| --- Log opened Sat Jun 15 00:00:31 2013 |
| 00:03 | < AnnoDomini> | I had to NAP my Debian every time I wanted to upgrade to a new version. |
| 00:03 | < AnnoDomini> | Because of all the proprietary, third-party, unsupported and plain strange packages I use. |
| 00:15 | | jeff [NSwebIRC@2D9871.A95144.1903C6.DB9FFC] has quit [Ping timeout: 121 seconds] |
| 00:19 | | Turaiel[Offline] is now known as Turaiel |
| 00:37 | | Derakon[AFK] is now known as Derakon |
| 00:38 | | mode/#code [+ao Derakon Derakon] by ChanServ |
| 00:42 | <@McMartin> | Yeah |
| 00:42 | <@McMartin> | And every three months or so my gentoo would tie its configurations in knots, so I had to NAP that every time they updated baselayout |
| 00:43 | <@McMartin> | Which is about when I switched to swapping between Fedora and Ubuntu, which was of course a NAP of everything that wasn't /home every time |
| 00:43 | <@McMartin> | And then until those two got their upgrade-in-place stuff working, NAP was actually the recommended upgrade technique, so~ |
| 00:49 | | You're now known as TheWatcher[T-2] |
| 01:01 | | celticminstrel [celticminst@Nightstar-e83b3651.cable.rogers.com] has joined #code |
| 01:01 | | mode/#code [+o celticminstrel] by ChanServ |
| 01:03 | | You're now known as TheWatcher[zZzZ] |
| 01:15 | <@celticminstrel> | Why do my java applications insist on running with Java 6 instead of 7? |
| 01:17 | <@McMartin> | If Debian tracks Ubuntu, then because installing the openjdk-7 package installs both JDK6 and JDK7 and selects 7 as the default in java-altarnatives. |
| 01:17 | <@McMartin> | Er, 6 |
| 01:18 | <@celticminstrel> | Well, I'm on Mac. |
| 01:18 | <@celticminstrel> | And java-alternatives does not exist, at least not in PATH. |
| 01:19 | <@celticminstrel> | Though I would be unsurprised if the problem is very similar. :/ |
| 01:26 | <@celticminstrel> | When run from the command-line, I do get Java 7. Maybe it's something to do with the app bundles...? |
| 01:27 | <@McMartin> | ... yeah, that sounds much more likely |
| 01:27 | <@McMartin> | Bundles or, uh, whatever OSX calls file associations |
| 01:28 | <@celticminstrel> | Hm... |
| 01:29 | <@celticminstrel> | Changing JVMVersion from "1.6+" to "1.7+" brings up an error saying it's not installed. |
| 01:30 | <@celticminstrel> | Maybe if I had a Java app bundle that requires Java 7 to compare... |
| 01:30 | <@celticminstrel> | ...unless it's something with the JavaApplicationStub thing... |
| 01:31 | < ktemkin> | celticminstrel: echo $JAVA_HOME? |
| 01:31 | <@celticminstrel> | Blank |
| 01:32 | <@celticminstrel> | Okay, so I can do "java -jar" and it works, but that's annoying. |
| 01:33 | < ktemkin> | IIRC, JAVA_HOME is what selects the default JRE; but I don't know if that holds for GUI applications on OS X. |
| 01:33 | <@celticminstrel> | I don't think it does... |
| 01:33 | < ktemkin> | (You could try redefining it in .profile.) |
| 01:33 | <@celticminstrel> | Anyway, on the command-line, it does use 7. |
| 01:33 | <@celticminstrel> | It's only the GUI things that use 6. |
| 01:34 | <@celticminstrel> | ...I seem to vaguely recall getting Eclipse to run in Java 7. |
| 01:35 | <@celticminstrel> | Apparently not. |
| 01:39 | | RichyB [RichyB@D553D1.68E9F7.02BB7C.3AF784] has quit [[NS] Quit: Gone.] |
| 01:41 | < ktemkin> | ls -lah /System/Library/Frameworks/JavaVM.framwork/Versions/CurrentJDK ? |
| 01:41 | <@celticminstrel> | No such file or directory |
| 01:42 | | RichyB [RichyB@D553D1.68E9F7.02BB7C.3AF784] has joined #code |
| 01:47 | | * celticminstrel rages for Yahoo for showing error messages instead of content if you refuse its cookie. |
| 01:47 | <@celticminstrel> | ^at, not for. How did I type for? |
