**** BEGIN LOGGING AT Mon Feb 08 02:59:58 2016 Feb 08 07:13:46 am i the only one who hates google's login screen, where username is entered, then you have to click Next in order to get password field Feb 08 07:14:00 any ideas how to restore the 'old' way? Feb 08 07:31:41 Sicelo: I do not hate that design but the Web site of my credit card company has been the same in that way for years. Feb 08 07:54:39 :) Feb 08 07:54:45 you're used to it Feb 08 07:58:56 3.5-inch HDDs seem very large (in dimensions, not storage capacity) now that I am used to 2.5-inch HDDs and SSDs. Feb 08 08:03:31 My Galaxy Note 3 seemed large when I first began using it but now “smaller-than-phablet” computers seem small. Feb 08 08:05:18 * Maxdamantus still has 14 TB of 3.5" HDDs in his computer. Feb 08 08:06:52 3.5" HDDs are probably still fairly large in storage capacity (versus volume). Feb 08 08:07:14 Can get 8 TB consumer drives atm I think. Feb 08 08:07:59 You'd probably have to pay excessive amounts to have that storage density on SSDs. Feb 08 08:08:15 for now, yes Feb 08 08:08:23 (that is, pay far more than you'd usually pay for the same amount of storage on smaller SSDs) Feb 08 08:09:33 sicelo: disable js for that google screen maybe? (didnt try) Feb 08 08:12:19 i wonder. will check Feb 08 08:12:45 Presumably there's something somewhere to say that it's a public computer you're logging onto. Feb 08 08:13:34 Oh, nvm. The boxes are on different screens even when it doesn't know your account. Feb 08 08:15:03 If you set your UA to the one used by an N9 it has both boxes on the same screen. Feb 08 08:15:59 Google should make this thing optional. I don't even see the logic behind it Feb 08 08:16:26 sicelo, maybe it forgets your username cookie Feb 08 09:02:41 Updated root certificates for Maemo Fremantle (and patches/instructions/everything else needed to make it work) should appear in CSSU Git sometime tonight assuming I dont hit any more snags/roadblocks Feb 08 09:05:20 Awesome work! Feb 08 10:14:45 wow! Feb 08 12:01:19 http://talk.maemo.org/search.php?do=process is borked. I mean, a search for "led" doesn't show any hits? Feb 08 12:08:37 I can't find how to patch mce - I posted it myself :-/ Feb 08 12:20:30 google to the rescue :-/ https://talk.maemo.org/showthread.php?p=1388368 Feb 08 12:28:50 freemangordon: are you maintainer of cssu-mce? could you please fix ^^^ + http://mg.pov.lt/maemo-ssu-irclog/%23maemo-ssu.2013-11-23.log.html#t2013-11-23T06:20:48 Feb 08 12:29:10 IroN900:~# md5sum /usr/lib/mce/modules/libfilter-brightness-als.so Feb 08 12:29:11 9db7e3c959bd5b47c7fcfb6ebc1ca764 /usr/lib/mce/modules/libfilter-brightness-als.so Feb 08 12:30:46 once this been fixed by binary patching in original MCE by me, now it's broken again on one of my N-th device migrations and CSSU updates Feb 08 12:33:48 ooh, happy new year Feb 08 12:35:55 Gong Xi Fa Cai Feb 08 13:02:36 Just about finished with the root CA stuff Feb 08 13:08:49 the what now :o Feb 08 13:39:25 kerio: the certificate store Feb 08 13:39:31 nice Feb 08 13:39:31 maemo/certman Feb 08 13:44:07 ok, the work is supposedly finished but for some reason it isn't working right (in that a legitimate https site gives me an "invalid" error in microb) Feb 08 13:45:12 ;< Feb 08 13:45:15 which site? Feb 08 13:45:19 ib.boq.mobi Feb 08 13:45:20 check it with ssllabs Feb 08 13:45:22 my bank Feb 08 13:45:42 it works just fine in microb with the old root ca set (or did last I checked which was a while back) Feb 08 13:45:53 and it also works just fine in my desktop browser Feb 08 13:46:04 https://www.google.com.au does work though Feb 08 13:46:27 ye it works Feb 08 13:46:29 like Feb 08 13:46:33 the site itself works Feb 08 13:47:01 Going to make a t.m.o forum post about all this and see if anyone can shed any light on the problem Feb 08 14:20:41 http://talk.maemo.org/showthread.php?p=1498178#post1498178 Feb 08 14:35:09 Even with the stuff that doesn't work yet, its still a major achievement for the future of Maemo Feb 08 14:40:09 ok, I fixed MCE ALS/indicatorlight stuff again, and there's a script now for you to fix it too :-) https://talk.maemo.org/showthread.php?p=1388368 Feb 08 14:43:54 well, rather a set of shell commands. Needs sed-gnu!!! apt-get install