[00:22] so I had the bizarre idea of having mouse support in grub2 [00:36] Yes, that is bizarre. [00:37] Well, it's not so bizarre, but it's (a) not *that* useful, and (b) hard. [00:42] mainly, cause part of me thinks Ubuntu seriously needs to look at what Windows 8 is doing with recovery options [00:46] Recovery is not best done in the bootloader. [00:47] ie: boot to a real environment, *then* do nice recovery stuff. [00:47] Hmm [00:47] I have been thinking that Windows 8's recovery stuff isn't in the bootloader, since you do get the mouse and stuff like normal [00:48] For example: you can't write to anything from grub. [00:48] and I meant, recovery through Grub as in, a sort of menu that could pass certain "hidden" recovery modes to the bootup [00:49] like: Single-user mode, command-line only mode, VESA graphics? [00:49] We currently do that. [00:49] If your last boot wasn't successful we default to the recovery mode, whuch has all that stuff. [00:50] I don't think adding a mouse makes that better in any significant way - the options are so simple that there's no benifit to having a mouse pointer, other than that it might look slightly friendlier. [00:51] Yeah. The Windows 8 recovery tools are loaded in a GUI (and by hitting F8 on the bootsplash) [00:54] however, my logic would be to use grub2's more advanced theming system to make a menu like that, even if it has to be still keyboard driven [00:57] I've not seen the recovery tools you're referring to, but again: does it have to be in the bootloader? Can't you just select "recovery mode" and do all this with an actual kernel underneath? [00:58] RAOF, http://www.neovelop.com/Media/Neovelop_En/Windows-Live-Writer/Windows-8-boot-menu_8A0F/WP_000224_2.jpg [00:58] http://www.neovelop.com/Media/Neovelop_En/Windows-Live-Writer/Windows-8-boot-menu_8A0F/WP_000230_2.jpg and [01:01] * RAOF wonders if that's run from EFI; that'd be more sane. [01:20] RAOF, its present on all Windows 8 installs, EFI or not [01:21] I seem to remember something about Win 8 requiring EFI :) [01:22] actually, it'll be able to use EFI for fancier stuff (like a faster bootup) [01:22] they've already confirmed 8 has the EXACT same minimum requirements as 7 [01:29] RAOF, http://i.imgur.com/oF4ki.png and before this, I was thinking a screen with a pictogram like the installer CD (but with like, Enter key -> Gear) [01:33] Although, I do have a wayward way this layout could be possible (in a way) without many changes to grub itself [01:34] That seems like a reasonable goal, although I'd ideally hide ?Test RAM? somewhere :) [01:35] RAOF, actually, although its seperated from the rest of them, even the Windows boot menu contains a link to its own memory diagnostic [01:36] my wayward way involves GRUB 2's support for bitmap fonts [01:36] Yeah, I just wouldn't present it on the main menu. [01:37] I was just going purely from how grub usually operates [01:37] Nor, really, would I present ?recovery mode? on the main menu; I'd save the main menu for top-level OS decisions. [01:37] Isn't windows 8 going to be 64-bit only? [01:37] though ... if you wanna go beyond that... [01:37] TheMuso, no. Windows SERVER is 64-bit only [01:37] Well, you can do submenus. [01:37] We currently use submenus in grub. [01:37] Lirodon: Ah ok, thats good to hear. [01:38] TheMuso, they wouldn't just randomly drop 80% of their userbase in one release [01:38] its Mac OS Lion that's 64-bit only [01:38] True, but I thought I heard rumblings along those lines... [01:38] I don't think ARM is 64-bit [01:38] Yeah I know about OS X> [01:38] No its not [01:38] But thats not x86. [01:38] So it doesn't count. [01:38] At least IMO [01:39] And windows 8 going x86-64 only wouldn't drop anywhere near 80% of their userbase. [01:39] Almost all systems bought in the last 5 years support x86-64. [01:39] (And lots before that, too) [01:40] I was thinking instead for the bottom, "E - edit boot parameters C - GRUB command line (advanced) F2 - Recovery mode" [01:41] though, I'm starting to think, maybe boot parameters/command line should be hidden or require CTRL+C/CTRL+E too, just in case [02:03] RAOF, TheMuso http://i.imgur.com/EH2AH.png did some adjustments [05:31] good mornng [05:40] didrocks, hello [07:08] good morning === chrisccoulson_ is now known as chrisccoulson === API is now known as Guest11151 === yofel_ is now known as yofel === Guest11151 is now known as apinheiro [11:01] jaytaoko: your gles branch doesn't build btw, needs this: http://paste.ubuntu.com/693692/plain/ [11:01] otherwise it looks good [11:15] where can I find the dx-team list archives? === MacSlow is now known as MacSlow|lunch [12:29] bug 682788 :( [12:29] Launchpad bug 682788 in unity "Global menu is not ergonomical on large screens" [Undecided,Confirmed] https://launchpad.net/bugs/682788 [12:38] Kaleo: in the spare time I could work on bug 849732 [12:38] Launchpad bug 849732 in unity-2d "Alt + doesn't work" [Critical,Triaged] https://launchpad.net/bugs/849732 [12:46] Amaranth: ping [12:46] jaytaoko: pong [12:46] Amaranth: Hello! [12:46] jaytaoko: howdy [12:47] Amaranth: if I am correct, you don't need the DXT1, DXT2 ... texture format with GLES. right? [12:47] Amaranth: I can ifdef the code that uses it [12:48] Amaranth: so it won't bother you on GLES [12:48] jaytaoko: was lots of code to ifdef [12:48] jaytaoko: since we never hit those codepaths it was easier to just leave the defines in the enum [12:49] Amaranth: ok, I will leave them. The DXT are just a texture format. You can actually use them no matter the platform... [12:50] Amaranth: have you seen the change I did for the headers of