| 01:58 | | Derakon is now known as Derakon[AFK} |
| 02:00 | <@Rodney> | Well... My laptop is running a lot smoother now. |
| 02:01 | <@Rodney> | Even if I think sudo apt-get install foo is a ridiculous invocation. >.> |
| 02:03 | <@celticminstrel> | ? |
| 02:05 | <@Rodney> | Long invocation is long. |
| 02:06 | <@Rodney> | Firstworldproblems I know >.> |
| 02:06 | <&ToxicFrog> | It's not really going to be shorter in any distro. |
| 02:06 | <@Rodney> | pacman -S :p |
| 02:06 | <&ToxicFrog> | sudo pacman -S, presumably |
| 02:07 | <@Rodney> | Ah, true. I'm used to using yaourt. |
| 02:07 | <@Rodney> | yaourt -S :) |
| 02:07 | <&ToxicFrog> | But, ok, I'm willing to grant that it's fewer characters if you use -flags instead of words. I like having an easy to remember set of commands to the package manager, though. |
| 02:07 | <&ToxicFrog> | I certainly wouldn't think of "-S" as "install" |
| 02:07 | < [R]> | Fuck yaourt's weird ass spelling |
| 02:08 | < [R]> | -S is sync actually |
| 02:08 | <@Rodney> | No. -y is sYnc. |
| 02:08 | < [R]> | -S, --sync |
| 02:08 | <@Rodney> | Oh. |
| 02:09 | | * [R] slaps Rodney stupid. |
| 02:09 | | Rodney is now known as Tarinaky |
| 02:09 | <@Tarinaky> | ToxicFrog: Does apt have -flags? |
| 02:09 | <&ToxicFrog> | Yes, but not for the actual verbs. Likewise zypper. |
| 02:09 | < [R]> | -y is acutally "refresh" |
| 02:10 | | Derakon[AFK} is now known as Derakon |
| 02:10 | < sshine> | I have some aliases, pacs="pacman -Ss", paci="sudo pacman -S", pacu="sudo pacman -Syu" |
| 02:12 | <&ToxicFrog> | At this point I am pretty comfortable saying "death to pacman"~ |
| 02:13 | <@Tarinaky> | I've moved away from Arch because nothing has cuased more stress and misery in my life than the Arch devs. |
| 02:13 | <@Tarinaky> | Well, as long as you discount my father. |
| 02:13 | | Typherix is now known as Typh|offline |
| 02:13 | <&ToxicFrog> | Hell, zypper allows abbreviations so it's just as short and way more memorable |
| 02:13 | <&ToxicFrog> | zypper in == zypper install, etc |
| 02:14 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has quit [Ping timeout: 121 seconds] |
| 02:15 | | * sshine would say that as far as distros that require a load of your free time, arch gives the most freedom. |
| 02:16 | <@Tarinaky> | I'm currently trying lubuntu. |
| 02:16 | < sshine> | or I don't know. I just really like using it. it's either this and spending time making everything work from scratch, or using xubuntu using crappy defaults. |
| 02:17 | <&ToxicFrog> | I'm generally in favour of stuff Just Working and arch fails the shit out of that test. |
| 02:17 | <@Tarinaky> | The real issue though is that I have no idea where the config files are... Took me over an hour to figure out how to configure grub2. |
| 02:17 | <&ToxicFrog> | On the plus side, the Arch wiki has been super useful to me. |
| 02:18 | <@Tarinaky> | Yeah. Lubuntu does, actually, just work - aside from some edge cases like my awkward wireless drivers and needing suspend-to-disk. |
| 02:19 | <@Tarinaky> | I quite like it. |
| 02:20 | < sshine> | ToxicFrog, yeah. I could work both with GNOME or a vanilla xmonad, but my win+space toggle to get a drop-down shell is just super useful. |
| 02:20 | <&ToxicFrog> | You can do that on any distro, though |
| 02:20 | <&ToxicFrog> | Just install guake/yakuake/tilda and bind it to something |
| 02:21 | < sshine> | Tarinaky, I agree that grub2 isn't easy to get up and running in arch. and I usually end up outside of LVM because the guides don't agree on what's best and only later do I find out I should have gone with it. ;) |
| 02:21 | <@Tarinaky> | sshine: I've not tried to set up grub2 on Arch. |
