=== ion_ is now known as ion === ion_ is now known as ion === doko_ is now known as doko === dendrobates is now known as dendro-afk === dendro-afk is now known as dendrobates [06:21] Hello. Since Unity started to be used by default in 11.10, I noticed that my application's menus are showing without their icons when the app uses the global menu. However, I've noticed that by using icons that are loaded from a system icon theme, they show correctly in the unity global application menus. So, in order to work around this, I figure I'll add Icons temporarily to a theme, but I need a few suggestions as to how I can go === FlannelKing is now known as Flannel === ubott2 is now known as ubottu [09:17] pitti: sorry, but I don't think I'll be able to get Firefox up tonight, will try to get it up later today so when you start tomorrow it should be ready (hopefully) [10:49] Do I see this right that casper has no means for autologin unless running a Display Manager? [10:59] hmm, whoever approved nvidia-tegra into the archive, a BIG THANK YOU ! ... but it somehow ended up in universe ... should be multiverse (unsupported binary drivers) [11:06] * ogra would really like to see who accepted a package in the accepted message [11:29] Hello im Niels and im new to ubuntu development [11:36] hi niels_ [11:36] Riddell: Hi! Was the removal of kalgebra-dev in https://launchpad.net/ubuntu/+source/kalgebra/4:4.7.90-0ubuntu2 by accident or is it gone forever? [11:36] hi Id like to join the ubuntu development but im really new at this [11:37] !development | niels_ [11:37] niels_: Interested in becoming an Ubuntu Developer? Get started here: https://wiki.ubuntu.com/UbuntuDevelopment [11:37] ok i have already configured launchpad and stuff [11:45] but how do i join a project [11:45] whats a good project to start on? [11:49] what are you interested in [12:12] Ampelbein: it's mostly now in analitz-dev [12:12] Ampelbein: which I'll upload tomorrow [12:12] Ampelbein: what do you need it for? [12:13] Riddell: I was looking at the NBS list, cantor lists it as a build-dep. [12:14] oh there's a new cantor to go with it [12:14] it's all in the PPA, kalgebra was uploaded a bit early by accident [12:15] I see, thanks! === rsalveti` is now known as rsalveti [13:00] Do you see a way which does not require hacking squashfsroot/etc/init/... to make casper auto-startx with the $USERNAME? [13:15] Casper could need quite some updates. Looks like most of the code dates back to 8.04 and has a lot of "extremely ugly hacks" as the comments put it === EyesIsAsleep is now known as EysIsPerson === EysIsPerson is now known as EyesIsServer [17:59] first time after I started using unity I consider stopping using ubuntu alltogether. it reminded me from one very nasty. [18:00] the problem existed before gui change as well. [18:01] but in different setting. [18:02] Say there comes a need to configure something in ubuntu. so okay you quickly find the setting you have to change because you have search engines and help files and whatnot. [18:02] the setting is layed into a graphical interface.. that's rather okay even if unnecessary [18:04] now then say you have 'Open With' and you'd like to use ~/software/newest-blender for it. [18:04] the actual feature you think should do this will bring you an 'assistance' menu that doesn't contain your blender in it! [18:05] but that's not bad still. [18:05] now comes the most deterring part. [18:05] when you search for solution, you get instructions for using that gui setting and nothing else. [18:06] I even found a forum post: [18:06] http://ubuntuforums.org/showthread.php?s=04e9c7d6b9d25bdf1a8def4d722a6cab&p=11545913#post11545913 [18:06] hey that's a good answer! except that the poster DID knew it already. :D [18:06] and it fucks everyone off. [18:09] you could overall help the user best by pushing the settings into single same place conceptually, yet keeping them physically separate. [18:10] global configuration doesn't always work though. [18:10] I'll think up something. [18:15] about every config in linux is a textfile anyway. but some actually require gui to get them properly configured. :) [18:16] '- [18:17] alright. now I found it. with some more effort. [18:21] Cheery: i don't quite understand what point you were trying to make :) [18:27] dobey: you have guis for changing settings and occassionally there happens that gui doesn't help you at all. [18:27] dobey: but it doesn't end there [18:28] what do you mean by "help you" there? [18:29] normally you'd change setting by typing config lines into a file. [18:29] it's supposed to help you if you make a gui for changing those settings without opening the file. [18:30] i don't think a gui is supposed to necessarily help with that [18:30] dobey: anyway. when you search for setting that gui app which was supposed to do what you wanted harms you because all your searches route to that app. [18:30] and i don't think users snould normally be changing configuration like that [18:31] were you complaining about unity, or configuraiton in general? [18:31] dobey: ubuntu. though this appears in windows, which is why I don't use windows [18:31] it may appear elsewhere as well. [18:32] changing things you shouldn't be changing, may break things, yes [18:32] that's not limited to software :) [18:33] as 'open this file with a specific software I downloaded myself to a local file?' [18:34] you may consider that's