the shaders. [12:51] mardy: if you have spare time :) I did not think you did [12:51] Amaranth: I added the macros VertexShaderHeader and FragmentShaderHeader [12:52] Amaranth: depending on the version of GL, these macros expand to strings that contains the right stuff for the shaders [12:52] ah [12:53] jaytaoko: I just made sure it built and the demos ran :) [12:54] Amaranth: Are you running on Intel GLES or ARM? [12:54] jaytaoko: intel gles, can't get an oneiric install running on arm [12:54] Amaranth: ok === m_conley_away is now known as m_conley [13:44] There have been a fair number of unity related ideas posted on brainstorm in last few months and at this point we have been redirecting them to the bug tracker [13:44] it would be great if someone from the unity team could act as a "liaison" between unity and brainstorm. [13:45] even just briefly going through the ideas tagged with "unity" once a month is all that would be necessary. [13:46] users have been posting quite a few good suggestions, and having to tell each one of them to re-post these suggestions a bug tracker is kind of annoying to these users [13:47] if anyone would like to work as this liaison, please let me (or any of the moderators on the brainstorm site) know as we would LOVE to hear from you! :D [14:36] hell'o [14:38] hell o nmarques === htorque_ is now known as htorque [16:33] smspillaz, do you have any clue about bug 627195 I have made it a very clean and clear bug report with a test case and some other finding, all in the description. [16:33] Launchpad bug 627195 in compiz (Ubuntu) "Apps raised from indicators sometimes dont have the focus" [High,Confirmed] https://launchpad.net/bugs/627195 [16:35] om26er: lets see [16:35] oh that [16:36] smspillaz, you already aware of that? [16:36] om26er: I talked with ted about this but there's really no way to do it other than defining a new standard for focus prevention [16:36] err protocol [16:36] I'll comment on the bug report about it [16:36] smspillaz, so we should have hopes for the LTS ;-) [16:37] om26er: no as in [16:37] om26er: it needs to be ratified by freedesktop [16:38] I guess we could implement experimental support for it, but its not high on my list of priorities [16:38] smspillaz, its annoying and everyone faces it though :/ [16:39] om26er: I know [16:39] freedesktop thing might make it longer time thing ? [16:39] well like, [16:40] my argument to ted is that the indicators should make themselves identify as pagers in focus requests [16:40] but he doesn't want to do that [16:41] my other argument was that the indicator should ask the application for the window it wants to activate it and then activate it itself (since it was the last window with the most recent _NET_USER_TIME) [16:41] but he wants it so that the indicator tells the application to activate itself which fails [16:41] his solution to the problem is to make it so that we replace _NET_USER_TIME with something totally different (like a sha1 hash) that can be passed around applications [16:41] which is a perfectly fine solution [16:42] except that it wouldn't work across all window managers [16:42] tedg: ^ did you ever write to wm-spec-list about this? maybe we should draft up a spec for how this would work [16:44] om26er: basically, I agree with his solution, but its just tricky because you are introducing a new protocol to talk to window managers and that means that you need to go through fd.o to get it ratified [16:45] smspillaz, No I haven't. I didn't think I had you convinced :-) [16:45] smspillaz, understood and thanks for the information (the little that got into my non-proogrammer mind) [16:45] heh :) [16:46] tedg: remind me later this week to draft up a spec for this [16:46] I bet it wouldn't be too hard to hack something together to make it work the way you wanted to [16:47] smspillaz, Let's do it at UDS -- perhaps a session? [16:47] tedg: be aware though that fd.o may well reject it on the grounds that you can acheive the same effect by having the indicator ask for the window id to activate and then activating that window itself as a pager [16:47] smspillaz, It doesn't have priority over Oneiric things. [16:47] tedg: we're pretty much finished O for compiz [16:48] * tedg files a few more bugs ;-) [16:48] tedg: I won't be at UDS but I will be there for the pre-sprint [16:48] tedg: no as in [16:48] tedg: it doesn't matter how many bugs there are [16:48] smspillaz, Ah, that's right. We can do it then. [16:55] tedg, hi, you probably had no time (and motivation ;)) to look at bug 835646 again? do you think that's release critical as it also affects unity 2d users, which likely use systems with lower resources, thus hitting swapping or worse earlier? [16:55] Launchpad bug 835646 in unity (Ubuntu) "unity-panel-service memory leak when starting various applications" [High,Confirmed] https://launchpad.net/bugs/835646 === zyga is now known as zyga-afk [16:56] oh, he's gone :( [16:56] htorque: really no motivation then :) [17:00] didrocks: probably, heh === zyga-afk is now known as zyga === macer1 is now known as Admc === Admc is now known as macer1 [18:52] whats unity 2D built on? === macer1 is now known as testing123111 === testing123111 is now known as macer1 [19:36] FloatingGoat, Qt [19:36] FloatingGoat, qt, because of that it looks 100 times better. [19:36] interesting, that's what i though [19:36] than 3d version [19:37] macer1: i know im jsut now trying the current build from the ppa [19:37] and the dash looks 100x [19:37] better [19:37] what is 3d? [19:37] well i thought 11.04 3d was qt too? === m_conley is now known as m_conley_away