| 02:21 | <@Tarinaky> | I stopped updating arch before that happened. |
| 02:21 | < sshine> | ToxicFrog, yup. I used yakuake for a while, but I really much prefer xmonad's scratchpads. |
| 02:22 | <&ToxicFrog> | I've been using guake quite happily. |
| 02:23 | | * McMartin pushes more exercise code, half of which doesn't actually work. |
| 02:27 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has joined #code |
| 02:27 | | mode/#code [+o himi] by ChanServ |
| 02:31 | < [R]> | Extlinux > GRUB2 |
| 02:31 | < [R]> | srsly |
| 02:34 | <&ToxicFrog> | grub was nice. grub2 is way the fuck too complicated. |
| 02:34 | <&ToxicFrog> | It does support way more filesystems and drive configurations than extlinux does, which is nice. |
| 02:34 | <&ToxicFrog> | But actually figuring out how to configure it is a pain. |
| 02:36 | | * [R] shrugs |
| 02:36 | | * [R] just has an Ext2 /boot with 32MB |
| 02:37 | < [R]> | Oh, 256MB. But still small. |
| 02:38 | < [R]> | It's not like /boot has to be super complicated. |
| 02:39 | <@Tarinaky> | I actually had ubuntu fail to install in a 33M /boot partition :/ |
| 02:39 | <@Tarinaky> | It defaulted to "install all ze modules". |
| 02:41 | | Typh|offline is now known as Typherix |
| 02:41 | < sshine> | I just didn't bother with a separate /boot... |
| 02:43 | <@Tarinaky> | I'm going to catch some Zs. |
| 02:43 | <@Tarinaky> | Later. |
| 02:46 | <@McMartin> | Welp, the music code looks good, now to add the SFX exercisers and maybe have a go at making fullscreen actually work |
| 02:59 | | Turaiel is now known as TurShower |
| 03:26 | | TurShower is now known as Turaiel |
| 03:34 | | Kindamoody[zZz] is now known as Kindamoody |
| 04:33 | | VirusJTG [VirusJTG@Nightstar-09c31e7a.sta.comporium.net] has quit [[NS] Quit: Program Shutting down] |
| 05:13 | | celticminstrel [celticminst@Nightstar-e83b3651.cable.rogers.com] has quit [[NS] Quit: And lo! The computer falls into a deep sleep, to awake again some other day!] |
| 06:05 | | Typherix is now known as Typh|offline |
| 06:19 | | Turaiel is now known as Turaiel[Offline] |
| 07:03 | | Derakon is now known as Derakon[AFK] |
| 07:33 | | Kindamoody is now known as Kindamoody|out |
| 09:15 | | You're now known as TheWatcher |
| 09:34 | | AverageJoe [evil1@Nightstar-4b668a07.ph.cox.net] has joined #code |
| 10:35 | <@froztbyte> | http://this-plt-life.tumblr.com/post/36425234595/when-somebody-tries-to-explain- javascripts-prototype |
| 11:05 | | AverageJoe [evil1@Nightstar-4b668a07.ph.cox.net] has quit [[NS] Quit: Leaving] |
| 11:12 | <@McMartin> | Seriously, prototypal inheritance is *easier than the other kinds* |
| 12:21 | < sshine> | yes. |
| 12:21 | < sshine> | and no. |
| 12:22 | < sshine> | class-based inheritance reminds me of factories much like commercial products are made. |
| 12:23 | < sshine> | prototype-based inheritance reminds me of cloning, which on the one hand is a simpler notion than a factory making things, but on the other hand, we don't see it a lot in society. |
| 12:23 | < sshine> | I bet mutation-based inheritance is a much simpler concept, since it is like having babies. |
| 12:23 | < sshine> | :-) |
| 12:23 | < sshine> | s/simpler concept/simpler concept to grasp/ -- naturally, not simpler than mere copying. |
| 12:25 | <~Vornicus> | I still haven't properly gotten my head around prototype object systems in teh first place. |
| 12:25 | < sshine> | Vornicus, you mean the implications that prototypes have on system designs? |
| 12:26 | <~Vornicus> | And on how to actually build objects at all. It is all kind of alien to me. |
| 12:26 | < sshine> | Vornicus, to be honest, I'd probably just re-invent classes if I had to do something in JavaScript here and now. :P (i.e., have a few objects with uppercase names that I only derive from) |