sed-gnu; make sure your path to sed points to /usr/bin/gnu/sed or use full pathname /usr/bin/gnu/sed in place of just "sed" Feb 08 14:57:45 http://wiki.maemo.org/User:Joerg_rw/tools#nasty_hack_to_fix_indicator_light_in_bright_environment Feb 08 15:04:34 DocScrutinizer05: do you ever get the impression the whole world is a nasty_hack ? Feb 08 15:05:09 I often think that most *I* do is Feb 08 15:05:13 lol Feb 08 15:05:51 the above thing works but is terribly wrong on a multitude of aspects Feb 08 15:06:16 i've not seen much wpwrak talking of late Feb 08 15:06:46 it shouldn't depend on sed-gnu, it shouldn't patch a binary to start with, and finally it just works by lucky coincidence that the pattern is unique in this very binary Feb 08 15:07:43 the world is a thin polished veil over a selection of nasty hacks, jury rigs, kludges, duct tape and wd-40 Feb 08 15:07:46 ooh and you generally prolly shouldn't patch binaries with *sed* Feb 08 15:08:11 i was half-expecting to see wpwrak discussing marketing-the-project-progress feedback and/or hackerbus-draft to discuss hey-ho get the re eventually =) Feb 08 15:08:15 inz: lol =) Feb 08 15:08:33 inz: duct-tape is like the force: dark on on side, light on the other, and holds the universe together Feb 08 15:08:52 hehehehe Feb 08 15:13:43 http://wiki.maemo.org/User:Joerg_rw/tools#epilog Feb 08 15:42:55 so, i have this notify script in irssi. And it sends dbus command to vibrate when someone says my nick, but for it to work it requires --print-reply. So im asking, if it possible to hide that dbus reply on irssi? Feb 08 15:44:57 >/dev/null ? Feb 08 15:45:47 u mean to add that after dbus command? Feb 08 15:45:55 i'll try Feb 08 15:46:09 I wonder why it needs --print-reply Feb 08 15:46:50 i don't know, but if i remove that part it doesn't vibrate Feb 08 15:47:01 weird Feb 08 15:47:46 yeah I tried as well Feb 08 15:47:51 and? Feb 08 15:47:57 confirm? Feb 08 15:48:43 ok say my nick Feb 08 15:48:51 Vajb: Feb 08 15:49:09 Vajb: Feb 08 15:49:11 Vajb: Feb 08 15:49:14 thx doc and thx bencoh for suggestion, works Feb 08 15:49:40 DocScrutinizer05: yeah, I confirm it doesn't vibrate without --print-reply Feb 08 15:49:48 freaking bug Feb 08 15:49:55 dunno Feb 08 15:50:22 dbus-send shouldn't return before the dbus message got sent? Feb 08 15:50:53 try a sleep 1 after dbus-send Feb 08 15:50:56 well, im happy as it works without that reply shown now :) Feb 08 15:51:48 DocScrutinizer05: sleep 1? what? Feb 08 15:52:02 I tried from a shell :) Feb 08 15:52:05 Vajb: http://wiki.maemo.org/User:Joerg_rw/xchat/notify.sh indeed also has print-reply Feb 08 15:52:22 bencoh: duh! Feb 08 15:52:45 ok try now. I replaced devnull with sleep 1 Feb 08 15:53:03 wtf how does dbus-send sent msg differ for receiver depending on whether sender prints reply or not? :-o Feb 08 15:53:25 Vajb: it won't work Feb 08 15:53:34 maybe it depends on having "someone" listening to the answer Feb 08 15:53:36 true Feb 08 15:53:51 Vajb: bencoh already tested it in a context that doesn't vanish (shell) Feb 08 15:54:03 it juust blurbed an error of baddlyformed data item Feb 08 15:54:25 Vajb: do you run irssi on device? Feb 08 15:54:36 LOL Feb 08 15:54:37 or is that irssi over ssh + vbell + xterm notification? Feb 08 15:54:49 hahahahaha Feb 08 15:55:03 DocScrutinizer05: ? :) Feb 08 15:55:04 I honestly doubt it would ever vibrate then Feb 08 15:55:23 bencoh: nah, i just like my phone to vibrate in my pocket when im irc'ing on desktop ;) Feb 08 15:55:47 DocScrutinizer05: with vbell+xterm notification on vbell, it would :) Feb 08 15:56:05 that would be beoynd my grasp Feb 08 15:56:07 well Feb 08 15:56:13 I'm just not sure xterm has external command support for that Feb 08 15:56:17 i have just irssi on my device Feb 08 15:56:24 it doesn't Feb 08 15:56:29 afaik Feb 08 15:56:41 I even doubt it supports bell at all Feb 08 15:56:56 doesn't (if we're talking stock XTerminal) Feb 08 15:57:07 :nod: Feb 08 15:57:33 Vajb: now i'm curious regarding your setup .. you mention dbus, irssi, desktop... please clarify what's running where :) Feb 08 15:57:58 I think desktop came in on an inverted basis Feb 08 15:58:20 DocScrutinizer05: talking about real xterm :) Feb 08 15:58:29 desktop was just part of my bad joke :) