=== Lcawte|Away is now known as Lcawte === Lcawte is now known as Lcawte|Away [08:08] cool - the Google Doodle for today is a working turing machine [08:08] not as cool as it could have been (i.e. fully programmable) but they turned it into a nice game :) [08:09] they solved the halting problem as well [08:09] ali1234: that's google for ya ;) [08:09] ali1234: They use Linux, which does infinite loops in under 5 seconds [08:09] ali1234: the halting problem can't be much harder [08:20] morning [08:21] morning o/ [08:22] what do to with the weekend? [08:26] jacobw: watch david bowie on iplayer/ [08:36] i am a bowie fan [08:37] jacobw: there was 4.5 hours of him on TV last night [08:37] \o/ BBC Four :) [08:40] * jacobw works out the vpn details [08:40] you're not in UK-land? [08:41] i moved to bavaria about a month ago [08:41] ah yes you said [08:42] ich kann nicht sprache deutsche jetze [08:43] aber.. ich learnen [08:43] DEUTSCH *bleep*, SPRECHST DU ES? [08:43] I wonder if they really did dub Pulp Fiction like that [08:43] hehe [08:44] i'm supporting germany in the soccer [08:44] Are they still in? I don't follow it really [08:44] yeah, they've just knocked greece out of the euro ;) [08:47] i'm not optimistic oabout england vs italy [08:47] i don't think they'll be an england vs germany match [08:48] they kicked "our" behind [08:48] but Oranje sucked anyway [08:48] (which would be the case if england beat italy) [08:49] i watched netherlands vs denmark a few weeks ago [08:49] somewhat suprising [08:49] they missed _every_ shot [08:53] i wasn't expecting england to beat sweden [08:54] it's inconistent with the ukraine performance :| [08:54] * jacobw thinks that rooney breaks the dynamic [08:55] I have no idea [08:56] i'm not usually interested in football [08:57] people here keep talking to me about it, so i've formed some opinions :| [08:57] jacobw: ear plugs ;) [08:58] and i have a colleague here with a 60" TV [08:58] which is bloody huge :p [08:59] I have a 48" one which is huge already [08:59] so 60" ... wut [09:01] it's crazy, SD looks so bad on 60" [09:01] you have to have HD at 60" [09:02] yay satellite ;) [09:02] also yay bluray [09:03] https://upload.wikimedia.org/wikipedia/commons/thumb/d/de/2005-05-15-raisting_900x460.jpg/800px-2005-05-15-raisting_900x460.jpg [09:03] this is the place where i live in bavaria [09:04] jacobw: you should get good reception on those :) [09:05] :D [09:05] satellite has alot of mind share here [09:06] yeah, they just switched off analog TV and got a bunch of new HD channels [09:06] and they're free-to-air (like the BBC/Freesat) [09:06] so _I_ get new HD channels as well [09:09] good morning everyone. [09:09] morning [09:10] morning jacobw [09:12] euro gp this afternoon :) [09:12] * jacobw is looking forward to watching paul di resta [09:27] oh look, here is my Terry Pratchett book :) [09:38] Good morning peeps :) [09:39] morning bigcalm :) [09:44] Aha, there is life :) [09:44] And a brief popey [09:55] At least he visited. [09:59] There he is again. Do you think he's real? [10:00] morning [10:00] bigcalm, poke him and find out! [10:00] * bigcalm tickles popey [10:00] * popey giggles [10:01] Spooky [10:03] hows everyone doing? :) [10:04] I have a new toy :) [10:04] How's you popey? [10:04] great :D [10:04] oh? [10:05] Ohhh, http://www.ebuyer.com/222310-seagate-500gb-momentus-xt-2-5-hybrid-ssd-st95005620as [10:05] Looks like I'm going to get a new toy [10:05] ahh, nperry got the ebuyer spam too [10:05] popey, nope I just checked hotdealsuk [10:07] £0.12 per/GB is fine for me. [10:08] bigcalm, what did you get? [10:08] AlanBell, did you get compiz built ? [10:10] popey: no [10:11] can I help? [10:12] hope so :) [10:12] so I have a compiz tree based on lp:compiz with some new bits in it [10:12] it does build using the instructions in the INSTALL file [10:13] by creating a subdirectory called build and doing cmake -i ..; make [10:13] i wouldn't do that [10:13] sure, but that confirms that it does compile [10:13] ok [10:14] so I pull a fresh copy of that and the packaging branch from didrocks into a directory next to each other [10:14] I have bzr-builddeb installed [10:14] my directory is called compiz, the didrocks thing is called ubuntu [10:14] same here [10:15] I want to compile the compiz directory using the packaging of the other one [10:15] cd ubuntu [10:15] I can go into the ubuntu directory and do bzr merge ../compiz and it pulls in the stuff I would expect [10:15] bzr bd [10:15] but then it won't build [10:16] bzr bd does something, but which one is it building? [10:16] isn't it building the ubuntu tree without my stuff? [10:16] popey: https://twitter.com/bigcalm/status/216473741734064129 [10:17] let me test here [10:17] bigcalm, it is sexy - isn't it? [10:19] nperry: veryily :) [10:19] guy at work has one of those, i had a play with it [10:19] not impressed, android still sucks [10:20] :) [10:21] yeah, I want my Ubuntu phone based on python [10:21] or a boot to gecko image on my S2 [10:23] i see the raspberry jam in london happened last week at the moz place [10:23] yeah [10:26] so is your patched compiz in the compiz dir AlanBell ? [10:26] and do dch -i in there to update the changelog [10:26] yeah [10:26] and debuild -S -sa [10:28] whut [10:28] how is there a debian directory in there now? [10:28] oh duh === webpigeo1 is now known as webpigeon [10:31] ls] [10:32] * AlanBell is confused [10:32] This isn't the terminal you are looking for [10:33] if it's a bzr tree, bzr bd -S .. is what i would do. [10:37] bzr: ERROR: unknown command "bd" [10:38] apt-get install bzr-builddeb [10:39] alan@deep-thought:/tmp/test/build-area$ ls -l *.deb | wc -l [10:39] 17 [10:39] \o/ [10:46] sure, but which tree did it build? [10:46] I can build didrocks tree [10:46] * MartijnVdS has set up cowbuilder + cowpoke at work [10:46] \o/ 32 core buildboxes with tons of RAM [10:46] I don't want to build that, I want to build my tree that doesn't have packaging in it [10:46] (and clean chroots) [10:48] do I copy the /debian directory from didrocks compiz tree into my compiz tree? if I do that it complains that it isn't committed but I don't want to add it and commit it or I won't be able to submit a merge proposal [10:48] or do I do some kind of bzr merge ../ubuntu to get the packaging across? [10:48] or is there some other magic command to build one tree based on the packaging in another? [10:49] just commit locally and don't push/send a merge request [10:49] ? [10:49] then after the build, undo the commit? [10:49] [10:50] maybe, I have tried going down that route but got in a bit of a mess [10:50] is building manually (not using bzr-builddeb) an option? [10:52] well that works, but then I don't know how to install and uninstall it and work out whether packaged compiz is running or my new compiz is running [10:53] and also 11:14 < popey> i wouldn't do that [10:54] I'd just change the version, add "~alanbell1" or something to the end and rebuild [10:54] (in debian/changelog) [10:54] it might not be beautiful, but at least you'll know what's running :) [10:54] so you have the source you modified in compiz [10:54] and the packaging branch in ubuntu [10:54] side by side, yes? [10:54] yes [10:55] bzr branch lp:~alanbell/compiz/texttracking compiz [10:56] bzr branch lp:~didrocks/compiz/ubuntu/ [10:56] https://wiki.ubuntu.com/DesktopTeam/Bzr [10:57] see the "updating a package" and the bit above " building a source package" [11:00] always good to see "Don't cry" on a wiki page [11:03] I don't really follow what it is trying to do :( [11:04] is this the first update? [11:04] I am guessing not, other people have updated it, just because it is the first time I have done it that isn't the initial commit [11:05] Ensure that debian/watch is looking at unstable version if available. Otherwise, update it. (uscan --verbose --report to see what is the latest version corresponding to debian/watch regexp) [11:05] that probably means something [11:06] so this goes on to tell you how to make a patch with quilt, which are the things you attach to bugs and nobody looks at them because you should have done a merge proposal instead :( [11:07] I don't want to do that, I just want to compile my code into a package (or source package I can put in a PPA to build) [11:08] the step after that is # do the traditional hack (One change!) [11:08] you should give up on bzr and use compiz upstream instead [11:08] it's much much easier to contribute that way [11:08] bzr is compiz upstream [11:08] git is abandonned [11:09] well, that depends who you ask [11:09] true [11:09] it's more accurate to say that bzr is a fork that has the same name [11:09] since nobody has officially announced that compiz is now officially hosted on launchpad [11:09] but I feel I have a better chance of getting a deb file if I use the infrastructure that everyone uses to make deb files [11:10] that may be true, but what is so good about a deb file? [11:10] so If I just build this thing and install it, I don't know the relationship between what I have just done, and the packaged compiz I also have installed [11:10] do I apt-get remove