| 12:28 | < sshine> | I just know that in JavaScript's case, you can redefine a bunch of properties that messes the objects up, like the method that enumerates an object's properties, or the method that says whether an object has a given property. |
| 12:29 | < sshine> | I read in JavaScript: The Good Parts a bunch of ways you can check if objects have properties and that some of those aren't so good to use compared to others. basically, IIRC, it advised to do this in one particular way. |
| 12:30 | | VirusJTG [VirusJTG@Nightstar-09c31e7a.sta.comporium.net] has joined #code |
| 13:14 | | iamhere [SinGle.Pv@583787.1E88D8.5038DD.E6C53A] has joined #code |
| 13:38 | | iamhere [SinGle.Pv@583787.1E88D8.5038DD.E6C53A] has quit [Killed (Vornicus (Go scam people somewhere else, this is ridiculous.))] |
| 13:39 | | iamhere [SinGle.Pv@583787.1E88D8.5038DD.3D348E] has joined #code |
| 13:41 | | iamhere [SinGle.Pv@583787.1E88D8.5038DD.3D348E] has quit [[NS] Quit: ] |
| 13:51 | | Orthia [orthianz@3CF3A5.E1CD01.B089B9.1E14D1] has quit [[NS] Quit: Going dooooown...] |
| 13:59 | | VirusHome [VirusJTG@Nightstar-09c31e7a.sta.comporium.net] has joined #code |
| 14:00 | | Vornicus [vorn@ServerAdministrator.Nightstar.Net] has quit [[NS] Quit: ] |
| 14:01 | | Pandemic [VirusJTG@Nightstar-09c31e7a.sta.comporium.net] has quit [Ping timeout: 121 seconds] |
| 14:09 | | Kindamoody|out is now known as Kindamoody |
| 14:18 | | Orthia [orthianz@3CF3A5.E1CD01.B089B9.1E14D1] has joined #code |
| 14:18 | | mode/#code [+o Orthia] by ChanServ |
| 14:40 | | greg [greg@Nightstar-25e7e415.rev.sfr.net] has joined #code |
| 14:40 | < greg> | Salut a tous |
| 14:41 | | greg [greg@Nightstar-25e7e415.rev.sfr.net] has quit [[NS] Quit: irc2go] |
| 14:52 | | cpux [cpux@Nightstar-98762b0f.dyn.optonline.net] has joined #code |
| 14:52 | | mode/#code [+o cpux] by ChanServ |
| 15:20 | | Kindamoody is now known as Kindamoody|out |
| 17:06 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has quit [Ping timeout: 121 seconds] |
| 17:19 | | himi [fow035@Nightstar-5d05bada.internode.on.net] has joined #code |
| 17:19 | | mode/#code [+o himi] by ChanServ |
| 17:25 | | jeff [NSwebIRC@2D9871.A95144.2D3AF0.96BD5C] has joined #code |
| 18:00 | | celticminstrel [celticminst@Nightstar-e83b3651.cable.rogers.com] has joined #code |
| 18:01 | | mode/#code [+o celticminstrel] by ChanServ |
| 18:25 | | Typh|offline is now known as Typherix |
| 19:17 | | Kindamoody|out is now known as Kindamoody |
| 19:44 | < sshine> | I am a bit uncertain as to how to characterise the ideal kind of computation performed on a GPGPU. |
| 19:46 | < sshine> | I assume that the computations must preferrably be arithmetic, that there should not be too much jumping around, and that one should not make the assumption of infinite stack/heap. but how does that affect which problems that are nice to solve on GPUs? |
| 19:47 | < sshine> | recursive divide-and-conquer algorithms, algorithms that use an unpredictable amount of memory in a sporadic way, problems that have subproblems of entirely different structure, I imagine, are not good candidates. |
| 19:48 | < sshine> | linear transformations are cf. computer games ideal. but this is not just what GPUs are good for, right? I mean, they don't have to be linear, so long as they are... (I'm missing some clarifying adjective here, I think.) |
| 19:51 | < sshine> | I have the impression that any kind of cascading, array-oriented computation where the problem consists of some large amount of identical sub-problems (proportional to the size of the GPU) is ideal. |
| 19:52 | < sshine> | can anyone recommend some reading on this? |
