**** BEGIN LOGGING AT Sun Mar 19 03:00:02 2017 Mar 19 06:56:52 https://2.taobao.com/item.htm?spm=2007.1000337.16.4.hx4eZE&id=540823535194 Mar 19 06:57:11 does it look real from photos? Mar 19 06:57:31 page requires login Mar 19 06:57:38 throw a link to the photo Mar 19 06:58:52 https://img.alicdn.com/bao/uploaded/i4/6000000002394/TB2u0CYXY_0UKFjy1XaXXbKfXXa_!!0-fleamarket.jpg_728x728.jpg https://img.alicdn.com/bao/uploaded/i4/6000000006759/TB2cPfVc4RDOuFjSZFzXXcIipXa_!!0-fleamarket.jpg_728x728.jpg Mar 19 06:59:48 i would bet on a fake Mar 19 07:00:20 also https://img.alicdn.com/bao/uploaded/i1/6000000000268/TB23TXrjmFmpuFjSZFrXXayOXXa_!!0-fleamarket.jpg_728x728.jpg Mar 19 07:02:05 https://nagomartin.wordpress.com/2010/07/09/be-careful-if-you-want-to-buy-nokia-n900/ Mar 19 07:02:26 few notes to spot most obvious differences Mar 19 07:04:17 hi Mar 19 07:05:25 and: https://www.youtube.com/watch?v=t27QMN5AH2c Mar 19 07:05:31 which confirms it's a fake Mar 19 07:06:21 thank you Mar 19 07:06:40 never supposed they could actually fake this form-factor Mar 19 07:06:50 you should message the dealer and/or taobao to inform about it Mar 19 07:07:29 there was discussion there already if it's a fake or not, but seller tried to prove otherwise Mar 19 07:07:42 post a link to yt vid Mar 19 07:08:02 that should nail him down Mar 19 07:09:01 otoh, it's funny people are still trying to sell fake n900 after it's EOL in ~2011 ;) Mar 19 08:04:44 nekit: remember that n900 was sold for quite a sum at the time of release Mar 19 08:05:59 KotCzarny: I wouldn't call it funny, rather sad Mar 19 08:06:07 those fakes seem to be based on MTK chips Mar 19 08:06:09 it's a sad laugh Mar 19 08:06:29 there is no replacement for 6 years already Mar 19 08:06:38 who needs a replacement? Mar 19 08:06:49 it's working perfectly fine for me Mar 19 08:06:53 how's your browser? Mar 19 08:07:03 and web browsing on any screen <10" is a pain Mar 19 08:07:04 etc. Mar 19 08:07:13 no, it wasn't back then Mar 19 08:07:17 other than browser it suits me perfectly Mar 19 08:07:42 fmg: but sad state of the web today is that everything has to have 2cm blank space around it + tons of js Mar 19 08:08:18 all the multimedia and communication functionality n900 has in 2010 are gone, with some exclusions Mar 19 08:08:26 irc is there Mar 19 08:08:31 yeah Mar 19 08:08:51 and other fluffs are just that, fluffs, that go with the wind after a year or two Mar 19 08:08:57 I am sure you got what I mean Mar 19 08:09:01 and i think xmpp is also supported Mar 19 08:09:21 it is all about the choice, now we don;t Mar 19 08:09:23 yup, it's not a feature phone with flashy interface for regular-joe Mar 19 08:09:31 but it was, back then Mar 19 08:09:52 was it? i got an idea that it was hacker's phone since the beginning Mar 19 08:10:15 besides that, it was fully-functional multimedia and communication device Mar 19 08:10:21 it still is Mar 19 08:10:37 mhm Mar 19 08:10:42 oscp runs perfectly fine, whereas many android phones lack such features Mar 19 08:11:53 so i can play lots of file formats in one app with flexible interface Mar 19 08:12:19 (audio of course) Mar 19 08:16:10 someone recently said that regular users are quite dumb computer-wise Mar 19 08:16:28 which also explains why simple devices/software are selling Mar 19 09:17:05 how to avoid DSME rebooting system on Devuan due to MMC failed to mount? Mar 19 09:18:18 recompile dmse to just oops instead of rebooting? Mar 19 09:18:23 *dsme Mar 19 09:18:57 test on the device with mmc reader ? Mar 19 09:19:10 fake it some other way? Mar 19 09:20:22 I asked because there should be a workaround already if freemangordon has it running on his VM, or I did something wrong Mar 19 14:55:17 NeKit: I doubt it is because of no mmc Mar 19 14:57:36 most probably it is because of a missing /var/lib/dsme/saved_state Mar 19 14:58:21 where should it come from? Mar 19 14:58:32 for now, create it by hane Mar 19 14:58:36 *hand Mar 19 14:59:28 it should contain "USER" Mar 19 15:00:25 it comes from /sbin/preinit script, which is missing in devuan Mar 19 15:04:13 NeKit: if doing "echo USER > /var/lib/dsme/saved_state" doesn't help, please provide syslog Mar 19 15:04:27 it helped, thank you Mar 19 15:04:53 cool Mar 19 15:05:02 we should fix that at some point Mar 19 15:05:18 Pali: what is the sane way ^^^ ? Mar 19 15:05:44 run getbootstate Mar 19 15:05:55 that program generates saved_state Mar 19 15:06:06 sure, but what it will do on x86? Mar 19 15:06:12 and should work with upstream kernel too (if you install it from cssu) Mar 19 15:06:25 Pali: x86-64 devuan here :) Mar 19 15:06:26 it is for n900, not idea then Mar 19 15:06:34 create USER with echo Mar 19 15:06:59 Pali: the point was - can we get similar stuff on x86? acpi call or something? Mar 19 15:07:27 no, there is nothing like that in x86 world, or at least not standard Mar 19 15:07:32 ok Mar 19 15:07:40 there are some proprietary extensions for notebook manufactors Mar 19 15:07:46 but nothing documented... Mar 19 15:08:02 maybe just default to USER in init script if there is no such file? Mar 19 15:09:53 mhm Mar 19 15:09:58 make sense Mar 19 15:15:26 NeKit: so, you have it booting to hildon-desktop now? Mar 19 15:15:48 not yet, still missing deps for it Mar 19 15:15:57 did you build osso-af-utils? it's required by osso-af-init, but not in fremantle-gtk2 Mar 19 15:16:13 yes, it should be in maemo repos Mar 19 15:16:19 * freemangordon checks Mar 19 15:17:07 it's there, yes Mar 19 15:17:14 community-ssu on github Mar 19 15:17:23 yes, but I guess I should put it in the repo Mar 19 15:17:29 is it there? Mar 19 15:17:32 hmm Mar 19 15:20:46 NeKit: it doesn;t build, I am going to fix it Mar 19 15:26:26 so it was not needed before? Mar 19 15:29:37 it was, I have fixed it locally Mar 19 15:29:43 but didn;t put it on github Mar 19 15:29:47 will do it now Mar 19 16:09:54 NeKit: it is ready Mar 19 16:10:24 NeKit: you will also need to pull maemo-system-services as well, I fixed dh_installxsession too Mar 19 16:12:57 NeKit: please report (or open issue on github) if there are more missing/broken/etc packages Mar 19 16:13:29 hmm, why are not issues enabled by default? Mar 19 16:15:02 NeKit: do you want me to add you as a member of fremantle-gtk2 org? Mar 19 16:19:55 Wizzup: the same question to you, I guess you're going to be the release manager, so you need write access to the repos to update debian/changelog files Mar 19 16:21:56 that will be useful, but I don't need access right now Mar 19 16:22:34 Wizzup: no maemo timeslice? Mar 19 16:23:03 I meant that I'll need it when amprolla is set up properly Mar 19 16:23:28 Wizzup: parazyd said he is working on it, so soon or later it will be ready Mar 19 16:23:41 I think it is better to have the packages ready by that time Mar 19 16:23:53 agreed. ok, yes please, then :) Mar 19 16:24:29 Wizzup: most, if not all of the packages need their debian/changelog fixed for devuan (and versions as well) Mar 19 16:24:41 what is you github nick? Mar 19 16:24:51 wizzup? Mar 19 16:25:05 no Mar 19 16:25:10 'MerlijnWajer' Mar 19 16:28:12 Wizzup: you should have an invitation Mar 19 16:30:33 freemangordon, thanks, but no need for now, I think Mar 19 16:31:27 NeKit: ok. but still, please report if you find issues so I can fix them Mar 19 16:35:50 should upstart-dev be replaced by maemo-system-services-dev? Mar 19 16:35:59 yes Mar 19 16:36:34 NeKit: you should install upstream upstart Mar 19 16:36:49 and maemo-system-services, maemo-system-services-dev Mar 19 17:17:03 how to go around osso-icons-default conflict with hicolor-icon-theme? Mar 19 17:22:01 hmm, I can;t remember of a such conflict Mar 19 17:23:11 NeKit: what is the conflict? Mar 19 17:24:14 dpkg: error processing archive osso-icons-default_2.5.0+0m5_all.deb (--install): Mar 19 17:24:14 trying to overwrite '/usr/share/icons/hicolor/index.theme', which is also in package hicolor-icon-theme 0.13-1 Mar 19 17:25:31 removing it will also going to remove gtk-3.0 and clutter Mar 19 17:25:40 you don;t need gtk3.0 Mar 19 17:26:02 hmm, wait Mar 19 17:27:05 now I remember I had to repackage it Mar 19 17:28:45 NeKit: try with http://46.249.74.23/devuan/osso-icons-default_2.5.0+0m5_all.deb Mar 19 17:29:12 can't remember why I had to repackage it, maybe because I had to remove some dependencies Mar 19 17:31:47 NeKit: there is something weird, dpkg -L hicolor-icon-theme | grep index gives empty result here Mar 19 17:34:11 that's interesting Mar 19 17:34:12 hicolor-icon-theme/stable,now 0.13-1 all [installed,automatic] Mar 19 17:34:16 is it same version? Mar 19 17:35:39 yes Mar 19 17:36:05 ii hicolor-icon-theme 0.13-1 Mar 19 17:37:39 https://packages.debian.org/jessie/all/hicolor-icon-theme/filelist Mar 19 17:38:06 no idea :) Mar 19 17:39:06 http://pastebin.com/nCXFrQmv Mar 19 17:41:39 NeKit: http://pastebin.com/acW8aW4R Mar 19 17:42:31 yes, the same Mar 19 17:42:46 is there a possibility that hicolor-icon-theme was also repackaged though? Mar 19 17:43:05 no Mar 19 17:43:30 lemme reinstall it Mar 19 17:44:01 hmm Mar 19 17:44:03 trying to overwrite '/usr/share/icons/hicolor/index.theme', which is also in package osso-icons-default 2.5.0+0m5 Mar 19 17:44:16 maybe I used --force-all to install osso icons Mar 19 17:44:49 can't really remember Mar 19 17:45:18 makes sense then, it installed with that flag Mar 19 17:45:34 do you still have the source you used to repackage icons package? Mar 19 17:45:44 I repackaged from .deb Mar 19 17:45:55 don't have a source package Mar 19 17:47:44 we need to find another packge with free icons Mar 19 17:48:02 gnome like or ? Mar 19 17:48:06 the license for osso-icons-default does not allow redistribution Mar 19 17:48:11 KotCzarny: maemo-like :) Mar 19 17:48:27 and gpled most likely? Mar 19 17:48:48 KotCzarny: yeah Mar 19 17:48:51 we need a graphics designer Mar 19 17:49:03 but I think there are already such packages in maemo repos Mar 19 17:49:36 KotCzarny: like https://talk.maemo.org/showthread.php?t=65801 Mar 19 17:50:12 i think i saw that icon pack somewhere Mar 19 17:50:45 google: clarity icon theme Mar 19 17:51:14 (and yeah, i've snatched them into oscp too ;) Mar 19 17:51:42 hmm, what? https://apps.ubuntu.com/cat/applications/lucid/hildon-icons/ Mar 19 17:52:10 hehe Mar 19 17:52:11 