compiz first? [11:10] so remove the packaged compiz [11:10] no, you apt-get purge compiz first [11:11] oh, ok [11:11] actually compiz-core [11:11] of course unity will then refuse to install because of dependencies [11:11] but that's ok, unity doesn't work with compiz upstream anyway [11:11] yeah, do I have to rebuild all of unity whatever I do? [11:11] no, it won't work even if you do that [11:12] I am thinking at this point of just throwing in the merge proposal without testing it (other than confirming it builds) [11:12] that's what most people do lol [11:12] explains a lot [11:12] thats not true [11:13] AlanBell, speak to sil2100 in -unity on monday [11:13] ok [11:13] he will be happy to help you [11:13] and we can test your change [11:13] what is it? [11:13] text cursor tracking for enhanced zoom [11:14] zoom in, type stuff, it follows you about [11:15] https://github.com/gloob/gloob-Ezoom-fork [11:15] https://github.com/gloob/compiz-accessibility-plugin [11:15] those work as an externally built compiz plugin, I want to get it into Ubuntu [11:15] there's no need to put it in compiz core to do that [11:15] just package it [11:16] if you put in a merge proposal we can look at it [11:17] we're about to ship a new compiz release [11:17] then it would conflict with the ezoom plugin that is in the base package [11:17] why didn't they rename it if it's a fork? [11:18] because it was intended to be merged back in [11:18] which is what I am trying to make happen [11:20] renaming it wouldn't preclude merging back the changes [11:21] well, they didn't, and it has been sat on github for ages not getting into anywhere people can actually use it [11:21] https://code.launchpad.net/~alanbell/compiz/texttracking/+merge/111710 [11:22] AlanBell, is there a bug filed to request this inclusion? [11:22] bug #727290 [11:22] Launchpad bug 727290 in Compiz "Allow enhanced zoom to follow text." [Undecided,In progress] https://launchpad.net/bugs/727290 [11:24] I've added it to our list to track [11:24] but ping sil2100 on monday and let him know you'd like to see it in 12.10 [11:25] i might attempt to backport that to 0.8 just for laughs [11:25] thanks popey [11:28] we'll be a bit busy on monday with SRUs for 12.04, but be good to get it on the radar [11:34] i didnt think anyone was still working on compiz 0.8x [11:41] popey: http://gitweb.compiz.org/?p=compiz/plugins/ezoom;a=log;h=refs/heads/compiz-0.8 [11:50] what happened to [11:50] Quinn Storm [11:57] lol, so "no" [11:59] AlanBell, quinn is still around it seems [11:59] http://ubuntuforums.org/member.php?u=79158 [12:37] hello how do i fine out everything about my pc ramp cpu graphics card etc ? [12:37] kane1309, you using ubuntu 12.04? [12:39] hmmmmmmmmmm [12:40] kane1309, in a terminal you can do "lshw --html > ~/hardware.html" and then you'll get an html formatted report you can open in a browser [12:41] argh this is some grade A microsoft stupidity [12:41] Cool, the GPG key for Spotify's apt repo expired. [12:42] so you can make a chain of trust within your exe by adding additional certificates [12:42] but you can't do it with spc, pfx, or cer files [12:42] you have to import all the certificates into MS certificate store [12:42] \o/ [12:42] lock-in is your friend, ali1234 [12:43] then use a pfx file with the private key and certificate used to sign, and then also specify another command line option for the *root* certificate you want to use. but not the name of it in the certificate store. you have to have it has a .cer file as well [12:43] then it will pull all the intermediate certificates from the cert manager and include them [12:43] so it is possible, it is just extremely long winded [12:44] and this does work in OVMF too [12:44] but now, the command to pick a root certificate for chaining... you can only supply it once to sign tool, even though the specification clearly states there can be MULTIPLE chains in the signature [12:52] ali1234: Is it better or worse than openssl? [12:53] i dunno, can openssl do authenticode signing? [12:53] well there's http://sourceforge.net/projects/osslsigncode/ [12:54] http://wiki.cacert.org/Authenticode ? [12:54] https://developer.mozilla.org/en/Signing_an_executable_with_Authenticode ? [12:54] oh wait [12:54] the last one uses the MS thing [12:55] oh it uses mono [12:55] at least you won't have to boot Win8 :) [13:12] MartijnVdS: the mono code signing tools are too old to sign for EFI [13:20] hmm wait a minute... [13:20] ok someone on stack exchange is really confused [13:21] making a pfx from a spc file does result in