**** BEGIN LOGGING AT Fri Jan 04 02:59:58 2013 Jan 04 06:02:35 Morning.g Jan 04 06:03:05 -g Jan 04 10:30:27 DocScrutinizer05, extras builder still not working Jan 04 10:30:44 see fremantle build queue: https://garage.maemo.org/extras-assistant/index.php Jan 04 12:05:52 Re. Jan 04 12:57:44 Pali: Bergie just informed me that they are about to transfer autobuilder to a VM to keep it running, >>albeit probably very slow<< Jan 04 12:58:30 DocScrutinizer05, ok, do you know when it will work again? Jan 04 12:58:46 trying to get detailed info Jan 04 13:00:53 ok Jan 04 13:00:56 Niels / X-Fade doing that moving to VM Jan 04 13:02:54 main issue for now is DNS, since Nokia not exactly responsive to any inquiries regarding that. Probably they simply don't answer Nemein's requests until the lawyers finished the paperwork with HiFo board Jan 04 13:08:27 ooh, and Nokia gave clearance for migration budget ~2 work weeks before new year, only then Nemein could start to do anything about it, so delay is all with Nokia, not Nemein Jan 04 13:10:03 almost nobody inside Nokia left over who would even know what's maemo doesn't exactly help either Jan 04 13:45:20 DocScrutinizer05: thanks for the info, I was really wondering what's going on with autobuilder Jan 04 13:46:10 Niles had to disable/stop it so the old server and the VM won't get out of sync Jan 04 13:46:21 we still have no control over DNS Jan 04 14:39:09 hi everyone, can anyone help me in setting up ubuntu lucid desktop VM for fremantle in windows Jan 04 15:11:42 hrmpf, it's the second time that sourav asked that question Jan 04 15:12:01 but really, how hard can it be to download virtualbox and adding a new VM with an existing drive? Jan 04 15:12:05 *and to add Jan 04 15:26:32 apparently it can be quite hard Jan 04 15:26:44 :D Jan 04 15:31:05 I can understand asking once, but doing it repeatedly over few days instead of just googling about it or figuring it out by himself... Jan 04 15:31:40 he esked me too, just an hour ago :D:D:D Jan 04 15:31:44 *asked Jan 04 15:32:29 he msg'd me too, though I never quite got what he was even asking Jan 04 15:57:27 thedead1440: hmm, what is that "However, Rob Baeur (SD69) has now informed me that Harmattan might not be involved in the Board's discussions with Nokia." supposed to mean? Jan 04 15:57:59 freemangordon: Let's clarify that the attached list is the one you are referring to. Jan 04 15:58:01 It has files which are on merlin1991's server and are on maemo.org. So those files should be deleted I believe before the list is forwarded to Nokia. Jan 04 15:58:02 I think we are unlikely to get Harmattan files now but you can try I guess. Jan 04 15:58:04 Rob Jan 04 15:59:00 but what will happen with Harmattan then?!? Jan 04 15:59:07 the fuck Jan 04 16:05:32 sd69 honestly needs help with the technical implications of the migration Jan 04 16:07:07 it's a tad unfortunate that probably neither side (board, nokia's lawyers) has full expertise about what maemo means and includes, on a technical level Jan 04 16:07:37 neither for HARM Jan 04 16:08:02 actually for HARM even I am basically clueless Jan 04 16:08:55 council got qwazix though, who seems pretty active and up to date with harmattan details Jan 04 16:10:24 ShadowJK: I've found some more info about the troubling card: Allocation units are as big as 12M, Write Size is 64K and Page Size is 16K. This seems unusually high to me for an SDHC-card. Could this have any influence on the IO errors? Jan 04 16:14:23 anyway... Jan 04 16:15:12 [general notice] autobuilder considered down according to read-only notice on top of any maemo.org