http://archive.debian.org/debian//pool/main/h/hildon-icons/ Mar 19 17:52:37 heeh Mar 19 17:53:00 maybe in sdk there is another icon theme package? Mar 19 17:54:09 hmm, how to remove a package while installing a replacement? Mar 19 17:54:20 --ignore-deps ? Mar 19 17:54:22 or something Mar 19 17:55:13 --forece-all did the job Mar 19 17:55:18 *--force-all Mar 19 17:55:26 yeah, i'm always forgetting the names Mar 19 17:59:25 i love how m$ 'antimalware service executable' brings the system to a grind, irony runs deep here Mar 19 18:14:47 And sadly it's the lightest one out there Mar 19 18:23:49 ShadowJK: hehe, but i think eset is quite decent and much better than m$ offering Mar 19 18:23:56 not free though Mar 19 18:44:48 * DocScrutinizer05 wonders how long till nausea kicks in, reading https://www.xda-developers.com/diving-into-sdcardfs-how-googles-fuse-replacement-will-reduce-io-overhead/ Mar 19 19:28:59 At what part does it kick in? Mar 19 19:31:26 At some point I lose track of why they want/need vfat/fat32 on internal emmc-type storage Mar 19 20:02:44 17% more slowly Mar 19 20:02:46 buahahahaha Mar 19 20:02:52 shittalk Mar 19 20:03:46 oh, hey, it will take 117s instead of 100s, its horrible Mar 19 20:19:57 and it all could be solved by some symlinking :P Mar 19 21:00:00 why the fuck would one want to emulate fat32... Mar 19 21:00:05 * merlin1991 headdesks Mar 19 21:09:58 640kB is enough for everyone! Mar 19 21:18:34 ShadowJK: (at what part) I simply can't make up my mind Mar 19 21:19:08 prolly depends on how many BS the particular reader can take before it's too much Mar 19 21:19:51 starting at ancient legacy, to most recent decisions, and the complete lack of common sense in all that Mar 19 21:22:23 with introduction of MTP the latest, they lost any reason to keep that VFAT crap around, and thus could have reverted to sane ACLs and whatnot, all proven age old unix stuff to solve age old "problems" like access permissions Mar 19 21:22:38 or DUH ... SElinux? Mar 19 21:22:44 apparmor? Mar 19 21:23:16 mount namespaces? Mar 19 21:28:25 A big part of why fat/fat32 continues to exist is that its the best filesystem when a device wants to expose its data via USB Mass Storage (i.e. MyDocs on N900) Mar 19 21:28:29 to start with: it's the responsibility of $APP to keep any security critical stuff confidential against other processes under same user/permission Mar 19 21:28:53 jonwil: see what I said (and they wrote) about MTP Mar 19 21:30:59 So does MTP have seek now? Mar 19 21:33:17 Mm, Wikipedia lists "GetPartialObject64" and "SendPartialObject" under the Android extension of it. Mar 19 21:34:14 oh, there's some MTP 1.1 that has at least "GetPartialObject" Mar 19 21:34:26 freemangordon: got the invite, will accept it when I get home tomorrow eve, I'll resume the package building and note taking, maybe automate the building with a simple script until parazyd finishes amprolla Mar 19 21:34:44 this week is sightly busy due to the tor dev meeting in amsterdam Mar 19 21:43:18 > An N900 with pali's U-Boot installed. Mar 19 21:43:22 how do i get this? Mar 19 21:43:54 i guess it refers to https://github.com/pali/u-boot-maemo/ ? Mar 19 21:44:15 https://talk.maemo.org/showthread.php?t=81613 Mar 19 21:44:48 oh hi pali :D Mar 19 21:45:35 hmm ok, so i should first get this n900 over to CSSU ? Mar 19 21:46:26 u-boot is independent of cssu Mar 19 21:48:49 hm ok Mar 19 22:09:08 blocked by some third party policy setting