all certificates going into the binary [13:21] so... no need t mess with the certificate store [13:21] and no need to use /ac like this guy is saying [13:21] mm [13:21] and no limitation on the number of chains (in theory, i need to test that) [13:21] just back from the west berkshire brewery tour, good fun if you can book for the next tour [13:35] hmm that's not it either [13:36] basically spc files work but it will only process certificates that can be traced back to a certificate in your certificate store [13:42] ah, no! [13:42] it will only add certificates if it can see the certificate of the issuer [13:43] so however much of the chain you put into the SPC file, it always cuts off the first one [13:43] that makes absolutely no sense at all [13:43] but - whatever [13:49] and confirmed with a real test [14:12] Hey ubuntu UK i have a live disk that has files on it i want to copy to how do i read these files? [14:13] I have a filesystem.squashfs file [14:15] mount filesystem.squashfs dir_you_want_to_mount_to [14:15] That should work, irc. [14:18] Oh, You may need to install squashfs-tools danawar2. [14:18] indeed i have just isntall them [14:18] trying your mount method first ^^ [14:19] how would i use a . bat ? [14:19] woops [14:19] i just mounted it to /media [14:19] and unmounted all my other drives [14:22] ???????????' [14:22] Batch files are not for linux [14:22] As far as i am aware [14:22] i no is there a way to convert them [14:23] hrmm not relly because a .bat is a load of windows commands [14:23] kane1309, You can port the commands to linux commands.. [14:23] nperry [14:23] It now says i have no permission to make a copy of the mounted file :/ [14:23] as root [14:24] how to i force ownership of an item [14:24] what do u mean ? [14:25] kane1309: you would need to rewrite the file as a bash script like change commands dir to ls [14:25] that was just an example [14:27] in the bat file it has [14:27] @echo off title Client java -jar -Xmx512m "play.jar" pause [14:27] now what do i do [14:27] danawar2, give this a go.. Not sure if it will work though. mount -t squashfs-o uid=1000 filesystem.squashfs dir_you_want_to_mount_to - I've got to pop out for 20mins. [14:27] ive lost the original now ;/ [14:27] because it unmounted my other drives haha [14:28] kane1309, java -jar -Xmx512m play.jar [14:28] thank you [14:28] As long as play.jar exists and you have java installed. [14:28] !java | kane1309 [14:28] kane1309: To just use java you need a "Java Runtime Environment" (JRE) and/or a browser plugin. If that is not sufficient you will need a "Java Development Kit" (JDK) aka "Software Development Kit" (SDK). Please see https://help.ubuntu.com/community/Java about how to install one of three current implementations. [14:29] What is the damn harddrive manager called un unity xD [14:29] unity ruined my life > . > [14:29] me too [14:29] when you say hard drive manager, do you mean file manager, or partition manager? [14:29] partition [14:29] sorta [14:29] hit super, type disk [14:29] i just wana mount my harddrives lol [14:30] or drive or partition [14:30] disk utility [14:30] you might want to install gparted [14:30] na it ok disk utility was what i wanted :) [14:30] ty [14:30] gona just reset machine [14:30] i think it is easier xD [14:30] than finding out why it wont mount [14:37] Can my CPU really be running at 10 degrees Celsius? Or is that a problem with the lm-sensors package? [14:39] no it can't [14:40] Hmm [14:41] Temperature sensor monitoring in Linux is weaker than in Windows [14:45] kvarley: You most likely have a strange setup, I would think lm-senors works with more hardware than windows [14:49] * bigcalm returns [14:49] o/ bigcalm [14:49] o/ [14:57] popey: free? [15:05] popey: well just slinging in a merge proposal on untested code seems to have worked rather well [15:05] https://code.launchpad.net/~alanbell/compiz/texttracking/+merge/111710 [15:06] lots of great feedback [15:06] just wish I knew C++ a little bit :( [15:10] oh god the tab formatting [15:10] why? why? [15:12] seriously i had no idea anybody could be insane enough to mix tab and space indentation in the same file [15:12] but i guess it makes sense that if anyone were it would be X11 devs [15:12] It happens when you have more than one dev working on the same files [15:12] not like this [15:13] Why vim has the default tab depth set to 8 chars, I have no idea [15:13] the indentation levels are: 4 spaces, tab, tab + 4 spaces, 2 tabs, 2 tabs + 4 spaces - and tab stop is 8 characters [15:13] Tasty [15:15] ali1234: yeah, that is an entertaining convention. On the plus side I have the skills to comply with that bit [15:15] it;s the worst possible way to do it [15:15] all tabs or all spaces will always be indented correctly regardless of tab stop [15:16] and considering that it's a scratch rewrite... i can't possibly imagine why you'd pick that convention on purpose for a scratch rewrite [15:17] I don't think that it's intentional [15:18] it has to be [15:19] ali1234: That actually kind of makes sense [15:19] ali1234: It means it's aligning to 4 character boundaries under the assumption that tabs are 8 character alignment and using tabs when it can [15:19] yes i know [15:19] ali1234: I mean I don't like it, I just can see the reasoning [15:20] the only sane alignment is of course 2 sparces [15:20] it's still incredibly stupid because it enforces a tab size [15:20] it is kind of the worst of all worlds [15:20] the only thing it has going for it is you save 3 byte per indentation level per line [15:21] well, actually 1.5 [15:21] don't you just hate losing half a byte [15:22] not in the source code [15:22] you could same more by notUsingObnoxiouslyLongCamelCaseNamesForEverything [15:23] wellAtLeastCamelCaseUsesLessSpace_Than_obnoxiously_long_underbar_separated_names [15:23] long camel case method names are fun.. [15:23] yeah but it is much harder to read [15:23] nnnnahhhh [15:23] but it's the obnoxiously long part i am against [15:24] nod [15:24] (although have you ever had to work on an entire project with 1 char variable/function names - it's not fun) [15:25] does decompiled machine code count? [15:25] no [15:26] Surely after a while they run out of numbers and letters, [15:27] nperry: I ended up printing out a table of all usable letters and characters allowed for eachof the different variable types [15:27] a = 1; b = 2; function c () { d = a + b; } [15:27] what i don't like is when functions have names that describe what they do and also what the arguments are [15:27] nperry: oh come on; get rid of the excess spaces [15:27] penguin42, It makes for easier reading. [15:27] It was a BBC Basic project (for schools) on Beeb Micros, so about ~12kB usable RAM [15:29] so instead of add(int a, int b); you get addIntegerToIntegerAndReturnTheValue(int the_first_value, int the_second_value); [15:33] how java [15:36] Damn you WordPress for using serialised arrays to hold multiple options in your database. Making replacing strings (like URLs) tricky. [15:37] MartijnVdS: there's a whole class of C programmers who write code like that too [15:38] bigcalm: it's easier than XML! [15:38] If they're storing serialised arrays their database isn't normalised [15:39] directhex: but a serialised array has the char count at the beginning of each value. Eg: ;s:8:"filename"; [15:40] I've been trying my best to use replace() within mysql to move a database from one domain name to another. But sometimes it's just not enough. Will have to write a PHP script to handle the arrays, which I didn't want to do [15:46] ali1234: Those people deserve pain [15:49] ali1234: or at least a stern talking-to === PreciseOne is now known as IdleOne [15:52] they should be forced to work only on projects where people write functions called add which have exactly two integer arguments and that do something other than add them together and return the result [15:53] that way we can get all the terrible coders into the same project and just not use that software [15:54] if we're really lucky the two types of awfulness might cancel each other out in a huge flash of pure energy [15:54] ali1234: No, that's what operator overloading is for, so you can redefine + to do something arbitrary === bou_ is now known as bou [15:57] penguin42: You'll love the ACME:: namespace of Perl modules :) [15:57] penguin42: https://metacpan.org/search?q=acme%3A%3A [15:58] oh - haven't seen that? [15:58] https://metacpan.org/module/Acme::Nooo [15:59] MartijnVdS: Wow, that's some of the clearest, most useful, most readable Perl code I've ever seen [15:59] penguin42: https://metacpan.org/module/Acme::Bleach [15:59] trying to install ubuntu on a windows machine. two options, install from an installer, while using windows, or a longer process involving a usb installer. windows installer seems easier, what advantages in using the longer process? [15:59] popey, cute kitten! [16:00] not sure if my post was complete so i'm going to start again ... [16:00] trying to install ubuntu on a windows machine. [16:00] two options, install from an installer, while using windows, or a longer process involving a usb