19:54:38 <Vorpal> in fact I'm not going to complain if they play COD either, because that means they are not playing a game I'm playing :P
19:54:39 <Jafet> I doubt the casual teenage idiot would whine anyway, it's just executive rationality
19:55:24 <Vorpal> Jafet, I think the real reasons quests don't affect the game world much except for the NPCs tied directly to the quest (and sometimes barely even then) is to avoid bugs with quests interacting.
19:55:35 <Vorpal> Such bugs still happen yes. But imagine how much worse it could be.
19:55:59 <Phantom_Hoover> <Jafet> More to the point, the developers take great pain to have quests affect as little of the game world as possible
19:55:59 <Phantom_Hoover> <Jafet> Because they don't want the casual teenage idiot to whine
19:56:08 <Vorpal> Phantom_Hoover, it doesn't make sense
19:56:25 <Phantom_Hoover> "OH MY GOD THAT THING CHANGED SLIGHTLY I HATE THIS GAME!!111!!111!!1"?
19:56:32 <Vorpal> Phantom_Hoover, the real reason is most probably to avoid quest interaction bugs.
19:56:41 <Jafet> Yes, they actually worry that people complain about that.
19:57:22 <Jafet> Vorpal already gave the reason
19:57:38 <Jafet> OMGZ I CAN'T FINISH THIS QUEST BECAUSE I CHOSE TO RAZE ITS CITY
19:57:50 <Vorpal> Phantom_Hoover, also it would involve a lot of work to handle all possible questline interactions. Like arch mage and leader of companions and just joined thieves guild
19:58:22 <Phantom_Hoover> Like when I couldn't progress in the main quest because Winterhold was blocking it.
19:58:36 <Vorpal> haven't heard of that specific one, but yeah exactly
19:58:44 <Vorpal> from what I heard the civil war quest line manages break loads of other quests.
19:58:46 <Phantom_Hoover> It didn't make in-game sense; the dialogue trees just hadn't been built to cover that particular case.
19:59:29 <Jafet> How much work would it take to get all the questlines to interact properly?
19:59:45 <Vorpal> Phantom_Hoover, I'm unable to start becoming a thane of Falkreath (sp?) in my current game. And I have no idea why
19:59:51 <Phantom_Hoover> Jafet, I don't know, but it's a good bet that it'd be harder than you think.
19:59:55 <Vorpal> the relevant options just never show up when UESP say they should
20:00:02 <Vorpal> no known bug that applies listed
20:00:48 <Jafet> It's only hard in that it takes actual developer time
20:01:11 <Vorpal> Jafet, and expensive to hire voice actors for additional lines.
20:01:11 -!- MoALTz_ has joined.
20:01:30 <Vorpal> Phantom_Hoover, speaking of which, while the voice acting in skyrim isn't as bad as in oblivion, it is still pretty terrible
20:01:43 <Vorpal> apart from for a few select NPCs.
20:02:03 <Jafet> It's not so much terrible in itself as it is used badly
20:02:25 <Jafet> What with the continuing mudcrabitis
20:02:41 <Vorpal> Phantom_Hoover, a lot of the non-star-actor voicing lacks emotion for example. And several voices are reused too much. I'm pretty sure Clavicus (sp?) Vile was Mercer Frey for example.
20:03:06 -!- MoALTz has quit (Ping timeout: 260 seconds).
20:03:18 <Phantom_Hoover> Mercer's the same guy as Belethor, Enzir and many others, IIRC.
20:03:32 <Vorpal> Phantom_Hoover, and he has a very distinctive accent.
20:04:18 <Vorpal> I'm not saying voice actors shouldn't be reused. They must be for budget reasons, but at least make the separate instances sound different enough. Some voice actors can pull that off. Some just sound too unique.
20:05:22 <Vorpal> Phantom_Hoover, also I wonder how many people voice acted the guards. I'm guessing like two persons of each gender or some such.
20:05:31 <ais523> btw, the GitHub hack is hilarious
20:05:43 <Vorpal> sure I realise you can't make every guard sound unique, but a bit more variation would have been nice.
20:05:53 <ais523> it seems that Rails has a feature that's suspiciously like register_globals
20:06:08 <ais523> and GitHub were using it without proper security checks
20:06:13 <Vorpal> and no I'm not talking about arrow in knee (though actually it is far less annoying than the "mudcrabs" discussions of oblivion)
20:06:34 <fizzie> Speech synthesis + randomized voice parameters == unique voices to all.
20:06:44 <Vorpal> fizzie, will they should as good though?
20:06:56 <fizzie> No, but that's the price you pay.
20:07:25 <Vorpal> btw, there witcher 2 stands out. Really awesome voice acting.
20:07:39 <fizzie> Set of recorded voices + slight tweaking can move the "amount of variation" / "natural-sounding speech" boundary around.
20:07:55 <Vorpal> fizzie, so sampling basically?
20:08:20 <Vorpal> hm couldn't that work, sampling on the word or phrase level?
20:08:51 <fizzie> Not unless you want people to sound quite multiple-personality-disordery.
20:08:52 <Jafet> It works when you produce a synthesizer that works
20:09:04 <fizzie> Of course it's nontrivial to do the right sort of tweaking without having a human present.
20:09:46 <fizzie> GlaDOS' voice is a good example of "start with a human, tweak for style" voice, even if the goal there is not to make another (but different) human voice.
20:10:16 <Vorpal> anyway I would strongly recommend Witcher 2 as one of the best modern RPGs out there, if you are grown up. I wouldn't recommend it to anyone under age. Because it is rated Mature for a reason.
20:10:41 <Jafet> Any voice, facial, hair or cloth synthesis in a video game is going to suck for the next few decades
20:11:01 -!- invex has joined.
20:11:03 <Jafet> It's cool to push the boundaries, but that's the truth
20:11:07 <Vorpal> Jafet, not sure about cloth actually. Hair yes.
20:11:28 <fizzie> Anyway, you can do transformations to voices. If they're slight enough, they might not sound too bad. Of course it's then arguable whether it adds enough variation to be worth it, since you might still be able to recognize that there's just very few actual speakers.
20:11:37 <Vorpal> that is maybe the weakest point of Witcher 2, the hair, but even then the hair in that game looks better than in most other games.
20:12:34 <Jafet> The real test is when you compare a game to the games five years *after* it
20:12:44 <Jafet> If nothing's changed, then your synthesis is real
20:13:09 <Jafet> Eg. vehicles now enjoy this status
20:13:53 <Vorpal> Jafet, the "game" mechanics of flight sims are also good enough to not change much any more.
20:14:51 <Vorpal> Jafet, as for face, I guess LA Noir doesn't count, since it isn't actually synthesis?
20:15:00 <Jafet> You'll see in five years.
20:15:08 <Vorpal> Jafet, it uses face scanning
20:15:23 <Vorpal> haven't played it myself
20:15:37 <Jafet> I'm still sure it's going to look bad in five years.
20:15:40 <Vorpal> Anyway, clothing in Witcher 2 looks awesome. Mostly it is armour though, but quite a lot of leather or other light armour, and dresses and such does seem to behave naturally.
20:15:50 <Vorpal> if that is what you meant with cloth synthesis
20:17:16 <Jafet> It was shorter than typing "voice synthesis and cloth animation"
20:17:33 <Vorpal> Jafet, cloth animation would be involved in dresses though
20:18:59 <Jafet> Cloth animation is a classic hard problem. Most games simplify it by having really coarse physics modelling or connecting canned animations
20:19:25 <Jafet> Sometimes it looks smooth enough that you don't notice it's fake, and I think that's the best you can achieve these days
20:19:41 -!- boily has joined.
20:22:41 -!- augur has joined.
20:23:50 -!- Taneb has joined.
20:23:59 <Taneb> Is ℒ meta-turing complete?
20:25:57 -!- Jafet has quit (Quit: Leaving.).
20:39:01 <fizzie> Was this "ntnu.no" oerjan's sorta-place, or how was it?
20:41:17 <fizzie> Lastlog suggests it was.
20:41:46 <olsner> yes, some kind of norwegian university-ish thing, presumably
20:45:21 <Friendship> Taneb: Whether it is TC or not is a philosophical question, but adding "meta" doesn't make it suddenly make sense.
20:45:33 <Taneb> http://esolangs.org/wiki/Meta_Turing-complete
20:46:43 <fizzie> Anyway, random googling hit a ntnu.no url of a (very short and detail-less) paper on "Voice transformation and speech synthesis for video games" from GDC 2008. It speaks of the sort of thing where you have a single recorded voice and it is transformed to generate multiple people.
20:49:10 <Taneb> He's probably asleep
20:49:38 <Friendship> Taneb: So yeah, those two pages seem to be describing the same phenomenon.
20:50:03 <Taneb> Not quite, but Meta Turing-Complete is a superset of Fancy L?
20:50:20 <olsner> meta turing complete looks like nonsense to me
20:50:34 <Taneb> Consider that ALPACA can describe many CA's, but a theoretical Fancy L could only describe one
20:51:08 <Friendship> Taneb: That's just what ℒ means, ℒ-hard seems to be equivalent to metaTC.
20:54:40 -!- monqy has joined.
21:02:13 -!- MoALTz_ has quit (Ping timeout: 260 seconds).
21:13:22 <Phantom_Hoover> <Friendship> Taneb: Whether it is TC or not is a philosophical question, but adding "meta" doesn't make it suddenly make sense.
21:14:27 <Taneb> We need to ASK THE GHOST OF ALAN TURING
21:14:43 <Phantom_Hoover> <Vorpal> anyway I would strongly recommend Witcher 2 as one of the best modern RPGs out there, if you are grown up. I wouldn't recommend it to anyone under age. Because it is rated Mature for a reason.
21:14:52 <olsner> Taneb: you mean the META-GHOST of ALAN META-TURING?
21:15:09 <Taneb> I never meta ghost
21:15:10 <Phantom_Hoover> You mean the really immature sex stuff made by and for 13-year-olds?
21:17:19 <Phantom_Hoover> Huh, Skyrim /has/ a disposition system, it just never /uses/ it.
21:19:07 -!- invex has changed nick to PiRSquared.
21:19:21 <Vorpal> <Phantom_Hoover> Huh, Skyrim /has/ a disposition system, it just never /uses/ it. <-- wow
21:19:44 <Taneb> Come one, Vorpal. He just said that.