page Jan 04 16:17:45 please any of you following tmo closely, would you point your humble councilor [me] to any threads/posts relevant to council that I should notice (particularly those related to migration/backup). Thanks. Jan 04 16:22:38 wirr: yes, it could. Biggest AU I've seen so far is 8M. A 32g sandisk I recentlyish purchased is 4M Jan 04 16:23:07 the max timeout is approx 2s iirc, read-mod-write of 12M is going to take 2 sec at class 6 speeds... Jan 04 16:23:20 or slightly more :) Jan 04 16:24:07 ShadowJK, I see... so this would mean ideally by optimizing alignment and fs options speeds _could_ be increased to stay below timeout? Jan 04 16:24:57 it'd mitigate it, I imagine Jan 04 16:25:41 do you know of any other configurable or hardcoded options which could be tweaked? Jan 04 16:26:39 is there a way to completely disable, or retrigger, the timeout? Jan 04 16:29:43 also AIUI an allocation unit is a chunk of flash space that's supposed to get written in one operation? could we just ignore the 12M AU the card requests and go for smaller size? Jan 04 16:40:19 AU is totally hidden from host Jan 04 16:40:47 (except on newer cards they do provide hint as to what optimal would be) Jan 04 16:41:37 AU, as i understand it, is smallest physically erasable block. Jan 04 16:42:23 write size 64k sounds like smallest writable block, where the destination must be erased, is 64k Jan 04 16:42:23 yep Jan 04 16:43:04 where destination must be already erased Jan 04 16:44:07 ShadowJK, I agree with that. do you know how i can configure the write size on linux? Jan 04 16:45:16 Well we don't have a filesystem that is flash aware Jan 04 16:45:30 jffs rules Jan 04 16:45:48 but maybe tinkering with the raid stripe/stride options in ext* and the ext* block size? Jan 04 16:48:11 jffs sucks in that it keeps all filesystem metadata in ram Jan 04 16:48:15 I've already adjusted stripe/stride to page size Jan 04 16:48:40 but write size is 4 times page size... so this not possible without flash-aware fs? Jan 04 16:49:04 and has to scan entire device on mount Jan 04 16:49:04 nilfs2 might be more sensible, and some version of it seems to be included in kp Jan 04 16:49:50 can stripe/stride be adjusted to 64k? Jan 04 16:51:23 ShadowJK, possibly, haven't tried yet Jan 04 16:51:49 I vaguely remember playing with this when I set up my adata card for my N800 Jan 04 16:52:22 but I forget the outcome :) Jan 04 16:57:16 wirr: does this card have a UHS logo on it too? Jan 04 16:57:33 first test (using mass-storage! and kp52) gives approx 4.7MB/s Jan 04 16:57:39 ShadowJK, yep Jan 04 16:57:58 way better than with default params Jan 04 16:59:56 what params did you change? Jan 04 17:00:17 Using nr-requests tweak too? Jan 04 17:00:52 * ShadowJK is at work and not supposed to irc, so I might reply with great latency Jan 04 17:04:32 would be very interesting to be able to read SD specs and see what they've changed over time Jan 04 17:04:35 ShadowJK, i've set nr_requests to 8192 using swappolube adjusted the partition layout to start with the first partition at 4megs (fat-optimized area from 4-12meg) and aligned all following partitions to 12meg boundaries (starting at 4, according to flashbench mailing list) Jan 04 17:05:02 adjusted fat & ext4 filesystems to page size Jan 04 17:07:30 ShadowJK, i'm wondering if we should change default scheduler from cfq to noop (noop is ideal for ssd, but deadline, which is not available, would be better for sd, i think) Jan 04 17:13:11 On N800 I only have noop, and keep wishing for cfq ;) Jan 04 17:13:25 ShadowJK, is that so? Jan 04 17:16:50 Well when I was