that i cant find Mar 19 22:09:09 dangit Mar 19 22:24:00 hm? Mar 19 22:24:48 np Mar 19 22:25:01 seems some earlier extra-devel install wrekked dependancies Mar 19 22:25:08 blocking u-boot to get installed Mar 19 22:27:45 !BEFORE INSTALLING THIS U-BOOT ALL OTHER INCOMPATIBLE U-BOOT VERSIONS MUST BE UNINSTALLED! Mar 19 22:27:50 ;-D Mar 19 22:28:32 Hildon Application Manager will not allow you to install this U-Boot if there are some problems. Mar 19 22:32:59 i just wiped the n900 using lazyway Mar 19 22:33:04 w/e :P Mar 19 22:33:17 almost have extras-devel back and will retry Mar 19 22:35:23 same result with a clean n900 Mar 19 22:35:25 :( Mar 19 22:41:30 right, apt-get install u-boot-flasher does something, lets see Mar 19 22:43:39 alright, this is getting somewhere :D Mar 19 22:52:17 merlin1991: iirc qemu has a fat32 emulator ;p Mar 19 22:55:02 Wizzup: on your devuan-on-n900 tries, did you run devuan from internal flash? or SD Mar 19 23:01:53 buZz: sd card Mar 19 23:02:00 k k Mar 19 23:02:09 i'm following https://github.com/dderby/debian900 atm Mar 19 23:02:37 patazyd has a ddable image Mar 19 23:02:43 parazyd Mar 19 23:02:49 going to sleep. Mar 19 23:02:49 dont mind, i have time Mar 19 23:02:53 sleep well Wizzup \o Mar 19 23:10:18 I didn't really understand where that article was talking about the FAT32 emulation happening. Mar 19 23:11:17 I imagine it would be sensible to just have FAT32 emulation for the UMS (so you'd point g_mass_storage to a dynamic file in the FUSE mount) Mar 19 23:11:45 but it sounded like every operation had to do with this "VFAT emulator" Mar 19 23:14:12 all you need is a FAT32 emulator for UMS .. handling permissions for different applications shouldn't really need a special filesystem. Mar 19 23:15:29 if you're restricting that at a kernel level, you'd probably just have each program have its own mount namespace where the diruectories it's meant to see are bind mounted into the expected place in a tmpfs. Mar 19 23:18:58 tbh on the former thing, I'd quite like an implementation of that that doesn't depend on any Android stuff ... so I could just do `virtfat normal-directory /tmp/fat-normal-directory` Mar 19 23:19:25 then make g_file_storage point to that /tmp/fat-normal-directory file. Mar 19 23:20:55 Usually when transferring things to normal people's computers using my N900, I'll just have various FAT images sitting in files on my SD card, and I'll point g_file_storage at them when needed. Mar 19 23:21:19 Would be nicer to export a normal directory as a FAT image instead. Mar 19 23:22:40 I've thought about looking into qemu's FAT virtualisation thing before, that enyc alluded to. Mar 19 23:28:38 Hm, there's a project on GitHub under bwimm/libvirtual_fat that says it does the read-only version of that. Mar 19 23:30:12 ~uboot Mar 19 23:30:13 N900 uBoot is a siamese twin binary [uBoot+stockMaemoKernel] that resides in kernel NAND partition /dev/mtd3 aka "kernel". You can't uninstall it, rather you'll nuke it when you flash/install another kernel like stock maemo kernel or powerkernel. To start other than stock maemo kernel via uBoot, you have to provide the according kernel image files, or http://talk.maemo.org/showthread.php?t=81613 Mar 19 23:33:19 it's really tricky :-) Mar 19 23:35:01 new kernel flasher pkgs have some nifty check to see if the installed (in NAND) primary kernel is ok to get replaced **** ENDING LOGGING AT Mon Mar 20 03:00:00 2017