21:19:45 <Vorpal> <Phantom_Hoover> You mean the really immature sex stuff made by and for 13-year-olds? <-- hm, yes there is a bit of that. But there are actually some real mature stuff as well.
21:20:21 <Phantom_Hoover> The disposition system for marriage is such that there's a woman in Whiterun who'll marry you immediately after you beat her in a fistfight.
21:28:04 <Phantom_Hoover> (Also I find the entire marriage system in Skyrim to be the funniest thing in the game; it's so ridiculous.)
21:30:22 -!- MoALTz has joined.
21:36:46 -!- boily has quit (Quit: WeeChat 0.3.7).
21:37:12 <fizzie> He looks like an axe murderer to me.
21:38:03 <fizzie> Phantom_Hoover: I thought it may, but wasn't sure enough to s///.
21:38:51 <fizzie> His hand appears to have slightly melted into the clothing, too.
21:39:11 <Phantom_Hoover> The dialogue with him is along the lines of "So, you want to get married? Well, everyone in Skyrim is too busy getting eaten by the fauna to talk to their prospective spouse, so you'll need a Marriage Thing. By the way, I happen to have a very nice Marriage Thing with me!"
21:43:55 * Phantom_Hoover reads [[Caster Semenya]], notes the gem "The federation also explained that the motivation for the test was not suspected cheating but a desire to determine whether she had a "rare medical condition" giving her an unfair competitive advantage."
21:44:10 <Phantom_Hoover> WELL WE CAN'T HAVE ANYONE'S INHERENT ABILITY GIVING THEM AN UNFAIR ADVANTAGE NOW
21:45:59 <zzo38> Do you have a .plan file on any accounts anywhere?
21:51:11 <Vorpal> <Phantom_Hoover> He then charges you 200 moneys for a Marriage Thing. <-- septims
21:52:06 <ais523> Vorpal: you're missing the point of the placeholder
21:52:21 <ais523> it's not necessarily "I don't know what it's called", but "other people won't know what it's called"
21:52:26 <Vorpal> ais523, uh, I was correcting "moneys"
21:52:33 <ais523> Vorpal: it's a placholder
21:52:39 <ais523> to make the sentence more legible
21:52:47 <Vorpal> ais523, so "currency units (septims)"
21:53:20 <Vorpal> and "Marriage Thing" is "Amulet of Mara" I /think/
21:54:16 <fizzie> I had a .plan file in some earlier home Linux box.
21:54:47 <fizzie> And maybe I have a .plan at the university's general-purpose shell home directory? Though I have a feeling they wiped those out at some point.
21:55:19 <fizzie> Oh, I still have a .plan file there.
21:55:22 <fizzie> vipunen ~ 53 % cat .plan
21:55:22 <fizzie> >#v1&#:<-1<-1\0\_.@ use email
21:55:22 <fizzie> >2-!#v_:2\^fibre^-< to connect.
21:55:32 <fizzie> It's even channel-appropriate.
21:55:47 <fizzie> Sadly, I don't think anyone has ever asked me about the meaning of the .plan.
21:56:26 <zzo38> I have an account on the Free Geek computers, so I have a .plan file there. It contains information about which days I am in.
21:56:52 <fizzie> The finger server even returns the plan, I was wondering if they had given up on that.
21:57:03 <fizzie> I don't think we have a publicly approachable finger server, though.
21:57:46 <fizzie> It's a file in your home directory that a finger server includes in the finger output.
21:57:52 <zzo38> Someone has said that "Which constellation on the ecliptic is an astrological sign not named after?" should be a question in Trivial Pursuit, after we have discussed a few things. Do you think so?
21:57:58 <fizzie> Typically a multi-line one; while a .project is a single-line variant.
21:58:12 <fizzie> You're supposed to set them so that people fingering you can know roughly what you're working on.
21:58:24 -!- Taneb has quit (Read error: Connection reset by peer).
21:58:30 <zzo38> The Free Geek Vancouver also has no public accessible anything in the user accounts, as far as I know.
21:59:26 <fizzie> <mooz> slaves to armok programmer's .plan is ~1000 lines long
22:00:22 <ais523> zzo38: /all/ the constellations in the ecliptic, there's an astrological sign not named after that constellation
22:00:26 <zzo38> fizzie: I think that is far longer than a .plan file ought to be; it should never be more than twenty lines in my opinion (although probably less)
22:00:27 <ais523> at least 11 of them, in fact
22:00:53 <fizzie> Also "what is a .plan file" was apparently taught at our "computer as a tool" (the mandatory introduction-to-the-university's-system) course.
22:00:54 <zzo38> ais523: Maybe I should rephrase the question, it isn't very good
22:01:05 <fizzie> (I'm grepping old logs to find out if anyone has asked me about the .plan above.
22:01:40 <fizzie> Apparently it was even one exercise on that course, to make a .plan file for yourself.
22:02:06 <zzo38> ais523: Do you know a better way to rephrase the question?
22:02:40 <fizzie> I see someone else's plan has been "Doing my best to become the fastest wank^H^H^H^Htyper."
22:02:46 <ais523> "which constellation on the ecliptic does not have an astrological sign named after it?"