researching this in maemo4 era last time (maemo5/n900 was such a huge improvement out of the box that I havent bothered much in that area), google searches found lots of eeepc users experimenting with this. Most users who tried noop came back saying "it's awesome" right after booting, then 15 minutes later started complaining about random apps or entire system mysterious stalling for tens of seconds at a time.. then they went back to cfq, Jan 04 17:16:56 which they deemed was "consistently fast, uh, consistently slow, but consistent" Jan 04 17:17:56 I did test bfq some time ago, seemed to be better that cfq but I didn't benchmark it so might have been placebo Jan 04 17:18:04 than Jan 04 17:19:12 okay Jan 04 17:20:44 The fundamental issue is still that they're all flash-unaware.. some operations on flash will be essentially "free" while others will have a huge cost Jan 04 17:21:31 and all the schedulers more or less assume a harddrive, and cost is more or less a linear measure of how far away from last accessed block a potential disk access is Jan 04 17:22:07 ShadowJK, but I think there's some development going on in that area, i think it's called 'fios' Jan 04 17:24:06 DocScrutinizer05 / SpeedEvil ; do you know if sd cards have internal oscillators or if they rely on host provided CLK? UHS cards running at really low speeds in SDHC hosts could be due to that too.. Jan 04 17:25:03 iirc bfq has at least some ssd optimizations too Jan 04 17:25:59 SSDs, flash, and HD are all 3 distinct Jan 04 17:26:15 ssd optimization might hurt flash more than HD optimization Jan 04 17:30:26 ShadowJK: sorry, no info here Jan 04 17:31:25 Afaict, the omap hsmmc hw's data timeout is specified in clock cycles, if we could drop hsmmc bus speed, the timeout would become longer.. but if card runs off of hsmmc clock, it'd run correspondingly slower too and it wouldn't change the situation :/ Jan 04 17:31:51 * ShadowJK wonders if mmc clockspeed is related to cpu speed too, on N8x0 it was Jan 04 17:32:08 on n900 too Jan 04 17:32:34 that's why cpu is locked 2 min 500 with USB connected Jan 04 17:32:56 2 min 500? Jan 04 17:33:14 @ min 500 MHz, sorry, typo Jan 04 17:33:41 ic, now I get it Jan 04 17:34:33 freemangordon; can you tell what speeds mmc runs at with cpu @ 250, 500, 550, 600, 600+? Jan 04 17:34:50 ShadowJK: though not to VDD1 OPP but to VDD2 OPP Jan 04 17:34:56 ShadowJK: ^^^ Jan 04 17:35:49 hold on, so setting the cpu to 500 or 600 would speed up the I/O? neat Jan 04 17:35:51 can't remember what were VDD2 freqs Jan 04 17:36:57 kerio: no. you should tweak VDD2 frequency, CPU @ 500 or 600 does not make much of a difference Jan 04 17:38:29 ShadowJK: can't remember at what speed mmc is clocked, but it is on the top limit Jan 04 17:39:26 freemangordon, pali: would one of you be willing to compile an mmc module for power52 with timeout set to 7 or 13? Jan 04 17:40:05 ShadowJK: from memory, external clock Jan 04 17:40:33 wirr: iirc it is set to 13 or 14 right now Jan 04 17:40:52 Yeah it's 14 in powerkernel Jan 04 17:41:00 freemangordon, 14... Jan 04 17:41:13 and that's the highes allowed value afaik Jan 04 17:41:29 freemangordon, yeah but strangely the card works fine on stock pr1.3 Jan 04 17:41:46 I see Jan 04 17:41:57 * ShadowJK wonders why the number 15 is reserved Jan 04 17:42:10 are you sure you tested it long enough? o stock that is. Jan 04 17:42:21 full 32gig read and write Jan 04 17:42:23 sequential Jan 04 17:44:40 ry hit with bonnie++ (in repos, devel probably or tools) Jan 04 20:31:32 ShadowJK: I've run tests with bonnie++: Despite the geometry, fs and scheduler tweaks this resulted in i/o errors and a complete crash of the device. next i've replaced the omap_hsmmc.ko with the one from cssu-thumb. and voilą now no more io errors and acceptable performance... Jan 04 20:32:16 Pali: perhaps you could compile new omap_hsmmc modules for power52 with timeout values of 7 and 13? i'd be happy to test. Jan 04 20:33:12 if someone's interested in the bonnie output just tell me Jan 04 20:42:16 I forget, is cssu-thumb kernel different from powerkernel? Jan 04 20:43:11 wirr: did bonnie fail on stock? Jan 04 20:43:22 haven't tried yet Jan 04 20:43:35 and yes, cssu-thumb and power52 modules differ Jan 04 20:44:45 what do you mean "7 and 13"? Jan 04 20:55:46 ShadowJK: if I remember right, cssu-thumb kernel is power-kernel (but not non-release kp52) Jan 04 20:57:22 so it probably has same hsmmc code as kp51 Jan 04 20:57:55 ShadowJK, timeout values Jan 04 20:59:40 ShadowJK: the best man to ask is freemangordon Jan 04 20:59:54 ShadowJK: or take a look into sources ;) Jan 04 21:00:01 wirr: btw where did you find out about the 12M thing? Jan 04 21:00:25 ShadowJK, flashbench, wait a sec Jan 04 21:00:48 ShadowJK, https://wiki.linaro.org/WorkingGroups/Kernel/Projects/FlashCardSurvey?action=show&redirect=WorkingGroups%2FKernelConsolidation%2FProjects%2FFlashCardSurvey Jan 04 21:01:25 Mine is the Sandisk Ultra SDSDU-032G-A11 "SU32G" (2012) Jan 04 21:06:18 can someone point me to the sources of cssu-thumb and power52? Jan 04 21:16:44 * ShadowJK vaguely recalls there used to be a benchmark utility on that flashcardsurveypage :/ Jan 04 21:36:59 ShadowJK, git.linaro.org Git - people/arnd/flashbench.git/summary git.linaro.org/gitweb?p=people/arnd/flashbench.git;a=summary Jan 04 21:37:17 did you run it? Jan 04 21:37:32 (it's probably destructive) Jan 04 21:39:31 yep Jan 04 21:39:45 (only through mass storage) Jan 04 21:40:56 https://github.com/coffeemug/rebench Jan 04 21:40:59 get this too Jan 04 21:41:08 I'll give you params Jan 04 21:45:38 Or here's the script I normally use: http://enivax.net/jk/n900/flash/bench.sh Jan 04 21:46:06 Obviously edit the device name so you dont overwrite harddrives, and ediit path to executable Jan 04 21:46:39 Basically goes through random writes with 4k ... 12M block size.. Jan 04 21:52:12 Warning: as it's a write test it will destroy existing data Jan 04 22:45:03 Night. Jan 04 23:40:05 Re. Jan 04 23:41:39 re Jan 04 23:43:25 thx Jan 04 23:46:04 uh? Jan 04 23:48:03 :) Jan 04 23:49:10 Looks like I'll have to build myself a gaming PC next year-ish :) Jan 04 23:49:14 Elite:dangerous getting kickstarter funded Jan 04 23:52:55 ;-) Jan 04 23:59:12 just take nvidia over amd Jan 04 23:59:20 then you don't run into as many driver problems :D Jan 05 00:01:30 you guys checkout the nexus4? Jan 05 00:01:53 I'm curious about your guys' secondary uses for the n900 now that it's not a phone for you Jan 05 00:02:41 nexus4 would be cool with a hwkbd :/ Jan 05 00:03:00 at 550E, I'd rather wait and see if jolla comes with something interesting Jan 05 00:03:24 meanwhile I'm still using N900 Jan 05 00:09:06 ShadowJK: yeah, I've the same issue w/ no HW keyboards... It'll be my first touch keyboard Jan 05 00:09:11 I gotta at least try it Jan 05 00:09:38 ShadowJK: any word of jolla? Jan 05 00:09:48 It would be nice if they gave us some specific hope Jan 05 00:09:55 so we could help them popularize and everything Jan 05 00:10:15 there's also the new ubuntu stuff Jan 05 00:10:29 they'll be using regular android hardware for the forseeable future, supposedly Jan 05 00:15:27 Its a pitty the few phones out there in the market that have hardware keyboards suck in other ways Jan 05 00:16:10 Yeah Jan 05 00:16:33 Motorola has produced a number of GSM Android phones with hardware keyboards Jan 05 00:16:46 but they are all locked down tight and you cant run alternate OSs on them Jan 05 00:17:40 Sony too, but they only put real kbd on their shit-end androids with 320x200 screens etc Jan 05 00:18:23 The HTC 7 pro would be nice if it wasnt for the crappy OS Jan 05 00:21:48 I suspect it will be a very long time before any manufacturer produces a phone with a physical keyboard that is anywhere near as good as the N900 Jan 05 00:22:39 I'm suspecting the very same Jan 05 00:30:29 https://www.google.gr/search?um=1&hl=en&safe=off&tbo=d&biw=1366&bih=628&tbm=isch&sa=1&q=android+qwerty+smartphone&btnG= Jan 05 00:31:10 Looking through those images it seems the companies think that the hwkb is only for geeks without a sense of beauty or sth Jan 05 00:31:44 except the MotoDroids that is... Jan 05 00:33:48 99.999% of those phones are obsolete or running on obsolete crappy hardware Jan 05 00:34:01 Yeah Jan 05 00:34:01 ya Jan 05 00:34:10 or have ugly dpads or stupid soft-buttons on the front Jan 05 00:34:24 95% were obsolete when their spec was decided Jan 05 00:34:40 or 2x2px screens Jan 05 00:37:12 I think that whoever does a high end beautiful qwerty phone today with a 4+ screen will carve market share for years to come Jan 05 00:37:43 Because I know nobody, geek or not, who used a good hwkb and wants to go back Jan 05 00:38:50 But even I wouldn't buy anything available to buy today with a hwkb. So companies fool themselves that there is no market Jan 05 00:39:00 qwazix: you mean who doesn't want to go back, right? Jan 05 00:39:25 I don't think ubuntu will consider making their own hardware, but they should Jan 05 00:39:29 because their crappy qwerty *of course* don't sell Jan 05 00:39:57 pewter_tao, I'm not good at those double negative things as in different languages mean different things Jan 05 00:40:07 http://www.computerworld.com/s/article/9235176/Update_Canonical_prepares_Ubuntu_for_smartphones Jan 05 00:40:34 but I mean that everybody I know that's been using hwkb wants to keep using hwkb Jan 05 00:41:37 ah, ok now I get why I was not clear. I meant "I know nobody, geek or not, who used a good hwkb and wants to go back to pure touchscreen" Jan 05 00:41:51 which probably doesn't make much sense... Jan 05 00:43:02 I think if you took a phone like the Galaxy Nexus, Galaxy S2 or Galaxy S3 and added a good hardware keyboard, it could well become the greatest phone ever made. Jan 05 00:43:41 qwazix: now that made sense Jan 05 00:44:12 Galaxy Note ftw. Fits a 104key qwerty :P Jan 05 00:44:15 jonwil: the bigger the screen, the more oddly a HW keyboard will make the bezel shape Jan 05 00:44:23 imho, n900 with a gig of ram and quad cpu Jan 05 00:44:36 thats a perfect phone :-) Jan 05 00:44:39 XenonXZ: yeah, why isn't anyway soldering new CPUs to the n900 board? Jan 05 00:45:23 I feel like we only need 20 people who want this and we can spec one together, get some biz dev people, and make a fucking company Jan 05 00:45:43 I'll make a "let's start a cell phone company website" and you guys can popularize it on maemo forum... Jan 05 00:46:45 :p Jan 05 00:52:12 ShadowJK: elite-what? bell and braben? Jan 05 00:52:27 yes Jan 05 00:52:36 \o/ Jan 05 00:52:44 Braben is doing a Elite 4 :) Jan 05 00:53:10 Braben, the last real coder Jan 05 00:53:12 ;-) Jan 05 00:53:37 or was it Bell? Jan 05 00:54:15 Bell is the harecore mathematician Jan 05 00:54:26 I even forgot how many million assembler lines elite been Jan 