| 19:52 | < sshine> | after having asked, I realise that there is the article http://en.wikipedia.org/wiki/General-purpose_computing_on_graphics_processing_un its |
| 20:02 | | Derakon[AFK] is now known as Derakon |
| 20:24 | | Kindamoody is now known as Kindamoody[zZz] |
| 20:46 | <@Tarinaky> | wtf? On Ubuntu l<enter> executes ls o.o |
| 20:46 | <@Tarinaky> | I mean... this is what I meant to type in the first place. |
| 20:46 | <@Tarinaky> | I'm just surprised. |
| 20:47 | < Syka> | Tarinaky: check your .bashrc |
| 20:47 | < Syka> | "alias l='ls -CF'" |
| 20:47 | <@Tarinaky> | So it is. |
| 20:47 | < Syka> | also, I have `alias derp='git'` in mine |
| 20:50 | | RichyB [RichyB@D553D1.68E9F7.02BB7C.3AF784] has quit [Ping timeout: 121 seconds] |
| 20:50 | <@Tarinaky> | "I am the man who arranges the blocks that're made by the men from Kazakhstan, they come three weeks late and they don't Tesselate but we're working to Stalin's five year plan..."~ |
| 20:51 | | RichyB [RichyB@D553D1.68E9F7.02BB7C.3AF784] has joined #code |
| 20:56 | | jeff [NSwebIRC@2D9871.A95144.2D3AF0.96BD5C] has quit [[NS] Quit: Page closed] |
| 20:57 | | jeff [NSwebIRC@2D9871.A95144.2D3AF0.96BD5C] has joined #code |
| 20:58 | <@Tarinaky> | Argh. Why won't you snap >.> |
| 20:59 | <@Tamber> | Simple. I just don't care enough. |
| 20:59 | <@Tarinaky> | I meant inkscape. |
| 20:59 | <@Tamber> | Oh |
| 21:03 | <@Tarinaky> | I blame inkscape. |
| 21:03 | <@Tarinaky> | I don't know what for yet >.> |
| 21:04 | <@Tamber> | It'll do something to deserve it. |
| 21:11 | | celticminstrel [celticminst@Nightstar-e83b3651.cable.rogers.com] has quit [[NS] Quit: And lo! The computer falls into a deep sleep, to awake again some other day!] |
| 22:15 | | ktemkin is now known as Veryone |
| 22:16 | | Veryone is now known as ktemkin[work] |
| 22:41 | <&ToxicFrog> | sshine: you want something embarassingly parallel with either no branching, or branches that each thread will take the same direction in each time. |
| 22:49 | | Vornicus [vorn@ServerAdministrator.Nightstar.Net] has joined #code |
| 22:49 | | mode/#code [+qo Vornicus Vornicus] by ChanServ |
| 22:51 | <@Tarinaky> | Argh. I fucking hate rebuilding my laptop :/ |
| 22:52 | <@Tarinaky> | Autocutsel isn't working and for some reason my .vim backup has failed to actually work :/ |
| 22:55 | <@Tarinaky> | Pffft... fucking git :/ |
| 22:57 | <@Tarinaky> | Apparently I'm missing a .gitmodules file :/ |
| 22:59 | <@Tarinaky> | Halp >.< |
| 22:59 | <&ToxicFrog> | .gitmodules defines the locations of the various submodules you're using |
| 23:00 | <&ToxicFrog> | It should be checked in |
| 23:03 | <@Tarinaky> | Well apparently it wasn't... |
| 23:06 | <@Tarinaky> | I just cloned the repositories I was missing into place... |
| 23:06 | <@Tarinaky> | But I still don't have any clue where I'm meant to be getting this submodule archive :/ |
| 23:06 | <@Tarinaky> | Err *file |
| 23:06 | <&ToxicFrog> | What is the context? |
| 23:07 | <&ToxicFrog> | .gitmodules is a sort of "base configuration" for all the submodules used by a repo, and like .gitignore it should be checked in by whoever created it. |
| 23:08 | <@Tarinaky> | I don't have one. |
| 23:08 | <@Tarinaky> | tarinaky@Mirkwood:~/.vim (master)$ ls -a |
| 23:08 | <@Tarinaky> | . autoload chars .git nohup.out syntax vimrc |
| 23:08 | <@Tarinaky> | .. bundle filetype.vim .gitignore README.md update.sh |
| 23:08 | <&ToxicFrog> | People cloning the repo can then use "git submodule init" to load that info into .git/config, edit it as needed, and then "git submodule update" to actually sync the submodules. |
| 23:08 | <&ToxicFrog> | Do you know that you should have one? |