22:03:22 <zzo38> ais523: Yes, that is better.
22:03:31 <fizzie> (xlock also (at least with some versions/settings/etc.) shows the locked user's .plan under the user name.)
22:03:39 <Phantom_Hoover> <Vorpal> and "Marriage Thing" is "Amulet of Mara" I /think/
22:03:46 <zzo38> [1] Do you think it would be a good question for Trivial Pursuit? [2] Do you know the answer?
22:04:21 <ais523> [1] probably not, there are probably minor constellations on the ecliptic that don't have names; [2] no
22:05:12 <zzo38> ais523: I think all constellations have names because the IAU defined all areas of the celestial sphere as belonging to one of 88 Latin constellations.
22:05:33 <zzo38> (They are useful for naming stars and for identifying stars)
22:06:21 <zzo38> See if anyone else in here knows the answer
22:10:23 <zzo38> There are other people who have accounts at Free Geek Vancouver, but as far as I know none of them have a .plan file.
22:11:15 -!- tzxn3 has quit (Quit: Leaving).
22:12:43 <fizzie> I did not know the answer, but now I do; still, it's slightly obscure for Trivial Pursuit, maybe? Then again, I'm no trivia game designer, and certainly it's got questions of various difficulties.
22:13:38 <zzo38> I want a more obscure trivia game anyways
22:15:32 <zzo38> Another thing: Two astrological signs are named after constellations which have slightly different names than the astrological signs named after them (I don't know if it was always the case). Do you know which two? (Hint: One of them is an astrological sign which one of the tropics is named after and corresponds to.)
22:15:48 <Vorpal> zzo38, wouldn't sell to the "masses"
22:18:02 <zzo38> Vorpal: I suppose you are correct; but you could still make minor selling things, there are such service to sell such games with actual cards. You could also make them as data files in computer; actually I think they should include both so that you can write computer program with it and print out the cards yourself too
22:18:42 <Vorpal> oh you mean a physical game
22:18:53 <Vorpal> I was thinking game show
22:19:05 <Vorpal> still I don't think it is mainstream enough
22:19:10 <zzo38> Actually, I found something, you can sell with cards, dice, custom board, whatever. For computer file you can just type into the computer. For television, of course you need far more
22:20:56 <fizzie> I was going to say "Trivial Pursuit, which was mentioned, isn't a game show", but of course it has game show adaptations. Anyway, it's got all these different nonsense editions, though admittedly a bit more "mainstream" than astronomy/astrology. (Like, uh, the Star Wars Episode I edition.)
22:20:59 <Vorpal> zzo38, but sure, include questions like "what does the following 4x4 transformation matrix do"
22:21:07 <Vorpal> (okay that is even more niche)
22:21:30 <zzo38> Vorpal: Yes you can include stuff like that too if you want to
22:22:59 <Vorpal> is there a collective name for the vectors (1 0 0 1) (0 1 0 1) (0 0 1 1)? That is the unit vectors of 3D with a homogeneous coordinate added?
22:24:43 <fizzie> Perhaps they're called "the unit vectors of 3D with a homogenous coordinate added".
22:25:00 -!- PiRSquared has quit (Quit: ChatZilla 0.9.88 [Firefox 10.0/20111221135037]).
22:25:56 <Vorpal> fizzie, that is rather long and unwieldy though
22:28:26 <ais523> they're just x̂, ŷ, ẑ in an xyzh coordinate system, aren't they?
22:28:38 <ais523> also, my IRC client renders those combined characters pretty badly
22:28:47 <Vorpal> ais523, uh is that how you write unit vectors? x with a ^ on top?
22:28:58 <Vorpal> I never seen that way to write unit vectors before
22:29:05 <ais523> it's what's taught in school in the UK
22:29:19 <ion> vorpal: How do you write unit vectors?
22:29:21 <ais523> ^ on top = unit vector, regardless of whether it's in an axis direction or not
22:29:58 <Vorpal> ion, e_1, e_2, ...., e_n (where _ is as in TeX) for (1 0 ...) (0 1 0 ...) ...
22:30:11 <fizzie> \hat{x} is I think vaguely common for the normalized x.
22:30:16 <Vorpal> that is what I have been taught
22:30:38 <fizzie> And I've seen the x-, y- and z-aligned unit vectors called \hat{i}, \hat{j} and \hat{k}.
22:30:40 <Vorpal> ion, had a German teacher though, so who knows (I live in Sweden)
22:31:10 <ais523> fizzie: oh, you're right
22:31:18 <Vorpal> Phantom_Hoover, no, the math teacher was from Germany
22:31:23 <Vorpal> he didn't teach German
22:31:30 <fizzie> "The notations , , , or , with or without hat/circumflex, are also used, particularly in contexts where i, j, k might lead to confusion with another quantity."
22:31:49 <Vorpal> Phantom_Hoover, he spoke pretty good Swedish though
22:32:03 <fizzie> Uh... that's hatted (x, y, z), (x_1, x_2, x_3), (e_x, e_y, e_z) and (e_1, e_2, e_3).
22:32:04 <Vorpal> I currently have a math teacher at university who is from China. And that guy doesn't speak very good Swedish.