installer [16:00] windows installer seems easier, what advantages in using the longer process? [16:10] Gah, time to descale the kettle [16:10] that's a chore [16:11] Not really, just unexpected scale in my tea [16:12] it ruins the tea experience [16:15] we don't have that problem up here [16:16] Scotland? [16:18] Manchester [16:19] water filters ftw. [16:20] but soft water tastes yucky [16:20] It's whatever you're used to [16:20] We have Yorkshire tea for hard water, makes a big difference [16:22] Odd, my phone appears in airdroid and my car's bluetooth as a Samsung GT-19300 [16:22] what is it really? [16:22] Samsung Galaxy S III [16:23] Hmmm... [16:25] is that just the part number for one variaton of the S 3 ? [16:25] bigcalm: Yeh, http://phonecomputerreviews.blogspot.co.uk/2012/05/samsung-galaxy-s-iii-gt-19300.html#.T-Xt8hUlg3g [16:28] Don't think I'll ever find a more useful app than AirDroid [16:29] The model number for the S III is the i9300... [16:30] It's an i not a 1, did wonder [16:34] so where does USC log to? [16:46] urgh. has anyone successfully installed psychonauts via USC? it gets to 2.1G downloaded then stops progresses, for me [16:46] extra stupid: aptd doesn't cache partial downloads === Lcawte|Away is now known as Lcawte [16:47] 2.1G !!! [16:48] huh, apparently USC just isn't updating the progress bar at all. stopping at 2.1G smells like 32-bit failure [16:48] bloody python [16:48] i got aptitude to do the install instead [16:48] 80% [1 psychonauts 3,540 MB/4,426 MB 80%] 1,777 kB/s 8min 18s [16:52] oh jesus, the download's fubar [16:53] 100% [1 psychonauts 4,843 MB/4,843 MB 100%] [16:53] er, no [16:53] redo from start :( [17:06] finally worked. [17:08] thankfully it seems psychonauts package was split, so patches to the executable don't mean another 4.4G [17:24] I'm using Precise, and I can't switch keyboard layouts. The indicator shows up when more than one layout is set up, and the icon will change if I select another layout from its menu, but the item with the dot in the menu doesn't change and nor does my effective keyboard layout. I can't change even if I remove the one I'm in from the list, and the extra options don't work. (mentioned this a few weeks ago, but no harm ... [17:24] ... trying again, right?) [17:30] Darael: that's odd, i'm using Precise and can switch keyboard layouts [17:33] jacobw: I've had the same problem on several machines. It seems to be user-account specific, and is happening only on accounts whose default layout before the upgrade was Programmer Dvorak (but the other accounts all had normal UK, so I haven't tested many things). [17:44] directhex: do you happen to know if mono will ever include the new signing tool work-alikes from .NET 4.5? [17:57] what's new in 4.5? [17:58] a new signing tool is one feature [17:58] ;) [17:58] * hamitron is guessing [18:00] directhex: ability to sign HelloWorld.efi. don't ask me why but the signtool.exe from 4.0 won't do it [18:00] i don't know the difference between the existing signing tools in .NET and in Mono [18:00] but really i want to know if mono is ever gonna reimplement signtool.exe the way they reimplemented signcode and chktrust [18:01] er... the difference is microsoft deprecated signcode and chktrust and signtool is the replace ment and it is slightly incompatible (for one thing it does sha256 signing while signcode only does sha1 and md5) [18:01] ali1234, i honestly don't know - a tool to sign EFI binaries isn't really mono's remit as-is, but i'm sure sn.exe could be adjusted [18:01] EFI binaries are PE/COFF [18:02] signcode from mono can already sign them [18:02] it just doesn't sign them "properly" [18:02] where "properly" = however microsoft chooses to interpret it's own spec this week [18:02] file a bug? seems like a great prospect for something to re-use for this job [18:03] i have no idea where [18:03] i actually tried googling this [18:03] bugzilla.xamarin.com [18:03] and i found a ML post from 2008 that linked to ... ^ that website and when i clicked it it just timed out [18:03] so then i gave up [18:04] xamarin didn't exist in 2008... [18:04] ximian? [18:04] ah could have been yes [18:15] ali1234, https://github.com/vathpela/pesign is what RH are using for testing [18:16] yes, i know [18:16] i haven't tried it yet, i'm still figuring out all the idiosyncracies of signtool [18:17] specifically i'm trying to figure out why people keep citing "you can only sign a binary with one key" as a reason for doing what they are doing [18:18] since while it's technically true, it can always be worked around if you have two keys - you just make a certificate for one of them with the other one [18:19] then put both certs in the signature [18:25] i think the bigger issue is whether EFI instances will parse two sigs correctly [18:26] well that's what i'm about to find out [18:26] and the killer for all this stuff isn't the boot loader or kernel anyway - it's firmware. your GPU needs a signed firmware... [18:26] well, actually no - there is only ever one signature - that's the point [18:28] but the signature can contain multiple certificates and EFI certainly handles the case where the key that signed the binary needs more than one cert to chain back to a trusted key [18:28] and according to the spec is should handle redundant certs and multiple chains back to different root keys [18:33] i assume anyone who uses the MS service will get their own signing key and certificate [18:33] otherwise they couldn't revoke a single company's key if it was stolen or something [18:33] ali1234, which efi binary are you signing? [18:33] directhex: HelloWorld.efi from tianocore [18:33] er... edk2 [18:34] ali1234, compiled with what? [18:34] compiled with gcc4.6 [18:34] for 64 bit [18:34] in fact all the instructions you need are at https://wiki.ubuntu.com/UEFI-howto [18:35] so gnu-efi. apparently the problem is that it puts gaps between PE-COFF sections, which most signing tools bork on [18:35] signtool doesn't [18:35] works fine :) [18:36] indeed. but signcode does [18:36] i'm sure there are alternatives that do work, or workarounds [18:36] but at this point i'm still trying to determine if there are problems with the standard reference implementations [18:36] though i don't really believe there are === popey_ is now known as popey [19:14] ok, signtool actually refuses to sign if you provide it with multiple trust roots [19:14] that's pretty lame since it's allowed by the spec [19:27] well, i must be doing something wrong, because i just don't believe this can't be done [20:29] Thats my deed done for the evening... I've passed the corsix-th onto laura cowen... rather than using Virtualbox :D [20:47] Bubba83 [21:35] good night everyone, sleep well. [21:52] is there any use for gconf on ubuntu server? [21:55] dogmatic69, I wouldn't think they'd be much call for it. [21:55] ye same here [21:56] I installed inkscape and figured it was from that [21:56] removed it now, need to use it for .ai -> .svg conversions [22:28] Anyone know where a good place to get distilled or deionised water would be? [22:28] Nobody seems to sell it D: === Lcawte is now known as Lcawte|Away [22:30] Azelphur: I vaguely remember seeing it in a supermarket [22:30] which one? tesco doesn't seem to sell it [22:30] ...for everything else, there's amazon [22:30] yea [22:30] shipping water tends to be expensive though [22:30] Azelphur: chemist? [22:30] how about a health food shop? [22:30] mgdm: maybe, will check some :D [22:31] * penguin42 would agree - a chemist should have it; they sometimes use it for making up stuff from powder [22:31] hmm yahoo answers says boiled water = deionised water, sounds scammy :p [22:32] Azelphur, garage? [22:32] tried the local garage today, nope :( [22:32] yes, well a garage doesn't need something that clean [22:32] * penguin42 guesses Azelphur is putting together a water cooling rid [22:32] g [22:33] penguin42: flushing my existing one [22:33] ah [22:33] prefer finding a decent supply somewhere cheap [22:33] rather than having to keep shipping it xD [22:33] if you're looking for it to pure pure / nonconductive, boiling isn't what you're after. unless you move up to distilling, boiling just kills bio [22:33] yea [22:34] I'm not /too/ worried about leaking, I've reassembled this loop twice now and never had a leak [22:34] and if you are doing that, should be using that equipment for something more "useful" ;) [22:34] haha [22:34] Azelphur: a model shop might have some for steam locos [22:34] don't think we have any around here :( [22:35] there's like 3 chemists though so I'm sure one of them will have one [22:35] I'm supposed to use deionised water in my steam models? [22:35] :/ [22:35] yup http://www.hornby.com/shop/live-steam/r8207-live-steam-distilled-water/ [22:36] * hamitron facedesks [22:36] if you don't, you trains will EXPLODE [22:36] it helps if boiling doesn't leave residue accumulations. I can imagine a miniature engine looking like the inside of my kettle would be less productive [22:36] yeh, I got something to flush it [22:37] gotta clean it all out as my loop has been growing things, http://www.overclock.net/t/1273331/my-water-cooling-loop-seems-to-be-growing-things T: [22:37] D:* [22:37] algae of some sort? [22:37] not unless it's super immune to biocide algae [22:38] I sprayed biocide at it, nothing happened [22:38] Azelphur, have you been to you're local motor factor? [22:38] what's a motor factor? o.O [22:39] like unipart [22:39] that doesn't look like bio to me. sharp corners look like no infection I've ever had (that's not what it sounds like). bio prefers round clusters [22:39] or partco [22:40] nperry: never heard of anything like that *shrug* [22:40] or if all else fails halfords. [22:41] :D [22:41] They don't sell deionised waters anymore in supermarkets [22:41] As irons don't really need it anymore. [22:41] hehe [22:41] I found out boots don't sell homebrew equipment either. and you really can't buy most chemicals at a chemists anymore. what is the world coming to [22:41] £3.99 for 5l of water. [22:41] there's more picture of the life forms I'm growing further onto the thread btw [22:43] Azelphur: I agree, that doesn't look biological [22:44] mind, you I guess nice does curding cheese, but still [22:44] the structure looks more like flakes than growths. if it is bio, you've got something much bigger growing inside [22:45] you mean those are only the babies and you don't want to see it's dad? [22:45] well flakes tend to flake off something ;) [22:45] haha [22:45] Azelphur: Is it possible a rubber seal or the like has failed and that's what those bits are? [22:46] penguin42: I think it could be, MacG32 on that forum is relatively convinced that it's leeched plasticizer from my tubing, and has photos from my exact tubing to back it up [22:47] oh, hadn't thought of plasticizer [22:48] seems like a bad batch of pipe perhaps [22:51] or maybe normal behaviour for cheap piping? [22:54] penguin42: could be [22:54] penguin42: I'll buy some decent XPSC piping this time round [22:56] * penguin42 wonders if the stuff they use for medical liquid pumps works [22:56] probably not designed to work in as warm an environment as a computer [22:57] I dunno. most the machines we built basically were computers [22:57] AAHH!!! [22:58] I have been having endless issues with my box, web server keeps dropping off the radar [22:58] I think it is related to 'wpa_supplicant[553]: last message repeated 61 times' [22:58] dogmatic69: well, what was the actual message? [22:58] that seems to show in the logs right around the time server goes down [22:58] StevenR: that is it, 'wpa_supplicant[553]: last message repeated 61 times' [22:58] hold... blade wpa_supplicant[553]: Failed to initiate AP scan. [22:59] its late :P [22:59] dogmatic69: can you pastebin a bit more of your logs please? [22:59] StevenR: k, give it a minute to log some more [23:01] StevenR: http://bin.cakephp.org/view/696643950 [23:02] tail -f /var/log/* ^ [23:03] it has not dropped now for 5 minutes... longest in about a week :/ [23:04] dogmatic69: need to see /var/log/syslog in much more detail please. [23:08] StevenR: http://paste.ubuntu.com/1056699/ [23:09] probably unrelated, but I'm curious why the wpa lines don't include a hostname? I thought syslogd added those [23:09] shauno: i previously hacked the wifi in, most likely did not do it properly [23:09] now using cat5 [23:10] would be good to turn it off but not bothered / dont know how [23:18] dogmatic69: if you no longer need wifi, then you could do sudo /etc/init.d/wpa_supplicant stop and then use update-rc.d to remove it from the startup process [23:18] dogmatic69: or if you're using network manager to manage your network config, just disable it in that. [23:19] sudo: /etc/init.d/wpa_supplicant: command not found [23:19] no network manager, its ubuntu server [23:21] dogmatic69: can you disable the wireless interface in /etc/network/interfaces then? [23:24] I have commented everything in there out [23:24] well, everything to do with wifi [23:25] it is now http://bin.cakephp.org/view/1624004382 [23:25] the rest of the config was ssid etc, but commented out now [23:25] is there somethign to reset? [23:26] dogmatic69: you'll need to restart networkin (or reboot) to apply it [23:27] $ sudo service networking restart [23:27] restart: Unknown instance: [23:34] dogmatic69: try sudo /etc/init.d/networking restart ? [23:35] hmm [23:35] * Running /etc/init.d/networking restart is deprecated because it may not enable again some interfaces [23:35] it killed it, now cant access the box [23:36] * dogmatic69 goes to kill the power [23:38] dogmatic69: it'll come back in a moment. [23:39] its already rebooted