| 23:09 | <&ToxicFrog> | If so, report that to the maintainer of the repository. |
| 23:09 | <@Tarinaky> | I am the maintainer of this repository. |
| 23:09 | <&ToxicFrog> | Then you should have added and committed .gitmodules when you created it. |
| 23:09 | <@Tarinaky> | How do I create one? |
| 23:09 | <&ToxicFrog> | With a text editor. |
| 23:09 | <&ToxicFrog> | git help gitmodules |
| 23:09 | <&ToxicFrog> | git help submodule |
| 23:10 | <@Tarinaky> | I don't need submodules strictly. |
| 23:10 | <@Tarinaky> | But it's not letting me commit the contents of the sub-repos directly. |
| 23:10 | <@Tarinaky> | So I'm a bit stuck. |
| 23:10 | <&ToxicFrog> | ...explain what you are doing |
| 23:10 | <&ToxicFrog> | If you aren't using submodules, you don't need .gitmodules |
| 23:11 | <@Tarinaky> | This is a backup of my .vim directory. |
| 23:11 | <@Tarinaky> | Most/all plugins are available from git. |
| 23:12 | <@Tarinaky> | cd bundle; git clone <plugin>; git add <plugin> ? |
| 23:12 | <@Tarinaky> | Adding the directory doesn't want to add it to my backup. |
| 23:13 | <&ToxicFrog> | Ok, so the issue is that you aren't using submodules but you should be |
| 23:13 | <@Tarinaky> | Right... so how do I fix this >.< |
| 23:13 | <&ToxicFrog> | You can add the directory if you delete the .git directory from inside it, I think, but you probably shouldn't be doing that |
| 23:14 | <&ToxicFrog> | Instead, 'git submodule add <repo url> <path>' for each one |
| 23:15 | <@Tarinaky> | What's path? |
| 23:15 | <&ToxicFrog> | Where you want it to be cloned into |
| 23:15 | <&ToxicFrog> | Same as git clone |
| 23:15 | <&ToxicFrog> | Anyways, this does two things: clones <url> into <path>, and records the submodule information in .gitmodules. |
| 23:16 | <&ToxicFrog> | Commit those changes and your repo now stores the set of all submodules and what commits they're at. |
| 23:17 | <&ToxicFrog> | Then when cloning onto another machine, clone normally, git submodule init, git submodule update |
| 23:17 | <&ToxicFrog> | And that'll automatically clone all of those repos into the right directories and check out the right commits |
| 23:19 | <@Tarinaky> | Right. What a f***ing nightmare. |
| 23:20 | <@Tarinaky> | I'm glad I was only dealing with 4 plugins and not 40 ::/ |
| 23:21 | <&ToxicFrog> | You could probably write a bash script to do it for you in that case~ |
| 23:22 | <@Tarinaky> | The issue is I'd still need to manually search google 40 times. |
| 23:22 | <@Tarinaky> | Just to build the array for the script to run over. |
| 23:23 | <@Tarinaky> | The issue is more that at no point did git think that maybe forgetting to add .gitmodule /might/ have been an error... |
| 23:23 | <@Tarinaky> | But it can throw a wobbly about git config >.> |
| 23:30 | <&ToxicFrog> | What? Why? |
| 23:31 | <&ToxicFrog> | find . -type d -name .git -print0 | xargs -0 <do a thing> |
| 23:32 | <@Tarinaky> | ToxicFrog: That requires me to first clone the missing plugins. |
| 23:32 | <@Tarinaky> | Getting the repo url requires google. |
| 23:32 | <&ToxicFrog> | And, yeah, (a) git will refuse to unrecoverably discard changes but it won't remind you to commit stuff except by listing it in "git status" and (b) if you aren't using 'git submodule' you wouldn't have a .gitmodules file to commit in the first place, so what would it warn you about? |
| 23:32 | <&ToxicFrog> | Oh. I kind of assumed you already had them cloned, from <Tarinaky> But it's not letting me commit the contents of the sub-repos directly. |
| 23:32 | <@Tarinaky> | I don't remember when I made the original repo. |
| 23:33 | <@Tarinaky> | I only discovered the problem when I tried restoring from the backup. |
| 23:33 | <@Tarinaky> | As is common when dealing with broken back ups. |