22:32:09 <fizzie> From http://en.wikipedia.org/wiki/Unit_vector
22:32:15 <fizzie> So that's a total of five notations for them.
22:32:48 <fizzie> And then the versions without hats for those where it makes sense.
22:33:23 <Vorpal> fizzie, yeah I never seen the hat one
22:33:30 <Phantom_Hoover> In these parts we write vectors with squiggly lines underneath.
22:33:58 <Vorpal> Phantom_Hoover, I seen vectors done as straight underline, bold font and line just above the letter
22:34:02 <fizzie> The hat is a common notation for normalization, though; it's just that when speaking of the unit vectors, it's a bit superfluous.
22:34:12 <fizzie> Sometimes there's a small arrow above.
22:34:26 <Vorpal> fizzie, seen arrow below too
22:36:06 <Vorpal> Why are there so many ways to write the same thing in math?
22:36:13 <Vorpal> It just hinders communication
22:36:26 <Vorpal> and then there is the case of the same symbol being used for different things
22:36:40 <Vorpal> which at least make slightly more sense
22:36:40 <ais523> Vorpal: the two are probably connected, if you think about it
22:37:10 <Vorpal> ais523, hm? the two questions I raised? Or the two usages of the same symbol?
22:37:13 <Phantom_Hoover> Vorpal, surely it's obvious even to you why there are so many notations for vectors.
22:37:24 <zzo38> I think the idea of "homogeneous coordinates" is also a similar idea occurs in my system of "matrix accounting"?
22:37:41 <Vorpal> Phantom_Hoover, presumably different people preferred different notations and thus we got the confusing mess we are in now
22:37:43 <ais523> Vorpal: writing the same thing different ways, and same symbol used for different things
22:37:44 <Vorpal> which is not a very good reason
22:37:52 <ais523> you have to invent new notations on the fly to avoid clashes
22:38:41 <zzo38> But, I suppose, many things in mathematics can be found to have multiple applications possible
22:38:56 <Vorpal> ais523, so you are saying 5 different persons needed to come up with a notation for vectors? That is a lot of independent research around the same time finding the same concept then
22:39:45 <ais523> it's bad enough, in my field, that for the same notion, some people are using ⩽ for the same thing that other people are using ⩾ for
22:40:05 <ais523> Vorpal: and I'm saying that notation that avoids clashes for one person, will have clashes for someone else
22:40:14 <ais523> people who use vectors almost exclusively aren't going to want to write in arrows all the time
22:40:29 <ais523> but people who use both vectors and other things may need them to avoid clashes
22:40:52 <Vorpal> <ais523> it's bad enough, in my field, that for the same notion, some people are using ⩽ for the same thing that other people are using ⩾ for <-- wow
22:41:04 <Vorpal> <ais523> Vorpal: and I'm saying that notation that avoids clashes for one person, will have clashes for someone else <-- ah, right
22:41:15 <Phantom_Hoover> Vorpal, specifically, vectors are a very widespread and very old (by mathematical notational standards) concept.
22:41:28 <Phantom_Hoover> They have been independently invented and studied countless times.
22:41:57 <Phantom_Hoover> Each of those times resulted in a new notation which persisted in the work which inherited most directly from it, and over time they intermingled.
22:41:59 <Vorpal> Phantom_Hoover, but over time you would think some ways of writing them would die out and you would in the end be left with a single one
22:42:28 <ais523> actually, might have been ≤ and ≥, I can't remember
22:42:32 <Phantom_Hoover> No, because for all that intermingling the people using one notation don't interact that much with those using the other, and most are content to make do.
22:43:08 <Vorpal> ais523, isn't ≤ just a different way to write ⩽? Different stylizing I mean
22:43:15 <ais523> not always, but yes sometimes
22:43:29 <ais523> why do you think they have different Unicode codepoints?
22:43:37 <Vorpal> Phantom_Hoover, fair enough
22:44:18 <fizzie> What's the fourth one for (i, j, k)? Is it just l? Or do people just go with (e_1, e_2, e_3, e_4) when they need four? (For (x, y, z) I'd assume w.)
22:45:01 <ais523> it's got to be either l or h, I think
22:45:55 <Vorpal> ais523, btw you said xyzh for homogenus coordinate above, it is just that I seem to remember it being xyzw in some contexts (such as opengl)?
22:46:21 <ais523> opengl has sets of four arbitrary letters that it uses for coordinates without knowing what they mean
22:46:40 <ais523> actually, probably it's /actually/ GLSL
22:46:45 <ais523> I know too many GPU-related frameworks
22:46:53 <ais523> so you can address a vector as xyzw or as rgba, for instance
22:47:11 <ais523> there are a few other schemes, they had to use weird letters sometimes to avoid clashes
22:47:25 <Vorpal> ais523, um, opengl accpets xyzw, rgba, 0123, <something I forgot related to texture coordinates>A
22:47:59 <Vorpal> at least those are the four variants I know of
22:48:13 <Vorpal> and presumably the texcoord one is uv??
22:48:29 <fizzie> OpenGL texture coordinate function parameters are s, t, r, q.
22:48:44 <fizzie> Though normally people don't go beyond s, t.
22:48:48 <Vorpal> so why is it called uv-mapping?
22:49:01 -!- Phantom_Hoover has quit (Remote host closed the connection).
22:49:02 <fizzie> Because the coordinates are quite often also (u, v).
22:49:29 <fizzie> I don't know how they'd continue from (u, v) onwards though.
22:49:37 <Vorpal> sadly mixing them doesn't work. You can't do foo.xyb
22:50:00 <fizzie> For some reason I would have guessed (u, v, s, t), but I'm not sure if that happens anywhere.
22:50:11 <Vorpal> fizzie, well, why should they be in order, strq isn't
22:50:19 <Vorpal> rgba isn't either of course
22:50:32 <Vorpal> so yeah uvst would work
22:51:04 <fizzie> 'strq' isn't even in the '3412' order.
22:51:24 <fizzie> Maybe someone added 'r' first, and then had to add 'q' later.
22:51:46 <Vorpal> fizzie, they could go stuv
22:52:34 <Vorpal> fizzie, then you could claim "uv-mapping" originated in someone reading it in middle endian
22:52:50 <Vorpal> which would be an awesome way to confuse new students
22:53:21 <fizzie> I see "uvw mapping" is also a term that is used, in 3D modeling contexts.
22:54:54 <Vorpal> hm, minecraft with 3D textures and a way to cut blocks in any way you like, that sounds awesome
22:55:27 <Vorpal> not sure how you would handle grass though
22:56:28 <fizzie> It seems to be about just plain regular textures; not sure how it differs from UV mapping.
22:56:54 <fizzie> A Blender tutorial about it seems to be using UVW and UV quite interchangeably.
22:58:09 -!- kallisti has quit (Read error: Connection reset by peer).
22:58:30 <fizzie> (A "Blender tutorial" above meant "someone's unofficial tutorial that uses Blender", not "a tutorial that's part of the official Blender documentation".)
22:58:37 -!- kallisti has joined.
22:58:37 -!- kallisti has quit (Changing host).
22:58:37 -!- kallisti has joined.
22:59:17 <Vorpal> fizzie, hm 3D textures mostly seem to be used for clouds and similar
23:00:01 <kallisti> I'm trying to get my second monitor working.
23:00:17 <kallisti> at first it was freezing up my computer with kernel messges about an invalid EDID checksum or something
23:00:24 <kallisti> I switched to usermode mode setting
23:00:30 <kallisti> which allows me to connect the monitor
23:00:37 <kallisti> now the display resolution on my normal monitor is shit.
23:00:42 <Vorpal> kallisti, nvidia, amd or intel?
23:01:02 <Vorpal> kallisti, always worked for me in the gnome 2 monitor control panel
23:01:09 <Vorpal> never used anything else for it
23:01:38 <Vorpal> kallisti, pretty sure there is a tool in xfce too
23:01:41 <kallisti> lol noob. real linux users have horrible problems for very basic things that they have to spend hours fixing by hand.
23:01:44 <fizzie> xrandr's command-line tool is reasonably reasonable, and should be able to do everything.
23:02:07 <fizzie> And there's... I forget the name. It was gxrandr or some-such.
23:02:43 <Vorpal> å<kallisti> lol noob. real linux users have horrible problems for very basic things that they have to spend hours fixing by hand. <-- quite, I remember dealing with some old slackware on a 2.4 kernel
23:02:43 <fizzie> "xrandr -q" is a good first step to check that it sees the outputs correctly.
23:02:58 <kallisti> xrandr: Failed to get size of gamma for output default
23:03:24 <Vorpal> value for gamma I would understand
23:03:28 <fizzie> Possibly as in "float, int, or whatever".
23:03:44 <kallisti> $ xrandr -q |& sprunge http://sprunge.us/ecVe
23:03:59 <Vorpal> http://sprunge.us/WXgi <-- my xrandr -q, as a reference
23:04:05 <fizzie> Well, that's not good.
23:04:14 <fizzie> It's seeing only a single output.
23:04:17 <Vorpal> btw this only have LDVS, VGA and DP1 afaik
23:04:21 <Vorpal> the HDMI and DP2 are made up
23:04:26 <kallisti> it worked fine about a week ago in Ubuntu.
23:04:40 <Vorpal> kallisti, reboot to ubuntu and check
23:04:44 <Vorpal> kallisti, what distro are you on now
23:04:59 <kallisti> I guess I could try liveCD or something
23:05:03 <kallisti> but man that's going to take forever.
23:05:07 <fizzie> I was under the assumption that Intel drivers should do XRandR pretty well.
23:05:17 <kallisti> maybe I'm missing some drivers?
23:05:19 <Vorpal> kallisti, hm 1024x768... I guess that one is a 4:3 laptop?
23:05:52 <Vorpal> <kallisti> but man that's going to take forever. <-- how will a livecd take forever?
23:06:18 <kallisti> you must have amazing internet.
23:06:18 <Vorpal> and that is assuming a slow computer
23:06:30 <Vorpal> kallisti, oh I thought you had a copy locally
23:06:39 <Vorpal> then an hour or so to download it
23:06:40 <fizzie> I don't suppose there's any official "fn-f7"-style display-mangling key? (Alternatively, X startup logs about the video bits can also help.)
23:06:51 <kallisti> there is... but it doesn't work now.
23:06:52 <Vorpal> 5 minutes to copy to USB stick
23:07:01 <kallisti> it's some kind of software key.
23:07:37 <Vorpal> kallisti, the hw (computer or monitor) could have died, or you are now using a buggy version of the kernel
23:07:59 <kallisti> http://askubuntu.com/questions/14299/my-monitor-is-plugged-into-vga-0-why-it-is-giving-me-errors-about-vga-1
23:08:09 <kallisti> which says switch from kernel mode setting to user
23:08:54 <Vorpal> kallisti, that is for radeon, which is just crazy usually
23:09:53 <kallisti> I'm beginning to think that user mode setting is not the solution here though
23:10:04 <Vorpal> kallisti, anyway radeon /is/ crazy, I should know, though I use the catalyst driver
23:10:06 <fizzie> Vorpal: Hour? Downloaded ubuntu-11.10-desktop-amd64.iso the other day to test the internets; 6.7 MiB/s at home, a bit less than two minutes for the CD.
23:10:26 <Vorpal> fizzie, hm, how large is that image?
23:10:37 <Vorpal> I thought it was a full-sized CD...
23:10:50 <kallisti> what was the name of that graphical utility?
23:10:58 <Vorpal> hm must have been using bt to download it last time
23:10:59 <olsner> fizzie: that's like ... A HUNDRED SECONDS
23:11:02 <Vorpal> fizzie, and that took about an hour
23:11:07 <Vorpal> at least I know it wasn't corrupt
23:11:12 <fizzie> 730/6.7 is around 108.
23:11:21 -!- itidus21 has quit (Ping timeout: 245 seconds).
23:11:27 <kallisti> mine's estimated at around 50 minutes currently
23:11:35 <Vorpal> fizzie, oh you have 6.7 MiByte/s?
23:11:59 <kallisti> youtube has this fancy thing that compares your internet speed to everyone elses
23:12:29 <fizzie> olsner: Thankfully it was 78.7 MiB/s at work; so if I'm in a hurry and can't wait that two minutes, I can go there to download it. Shouldn't take more than a hour to get to work and back.
23:12:45 <olsner> kallisti: 50 minutes, for one cd? so you have about 1-1.5MBit then?
23:13:40 <Vorpal> olsner, how long would it take with a 300 baud modem (include protocol overhead, because it is going to be quite significant in absolute terms in this case I bet)
23:13:40 -!- itidus21 has joined.
23:14:00 <olsner> hmm, I don't think my office has gigabit internet, probably at most 100Mbit
23:14:48 <kallisti> I have no ethernet access at home currently
23:15:00 <Vorpal> I wouldn't survive that
23:15:11 <Vorpal> I use that for my laptop too
23:15:15 <olsner> Vorpal: 300 baud? depends on how many symbols you're using for the encoding
23:15:27 <Vorpal> because it is more stable, and faster when transferring between the desktop and laptop
23:15:48 <Vorpal> olsner, hm good point, what were the standard coding at the time those were modern?
23:15:58 <fizzie> Vorpal: I liked the 6.7 MiB/s home-speed especially since the contract is for 50 Mbps == 5.9 MiB/s.
23:16:41 <kallisti> hm, is it possible that I need some driver to get this monitor to work correctly?
23:16:50 -!- itidus20 has joined.
23:17:08 <Vorpal> kallisti, since it is intel graphics: no
23:17:16 <fizzie> The VDSL box reports it negotiates the link at around 76000kbps. I'm hoping nobody at the ISP is going to notice and downgrade it.
23:17:59 <kallisti> I saw something about i283 in it.
23:18:11 <Vorpal> kallisti, presumably the standard place?
23:18:19 <Vorpal> I doubt debian vs. ubuntu would change it
23:18:44 <kallisti> the messages that appear when Debian is starting up
23:18:54 <olsner> Vorpal: looks like some of the early modems were 300bit/s
23:18:56 <Vorpal> kallisti, not sure what you mean, not the kernel messages?
23:19:09 <olsner> (though I wonder why I'm looking this up on wikipedia for you)
23:19:29 <Vorpal> olsner, I'm conditioning you to become like fizzie
23:19:30 <fizzie> It's all dmesg up until the init.d scripts start.
23:19:51 <fizzie> Don't know where (or if) the initscript outputs go.
23:19:56 -!- itidus21 has quit (Ping timeout: 246 seconds).
23:20:12 <fizzie> They're not usually so interesting.
23:20:33 <olsner> Vorpal: google for 730MB / 300 bit/s now
23:20:34 <kallisti> this one went something like "failed to find i283 symbols. graphics turbo disabled" or something.
23:20:35 <Vorpal> I would see them just above my login prompt
23:20:40 <fizzie> I mean, even when they e.g. load modules, the interesting bits are the kernel messages from the modules.
23:20:53 <Vorpal> olsner, stop trying to condition me into using google!
23:21:19 <olsner> Vorpal: google for 730MB / 300 bit/s now
23:21:30 <Vorpal> (730 MB) / (300 (bit / s)) = 236.253235 days <-- nice though
23:21:53 <Vorpal> kallisti, Now 50 minutes doesn't seem that long does it?