05 00:54:54 People who reverse engineered 2 and 3 found many bugs in "simple" things like the 64 bit multiply function Jan 05 00:55:08 Well the original fit in 22k ram Jan 05 00:56:17 damn new elite, forcing me to buy new hardware Jan 05 00:56:24 I recon the best hope for a decent phone with a hw keyboard is if the GTA04 people ever get to the point that they can afford to do one... Jan 05 00:57:19 The hardware of the N900 is starting to look a bit dated (and if it was possible to upgrade the parts in the N900 with new ones someone would have found out how to do it by now) Jan 05 00:58:44 Still love my N900 even if it is dated Jan 05 00:58:48 yes me too Jan 05 00:59:02 The only way I will give up the N900 is if it breaks and cant be fixed anymore Jan 05 00:59:45 DocScrutinizer05; elite 2 Atari version approximately 600,000 lines of asm Jan 05 01:02:29 http://www.kickstarter.com/projects/1461411552/elite-dangerous anyway Jan 05 01:03:12 ShadowJK, thanks four your links - I wont be able to run the tests today but i'll be back tomorrow Jan 05 01:05:05 wirr: sure :) Jan 05 01:06:42 jonwil, uhh, that's if the GTA04 devs figure out power management Jan 05 01:07:07 I had a Freerunner, and it simply pales in comparison to what the N900 can do Jan 05 01:07:57 nn all Jan 05 01:08:11 night Jan 05 01:11:03 What makes power management so hard? I thought it was all done by the kernel on a modern Linux setup Jan 05 01:11:37 Although I guess the real issue (battery life) is all to do with optimizing the whole system to run better Jan 05 01:13:44 Have you use hibernate/suspend-to-disk and/or suspend-to-ram on a linux machine? Jan 05 01:14:05 nope I haven't Jan 05 01:14:39 Ah well, N900 does suspend-to-ram seamlessly, even when screen is on Jan 05 01:15:17 Getting linux kernel to run on a cpu is much easier than getting it to shut down the cpu and wake it up again reliably Jan 05 01:15:47 jonwil: last I heard the wifi module sucked an unholy amount of power for piss range. Jan 05 01:16:00 Right-e-o Jan 05 01:17:05 Without the DVFS, and the sleep modes, N900 idle batterylife would be 5 hours instead of the actual 10-11 days (radios off, screen off, etc) Jan 05 01:17:29 ShadowJK, and while in use, 30 minutes. Jan 05 01:17:47 (this can be demonstrated by booting Debian with the Nemo adaptation kernel) Jan 05 01:17:55 lol Jan 05 01:18:07 30m is impossible :) Jan 05 01:18:54 IIRC Stskeeps said that even with some original RX-51 team members working on MeeGo for the N900, they still couldn't make MeeGo run on the same power envelope as Fremantle Jan 05 01:19:06 so yes, embedded power management is hard. Jan 05 01:21:12 I am guessing that even if you have the same hardware its not just a case of "hey, xyz has already written power management code for our SoC, lets pinch theirs" Jan 05 01:22:07 I suspect omap3 is the only cpu with proper-ish power management code Jan 05 01:23:11 The Jan 05 01:24:12 The original OLPC had similar abilities, they had to completely ditch the bios and rewrite from scratch to accomplish it :) Jan 05 01:24:44 and yeah if the MeeGo guys couldn't match Fremantle for power management when they were using identical hardware and much the same kernel and userspace bits (dsme, mce, bme etc) I dont think the GTA04 guys have a chance Jan 05 01:26:12 Be interesting to see some hard numbers comparing the power management/battery life profile of an N900 with stock Fremantle and a N900 with Pali's BME replacement work, see how the new stuff affects battery life Jan 05 01:27:26 Not significantly Jan 05 01:27:46 I've been running my own for years Jan 05 01:28:29 