not a common thing to do. but you just screw yourself there because that's just plain ignorancy. [18:34] open the application, go to File->Open, find the file, click "Open" [18:34] dobey: and what do you do with the nautilus app you've got? [18:35] dobey: or auto-open in your browser? :) [18:35] i think everyone wants to get rid of file browsers in general [18:35] because you can mutilate your way through doesn't mean things should be broken. [18:35] but all the necessary underlying pieces aren't quite there yet to enable that [18:36] how do you get rid of file browser? [18:36] you can replace it with a better app, yes. [18:36] that'd be preferred actually [18:36] meaningful and useful search and indexing of files [18:36] dobey: I see no reason to get rid of file browsers, although they could become somewhat more intelligent sometimes... ;) [18:37] JanC: because browsing files is pretty much never what you want to do [18:37] dobey: and what do you do when you want to create a directory? [18:37] file browsers are just very tedious and manual search interfaces [18:37] Cheery: you don't, because directories are meaningless [18:38] dobey: if they are, why are they there from the start? [18:38] meaningless to users, not to computers [18:38] directories are more meaningless to computers than to users [18:38] broder: Specifying ip=frommedia on commandline works even better than symlinking true from NetworkManager [18:38] i've never logged into my computer and thought "you know, today i'll make a directory" [18:39] Does anyone know why, after I boot with casper, I got the CD added to my sources.list ? [18:39] JanC: not according to ls -d "~/.*" [18:39] dobey: yeah. but you've probably thought today you're starting something, and create a directory for all work files you need for that thing. [18:40] dobey: directories solve a *very hard* problem in somewhat sensible way. [18:40] Cheery: only because there is no better way to do it currently, because IDEs pretty much all suck. [18:40] dobey: those directories are there for the users, not for the computer (a computer doesn't care whether those files are in directories) [18:40] directories are easy because they're what everyone has used for the past 20 years [18:40] it doesn't make them the best solution [18:41] JanC: users don't care aabout the stuff in there [18:41] dobey: this far I haven't seen better even if I have searched for one. [18:41] JanC: except for maybe .porn or something :P [18:41] dobey: but they care they are not visible when looking for their own files, which is why they are in hidden files/directories ;) [18:42] so directories were invented so that users could file away documents in logical places [18:42] JanC: not really, no [18:42] dobey: File browsers for stuff in .* blah you're probably right, but for organising there own files I think directories still make some sense [18:42] it might not be perfect, but it was better than what was there before ;) [18:42] directories were invented because there were no GUIs at the time [18:43] penguin42: not really [18:43] dobey: directories persisted for long after there were GUIs [18:43] penguin42: organizing things is work. why should i have to do that work, when i have this super fast computer that can do it all for me? [18:43] dobey: How would you organise the files with a GUI in a way that was different underneath from directories? [18:43] penguin42: you don't. the computer does it. users shouldn't have to do that work [18:43] and when I say I still want a file browser, I certainly want it to be more advanced than what we have now (static categories of files) [18:44] Cheery: yes, because humans are lazy [18:44] BeOS had "dynamic categories" in its original file system ;) [18:44] dobey: hmm but what do you do with all those files you've got - I don't search for data a lot of the time, I know where I put it - it's the same with real paper [18:44] Cheery: and change is hard. cf. unity/gnome3 [18:45] penguin42: i don't have files. i have content, most of which happens to be encapsulated in files, because that's how computers have worked for 60 years [18:45] (or "dynamic directories" if you want) [18:45] dobey: oh bend over with that. you said you'd want to index all your files and use search every time to get on your files. [18:45] dobey: data is stored in "files" in the real world too ;) [18:45] dobey: That's OK, but with your content would you want to organise that content in some way? [18:45] penguin42: don't assume that because i said search and indexing is the answer, that users are the ones doing the searching. [18:46] dobey: or lets even forget the concept of 'file'. [18:46] dobey: now. what do you got in there? [18:46] dobey: But I don't think in all cases the users or even the computer has to do search - I have an organisation and I know where I put some stuff (admittedly not everything) [18:46] Cheery: in where? [18:47] dobey: say if you had thing you're fantasizing about, how would it perform? [18:47] penguin42: why do you care *where* it is on the hard disk? it is irrelevant. all you care to do is use the data. [18:47] Cheery: pretty much the same way my phone does. [18:48] dobey: Agreed, but I have a lot of data - I have the note about what I was doing with my bills, I have some notes about web pages I'm going to visit [18:48] dobey: well you care about the location for backup purposes. [18:48] Cheery: That's avoidable [18:48] dobey: also, you care that all the data is in consistent place so you can use it. [18:48] Cheery: not really; the backup software cares about location. all it has to do is back it up, and restore it, if need be [18:48] i don't care where on disk it is [18:48] dobey: I mean I have to separate some of my information into categories about different things I'm doing; I'm not sure I'd be able to give it information that a computer would be able to arrange any better [18:49] penguin42: you mean you want to tag your data? [18:49] dobey: for example sometimes this fails because person sends a document file but not images he has linked into document. [18:49] Cheery: it doesn't fail. broken applications are still broken [18:49] as said before: the original BeOS filesystem worked that way, ages ago... ;) [18:50] a document should be self-contained [18:50] dobey: Yes, and I'm happy with directories effectively being places where all the things with the same tag are listed; but the interesting questions are what happens as you rearrange and delete things with a tag - it's not always obvious [18:50] dobey: which you can do by wrapping your doc into a directory. [18:50] dobey: I mean 'directories' are after all a card-file analogue [18:50] penguin42: i don't understand your question. if you delete something, it gets deleted. [18:51] dobey: Or does it just remove the tag from some subset of the items that were so tagged? [18:51] Cheery: then it's not a document, it's a zip file full of crap [18:51] dobey: OK, if you just had tags what would you do to arrange a set of information in time order ? [18:52] search for that tag, and sort results by date? [18:52] dobey: except if .zip file also contains an extra file which makes it appear as something else. [18:52] dobey: date of what - when you wrote it? When the stuff in the file talked about? [18:52] Does anyone know why, after I boot with casper, I got the CD added to my sources.list ? [18:52] Cheery: stop making up hacky solutions to very rare problems, as some excuse as to why you need a file browser [18:52] dobey: IMHO what you end up with is a hierarchy of tags - which is OK, but a flat tag cloud is pretty useless [18:53] penguin42: whatever date you wish to sort buy [18:53] by [18:53] dobey: OK, so you've got a set of stuff tagged and with a date, and a friend comes along and gives you a copy of a similar thing that he's worked on - how do you differentiate the two sets ? [18:53] penguin42: directories are hierarchies of tags, but only one tag per N items [18:54] you don't tag it with a date; dates are inherent properties [18:54] dobey: even with your metadata-based document store, you will need a "file browser" (or "document browser", or whatever you want to call it) [18:54] dobey: Well I include the filename as well as the directories I guess [18:54] dobey: But a directory is not one tag - it's the hierarchy of how you got there - dave/photos/2011/december [18:54] JanC: no i won't [18:55] dobey: That's very different from dave/errors/2011/december [18:55] dobey: It's interesting gmail used tags initially instead of folders, but have now made them hierarchical and ended up with something closer to folders [18:55] penguin42: I don't like that sort of sorting directories. [18:56] penguin42: and again, you don't actually care about the directories at that point. you care about the 'tags' as you call them [18:56] Cheery: It's sometimes useful though [18:56] penguin42: I generally do not care about when it's created. :) [18:56] penguin42: and it's stored by the file system already [18:56] dobey: I agree but I'm saying hierarchical linked directories and files are very similar to hierarchical tags anyway [18:56] ManDay: that's done by one of the scripts in casper-bottom - if you're digging into what casper is doing, it's worth skimming all of them [18:57] (and I don't agree directories are a computer thing - we've stored things in books in organised libraries for centuries) [18:57] penguin42: this isn't a similarity contest though [18:57] broder: I tried, haven't found it. [18:57] it's about improving UX, and reducing workload for the user [18:57] I grepped for everything sensible but nothing came up [18:58] dobey: I'm OK with reducing workload for the user and improving the user experience, and I think a tagged approach can help - but I think as soon as you have more than a little bit of information you need something a bit more powerful than a flat tag cloud [18:58] ManDay: ./scripts/casper-bottom/41apt_cdrom [18:58] penguin42: this also isn't about libraries; and they've only been that way, because we're only at the point of replacing them with electronics now. [18:58] penguin42: i don't know why you're assuming some sort of flat tag cloud [18:59] tags are a single piece of data [18:59] dobey: OK, can you explain what you mean by a tag just so we understand [19:00] I would say all forms of metadata is what you want [19:00] a tag is a user-defined piece of metadata they add to something; like tagging some recipes as breakfast, lunch, dinner, brunch, or dessert or something [19:00] JanC: which is why i said indexing, not tagging [19:00] and if you like to add directory=dave/errors/2011/december as metadata, that's fine ;) [19:01] hmm found a way to add .desktop, but did not find how to get the 'open with' work properly [19:01] Cheery: what are you trying to open with, with? it works