23:21:53 <kallisti> [ 17.356635] intel ips 0000:00:1f.6: failed to get i915 symbols, graphics turbo disabled
23:21:56 <kallisti> [ 22.553315] intel ips 0000:00:1f.6: i915 driver attached, reenabling gpu turbo
23:22:05 <Vorpal> kallisti, stop whining you modern kids...
23:22:16 <fizzie> Doesn't look too problematic.
23:22:39 <fizzie> How about the Xorg.0.log's "intel" lines?
23:23:14 <fizzie> (And/or xrandr -q if you changeded something, I suppose.)
23:25:13 <kallisti> $ sudo find /var/log/ -exec grep i915 {} \; | sprunge http://sprunge.us/JZcU
23:27:25 <fizzie> Was this still without KMS?
23:30:05 -!- Vorpal has quit (Ping timeout: 246 seconds).
23:30:11 <fizzie> The ignored EDID blocks don't sound too scary. It's only every odd one, after all. But the X logs should list all (i.e. >1) detected outputs.
23:30:58 <kallisti> ...misread. uh, some of it is without KMS and some of it is
23:31:08 <kallisti> I only switched to no-KMS like 30 minutes ago maybe.
23:31:27 <kallisti> but with KMS it would spam invalid EDID messages and everything froze
23:31:33 <kallisti> until I disconnected the monitor.
23:32:54 -!- itidus21 has joined.
23:33:11 <kallisti> so the monitor is definitely sending incorrect information.
23:34:08 -!- itidus20 has quit (Ping timeout: 255 seconds).
23:35:01 <kallisti> I'm not entirely sure why I switched to Debian
23:35:04 <fizzie> It shouldn't really be actively sending anything, just when the card polls it. But I suppose it's possible for the KMS bits to get confused by EDID issues and keep asking.
23:35:32 <kallisti> I could have easily set up xmonad in Ubuntu, and kept all the fancy autoconfigure stuff.
23:37:00 <kallisti> unless the fancy autoconfig stuff is mostly GNOME
23:38:39 <fizzie> There's some amount of Ubuntu involved too, when it comes to drivers and settings and such. Anyhoo. Without KMS, tried plugging to monitor in and then restarting X and seeing if it even notices there's two outputs. (I suppose there's no xorg.conf to confuse matters?)
23:39:04 <kallisti> I haven't touched xorg.conf so, probably not.
23:39:21 <kallisti> ...restarting X would involve restarting firefox
23:39:25 -!- sebbu has quit (Ping timeout: 244 seconds).
23:39:44 <kallisti> perhaps I should have set up screen with wget. :P
23:41:22 -!- sebbu has joined.
23:41:22 -!- sebbu has quit (Changing host).
23:41:22 -!- sebbu has joined.
23:41:51 <kallisti> hm, I could switch to bittorrent. :>
23:43:32 <fizzie> It's possible X might notice something new at startup time. Though I think "xrandr -q" itself should make it "rescan monitors".
23:44:37 <fizzie> "arandr" was the lightweight XRandR GUI, but it doesn't do anything magical.
23:45:34 <kallisti> screen session with rtorrent resuming where the direct download left off.
23:46:25 <fizzie> (The "ignoring invalid edid block" of every odd number just sounds like the monitor had all modes in two different formats, the correct one and (maybe for compatibility?) some funky one.)
23:48:38 <kallisti> no search seems to bring up any linux related issues specifically with the monitor.
23:49:25 <fizzie> Regarding your TURBO, "Note: If you have a first generation Core i{3,5,7} series processor with an integrated GPU, failure to add i915 to the MODULES array in /etc/mkinitcpio.conf will likely cause the error kernel: intel ips [...]: failed to get i915 symbols, graphics turbo disabled."
23:49:42 <fizzie> (The instruction was for Arch.)
23:50:13 <fizzie> Anyway, but it's probably that it doesn't have the i915 module available at that early time.
23:50:28 <kallisti> also I've been having issues with ACPI
23:50:38 <kallisti> my laptop won't suspend when I close the lid.
23:50:42 -!- myndzi\ has quit (Ping timeout: 248 seconds).
23:51:14 <fizzie> Still, it doesn't sound like it should be a *fatal* error, and esp. not one that'd cause things to hang.
23:51:44 -!- blackwhit has joined.
23:53:06 <kallisti> [ 17.749133] ACPI: resource 0000:00:1f.3 [io 0x1840-0x185f] conflicts with ACPI region SMBI [io 0x1840-0x184f]
23:53:08 -!- blackwhit has left.
23:53:20 <kallisti> [ 17.749139] ACPI: If an ACPI driver is available for this device, you should use it instead of the native driver
23:53:55 <kallisti> I don't think there is one though... I wouldn't know what to look for.
23:54:41 <fizzie> You could lspci |grep 1f.3 to maybe at least see which device it's about.
23:55:38 <fizzie> I seem to recall encountering some dangerous-sounding ACPI resource conflict problems too, though I think that was just for hardware monitoring sensors.
23:58:10 -!- kallisti has quit (Ping timeout: 248 seconds).
23:59:49 <fizzie> I would guesstimate it's about the same issue as https://bugzilla.kernel.org/show_bug.cgi?id=12376