perhaps only difference is that bme shuts down earlier to avoid suddenly running out of power and the potential filesystem corruption from that Jan 05 01:29:43 I do wonder which bits of the system are the ones causing MeeGo-on-N900 to have a worse power envelope than Fremantle. Or maybe its just that userspace programs (GUI, apps etc) aren't as well written for MeeGo as they were for Fremantle and therefore use more juice... Jan 05 01:30:20 I remember DocScrutinizer bitching about the sensor framework atleast Jan 05 01:30:36 indeed Jan 05 01:31:21 qtm and sensorfw all based on polling rather than IRQ event driven Jan 05 01:31:39 What bit of phone handles sensor framework on Fremantle? Jan 05 01:31:50 let's take an example of fremantle and idle N900: 11 days life. fremantle otherwise idle, but following script running: "while true; do sleep 5 ; done": 4 days life. Jan 05 01:32:14 a sane design decision for all sorts of devices with different sensors that not always have IRQ, but all the devices have lots of battery power or even PSU Jan 05 01:32:50 So if you've got 4 random things polling accelerometer, proximity, camera door, etc, even with half-minute poll interval you've just dropped idle life to four times less Jan 05 01:33:14 I wish there was a Compass Jan 05 01:33:30 My G1 had one, but I threw it away in a psychotic Episode. Jan 05 01:34:19 the compass in my android tablet is about as useful and accurate as a Ouija board Jan 05 01:34:27 Ok Jan 05 01:35:07 jonwil: (which bit of sw) depends, some is done by mce, some by liblocation, some by libisi, etc pp Jan 05 01:35:59 and quite some of them are also not as strictly event driven as they should be Jan 05 01:37:14 e.g. accelerometer, it's way better than the one used in mer, but still miles from optimum Jan 05 01:37:48 remember the Ubuntu Phone OS I talked like 2 days ago? Jan 05 01:37:57 it just doesn't matter that much since they probably poll accelerometer only when screen active Jan 05 01:37:59 here's a demo: http://youtu.be/kXWnMTm7We8 Jan 05 01:38:06 it should work on any hardware. Jan 05 01:38:25 I think it really looks gorgeous Jan 05 01:38:41 we should work in porting that to the N900! Jan 05 01:39:34 "any hardware" is kinda... Jan 05 01:39:40 not sure if I should post *yawn* now, or get me a coffee prior to clicking that link Jan 05 01:40:03 in practice it's like N900 and Google Nexus*, right Jan 05 01:40:09 it's just 2 minutes Doc Jan 05 01:40:09 and N900 is underpowered Jan 05 01:40:26 this is 100% Ubuntu no java / dalvik (Android) crap inside Jan 05 01:40:40 just like Maemo, if it runs Maemo, why not Ubuntu phone? Jan 05 01:41:44 the only major thing hardware-wise is that they've made ubuntu run on android kernel instead of needing linux kernel Jan 05 01:41:45 well, because I hate buntkuh, maybe? Jan 05 01:42:05 you hate Ubuntu? Jan 05 01:42:10 yep Jan 05 01:42:15 wow that's weird, aren't you a Debian guy? Jan 05 01:42:28 or are you a Fedora/Red Hat (RPM) guy? Jan 05 01:42:44 so what? canonical hijacked linux and made it a me-too-windows Jan 05 01:42:48 ~buntkuh Jan 05 01:42:49 hmm... buntkuh is http://xkcd.com/424/ ... Redmond, or http://de.wikipedia.org/wiki/Holstein-Rind Jan 05 01:43:21 android has a "force everything to sleep" thing going on for power saving, so I imagine android kernels aren't that complete and well tweaked for always-on OS :) Jan 05 01:44:14 who said it kept the Android kernel... Jan 05 01:44:24 that's a different project called Ubuntu On Android Jan 05 01:44:37 this is Ubuntu Phone, a complete OS Jan 05 01:44:52 Ubuntu Phone <> Ubuntu for Android Jan 05 01:45:46 my brother has a Samsung Note 10.1 tablet and installed that Ubuntu for Androind and lags like HELL Jan 05 01:45:48 to me, it's unusable. Jan 05 01:46:00 and it's a quad-core tablet with 2 GB of RAM Jan 05 01:47:14 The whole point of their "any hw" claim is that they run on existing android kernel of the "any hw", that way they don't have to write device drivers and fight with device vendors Jan 05 01:47:24 Sc0rpius: I'm a system guy Jan 05 01:48:29 ShadowJK, well yeah so far it only works with Nexus because the whole OS is OpenSource and they can grab drivers from there Jan 05 01:48:36 but that's not the point in the future Jan 05 01:48:40 Sc0rpius: ...and i'm kinda oldfashioned, so I hate to find upstart on a friend's buntkuh box when he asks for help, while *all* by boxes have proper sysV-init, and I don't see *any* advantage of that new crap Jan 05 01:48:52 you're a Slackware guy. Jan 05 01:48:58 the I hate gnome Jan 05 01:49:02 then* Jan 05 01:49:22 that's normal in experienced Linux users. Most people install Xubuntu (no Unity, no Gnome) with Xfce Jan 05 01:49:32 Sc0rpius; I'm pretty sure the graphics drivers are blobs even on nexus? :) Jan 05 01:49:37 and generally I hate that windowsy philosophy in ubuntu Jan 05 01:50:08 so you like your several terminal open and nothing else hehehe Jan 05 01:50:53 several terminals and a browser Jan 05 01:51:04 while I've always been proud of my linux systems being *ahead* of any windows desktop usability wise, ubuntu tries to mimic windows as much as they can Jan 05 01:51:21 GUI is great. Jan 05 01:51:32 I install plain ubuntu but never ever use the default desktop Jan 05 01:51:33 DocScrutinizer05, uuhhh? Jan 05 01:51:41 wizards are da shite though Jan 05 01:51:42 Unity is not a mimicry of Windows. Jan 05 01:51:53 Ubuntu GUI (with Xfce or LXDE) is great Jan 05 01:52:04 Unity is very bad but there are several options Jan 05 01:52:48 canonical, red hat, all the same crap, different package (manager) Jan 05 01:52:54 I use mostly: notion lxdm uzbl mupdf wicd Jan 05 01:53:08 then it doesn't matter if the ubuntu machine has 32MB of ram Jan 05 01:53:21 i'd not even be astonished if ubuntu came with double-click to launch, by default Jan 05 01:53:57 I use LXDE in my netbook, it's extremely light Jan 05 01:54:03 but Xfce in this desktop Jan 05 01:54:07 LXDE rules, especially OpenBox Jan 05 01:54:10 not light and usable enough imo :p Jan 05 01:54:20 actually this is a Xubuntu running in VMWare full screen in a 2nd monitor, my host is Windows 8. Jan 05 01:54:30 I have to have both worlds at the same time. Jan 05 01:54:46 DocScrutinizer05; seems like they have voice to launch now Jan 05 01:54:53 (say app name to start it) Jan 05 01:54:54 \o/ Jan 05 01:54:57 that's nice Jan 05 01:55:10 until you launch porn Jan 05 01:55:15 ;-) Jan 05 01:55:19 if it's like iphone or android's voice recog, it's about as useful as a random number generator Jan 05 01:55:37 I tried "FNORD" with Android. Worked grate! Jan 05 01:55:46 I literally screamed Jan 05 01:55:49 Like a Madman Jan 05 01:55:55 ;-) Jan 05 01:57:39 ShadowJK: I bet in 6 months they have tiles, like windows8 Jan 05 01:58:04 I hope so! Jan 05 01:58:17 * DocScrutinizer05 too Jan 05 01:58:24 might finally kill them Jan 05 01:58:37 you already can take the labels out in the taskbar and leave only the icons just like W7 Jan 05 01:58:51 and you can, already, group Windows in the taskbar as well Jan 05 01:59:00 at least in my Xfce that I'm using right now Jan 05 01:59:21 I can't tile windows though Jan 05 02:25:47 Night. **** ENDING LOGGING AT Sat Jan 05 02:59:59 2013