perfectly fine here [19:02] dobey: I just found that tags work well for small amounts of data/files - like the few you might have on a phone, but when you have thousands of tag matches life gets harder [19:02] but even with all that meta-data, I want to be able to "browse" my files (even if it doesn't involve a hierarchical tree) [19:02] dobey: I have application binary in /home/software/ which I want to use when opening a certain kind of file. [19:03] JanC: why? [19:03] dobey: that binary doesn't show up in 'open with' -menu, and I can't see a button which adds it. [19:03] dobey: what's the alternative? [19:04] how do I find files/documents/data otherwise? [19:04] dobey: I find myself wanting to do things like restrict matches to only things with a particular tag, but not in the confidential set or etc etc and you start having to remember which tags you have [19:04] Cheery: the .desktop needs to have the mime type listed for the file you want to open [19:04] Cheery: and you need to run update-desktop-database on the directory where the .desktop file is, to update the cache [19:05] JanC: WHY do you want to browse files? your music should show up in your music player, documents in editor/viewer, videos in video player, etc… [19:06] dobey: Why do you want to separate the types of media? [19:06] penguin42: making your life complicated with directories or tags is still making your life complicated; but that is a choice you made [19:06] dobey: what if I want all files related to some topic/project, but don't know what the types of files are? [19:06] dobey: No, I believe it's a necessity when you have a lot of data [19:06] penguin42: i disagree [19:07] dobey: And that's why I like directories :-) [19:07] JanC: what is a topic/project? [19:07] dobey: it could be anything [19:07] that's the point exactly [19:07] dobey: No, but tags are good - I just don't think they can be used to remove directories completely, only to partially arrange stuff [19:08] JanC: then how do you know it is a topic/project if you don't know what a topic/project is? [19:08] dobey: I could give an example, but not exactly a definition (outside of "about everything") [19:09] say i want to find everything I have about Ubuntu [19:09] search -> "Ubuntu" [19:09] now it suddenly just figured it out :o [19:09] oh well [19:09] dobey: search where? [19:09] JanC: in the search application [19:10] enough! I'm frustrated enough it with. >:/ [19:10] now going to do the actual stuff I was planning. [19:10] so what's the difference between a "file search application" and a "file browser"? ;) [19:10] dobey: Hopefully that search application will have remembered all the things tagged with Ubuntu and organised them into a cache that will let it find all the things labelled with ubuntu quickly - lets call it a directory [19:10] it's not a file search application; it's a search application [19:11] penguin42: it's called an index [19:11] dobey: I think you'll find the thesaurus has those as the same thing [19:11] JanC: idk. but it's sort of silly thing to think different if your filesystem is directory based. [19:11] I suppose that all depends on your definition of "file" [19:11] directories don't work here, without having N copies of the file, for every tag you want it to have [19:12] or N directories for all tags [19:12] dobey: No, we've got hard links - they do work [19:12] which are wastes of disk space [19:12] (or at least links of some type) [19:12] penguin42: and it's a complete hack solution [19:12] and a waste of disk space [19:12] dobey: As opposed to the waste of CPU time in searching for stuff you've previously organised? [19:12] and it works, which the alternatives don't (yet) ;) [19:12] JanC: local content are what people generally mean by 'file' [19:13] alternatives work fine [19:13] they just only exist on phones/tablets [19:13] penguin42: you clearly don't understand how optimizations work i guess :) [19:13] dobey: they don't work for the simple reason that most data lacks the required metadata [19:13] JanC: they work perfectly fine [19:14] there is no required metadata [19:14] i have never needed/wanted a file browser on my phone [19:15] broder: Can I delete the concents of /run in the squashfs? [19:15] dobey: neither have I, but then again, I use my phone as a phone... ;) [19:15] ManDay: yes, it will be shadowed by a tmpfs very early in the boot [19:15] thanks [19:15] dobey: Well, not on my phone, but I do access google docs from my phone - albeit only with 2 or 3 files [19:16] [19:16] Any other directories which I may purge for the liveCD image? [19:16] (besides the usual tmp ones) [19:16] ManDay: probably, but i don't konw what they are off the top of my head [19:17] ok, i guess /run is good neough [20:25] Hello, I am trying to set up a PPA, and I do not know if I have correctly published my key to the keyserver, is there some way to determine this? [20:28] launchpad won't accept my key fingerprint, and I think I published directly to the ubuntu keyserver about 20 minutes ago. [20:31] tehuser_u: Go to http://keyserver.ubuntu.com/ and search for the user id to see if it is published correctly. [20:31] ok. Cheers. [20:36] ok, no it hadnt. thanks. [21:29] Thanks broder [23:10] slangasek: I syncd libverto which is a new build-dep of krb5, but it'll need a push through NEW and an MIR before krb5 can build unless you want to temporarily switch to the in source version