| 23:33 | <&ToxicFrog> | This is why you test the integrity of the backup while you still have the original around. |
| 23:33 | <&ToxicFrog> | (and use actual backup software rather than a version control system) |
| 23:34 | <@Tarinaky> | There are very good reasons for using a version control system to back up UNIX config files. |
| 23:34 | <@Tarinaky> | Particularly with the existance of github >.> |
| 23:34 | <&ToxicFrog> | I use git to sync my config files across systems but you'd better believe I don't rely on that as my backup for ~/ or /etc |
| 23:35 | <@Tarinaky> | Yes. Well. That's not relevant to the discussion. |
| 23:35 | <@Tarinaky> | Because I haven't. And I'm not going to. |
| 23:36 | <&ToxicFrog> | (also, github works great for hosting your config files until you need to put passwords in them) |
| 23:37 | <@Tarinaky> | I... can't think of a reason I'd put a password in my vim config file. |
| 23:37 | <@Tarinaky> | Isn't this kind of thing what rsa is for? |
| 23:38 | <@Tarinaky> | Putting passwords in plaintext config files is bad no matter how you slice it >.< |
| 23:40 | <@Tamber> | If an attacker can get into your home directory and read your config files, you're already in trouble. |
| 23:40 | <@Tarinaky> | Really don't understand why external harddrives need mains power instead of just running off usb voltage :/ |
| 23:41 | <@Tarinaky> | It's only a harddrive >.< |
| 23:41 | <@Tamber> | Get a smaller one, then. :) |
| 23:41 | <@Tamber> | And, probably, be prepared to use a Y cable to use 2 ports to power it. |
| 23:42 | <@Tarinaky> | They cost ?30 to ?20 more for a 'portable' harddrive and it's not like they need an order more power. |
| 23:42 | <@Tamber> | It's not about the voltage, but the current the USB ports will supply. |
| 23:42 | <@Tarinaky> | ie Power. |
| 23:42 | <&ToxicFrog> | Tamber: the problem isn't voltage, it's current |
| 23:43 | <@Tamber> | ToxicParrot? ;) |
| 23:43 | <@Tarinaky> | Anyway. I would be happy with using two usb ports. |
| 23:43 | <&ToxicFrog> | USB1 spec limits you to 100mA |
| 23:43 | <@Tarinaky> | I have more usb ports than power sockets. |
| 23:43 | <&ToxicFrog> | Er, USB2 |
| 23:44 | <@Tamber> | With up to 500mA if you communicate enough to request it, isn't it? |
| 23:44 | <~Vornicus> | 500mA; you start with 100 and can request more, up to 500 |
| 23:44 | <&ToxicFrog> | A later update permits (but does not require) the existence of "charging ports" that max out at 1.5A with concurrent data or 5A with no data |
| 23:45 | <~Vornicus> | Why is getting this C++ code out so hard. |
| 23:45 | | * Tarinaky watches the file names of all his pornography scroll by >.> |
| 23:45 | <~Vornicus> | ...probably because the python code I'm basing it on is itself Pretty Terrible |
| 23:46 | <@Tarinaky> | I think, maybe, I regret using rsync with the --progress flag. |
| 23:47 | <&ToxicFrog> | Anyways, point is, spinning platter hard drives tend to require more power than a USB port can provide; even if their requirements fall within the USB "high power" range, there's no guarantee the specific port they're connected to supports that. |
| 23:48 | <&ToxicFrog> | The exceptions tend to be laptop drives, which means more $/GB |
| 23:48 | <&ToxicFrog> | Or SSDs, which are even more expensive. |
| 23:48 | <&ToxicFrog> | Also, Tarinaky, re passwords: not all services support RSA. |
| 23:49 | <&ToxicFrog> | If you're using fetchmail or getmail you need a file somewhere with the passwords in it, for example. |
| 23:49 | <@Tarinaky> | Yeah, but you don't store them in a config file. |
| 23:49 | <@Tarinaky> | You have a special, encypted, passwords file. |
| 23:49 | <@Tamber> | Which then everything needs to know how to access, and open. |
| 23:50 | <&ToxicFrog> | If you're using a chat client, the password is stored in a recoverable way because the client needs to be able to use it to authenticate you. |
| 23:50 | <&ToxicFrog> | Etc. |
| 23:50 | <@Tarinaky> | It's still not stored in the config file. |
| 23:50 | <&ToxicFrog> | Tarinaky: ...which means the keys need to be stored somewhere, and now all you've done is added a layer of indirection to the problem |
| 23:50 | <@Tamber> | Tarinaky, "libpurple" |
| 23:50 | <@Tarinaky> | It's a very useful layer of indirection. |
| 23:51 | <&ToxicFrog> | Also, this assumes you have the time, knowledge and will to make custom versions of everything you use to support this. |
| 23:51 | <@Tarinaky> | Means you can share default configuration in a default profile and keep the passwords in the actual user profiles for example. |
| 23:51 | <@Tarinaky> | If your OS supports multiple user profiles. |
| 23:51 | <@Tamber> | Passwords are stored in accounts.xml in your .purple. As the purple devs put it: What's the point. |
| 23:51 | <&ToxicFrog> | Or...you could fix the problem at the root and use encrypted offsite backups. |
| 23:51 | <@Tamber> | You have to store the passwords *somewhere* |
| 23:52 | <@Tarinaky> | Tamber: That's not the point. The point was storing them in the configuration file. |
| 23:52 | <&ToxicFrog> | Rather than manually patching every single program you use that needs to automatically authenticate to a remote service to fix the, fundamentally, unrelated problem of "I want to store my backups somewhere that everyone can read them" |
| 23:52 | <@Tamber> | Tarinaky, and suddenly when it's in a different file, there is no problem? |
| 23:52 | <@Tamber> | You've just moved the problem elsewhere, made life harder for the applications, for little/no benefit. |
| 23:52 | <@Tarinaky> | Tamber: It means you can backup your config files without worrying about a password being in them.# |
| 23:53 | <@Tarinaky> | Because they're a seperate file. |
| 23:53 | <@Tarinaky> | This is what we were discussing. |
| 23:53 | <@Tamber> | ...why would I worry about my password being in a backup? |
| 23:53 | <@Tarinaky> | No more, no less. |
| 23:53 | <@Tamber> | Unless, of course, you're uploading that backup to the web unencrypted. |
| 23:53 | <@Tarinaky> | Because we were discussing using github to backup the plaintext config files of a well known plain text editor. |
| 23:53 | <&ToxicFrog> | Tarinaky: except this is only a problem if your backup solution is fucked1 |
| 23:54 | <@Tamber> | In which case, the proper solution is: Then Don't Fucking Do That, Then. |
| 23:54 | <&ToxicFrog> | It adds a great deal of client-side complexity to solve a problem that should not exist. |
| 23:54 | <@Tarinaky> | Tamber: Pretty much. |
| 23:54 | <@Tarinaky> | Tamber: That was the first thing I said. |
| 23:54 | <@Tarinaky> | Except in one word. |
| 23:56 | <&ToxicFrog> | ...except you are, in fact, Fucking Doing That. |
| 23:56 | <@Tarinaky> | I don't have any passwords in my vim config files. |
| 23:57 | <&ToxicFrog> | Right. |
| 23:57 | <&ToxicFrog> | My point was that github is not a viable backup solution in general. |
| 23:58 | <@Tarinaky> | Nobody argued against that. |
| 23:59 | <@Tarinaky> | My 'general' backup solution is an external harddrive in a bag. |
| 23:59 | <@Tarinaky> | Usually kept in the same building as the computer. |
| 23:59 | <@Tamber> | My point was that worrying about passwords being in config files is pointless; unless you're doing something daft. |
| 23:59 | <@Tarinaky> | But that's what I get for not having an enterprise budget :p |
| --- Log closed Sun Jun 16 00:00:46 2013 |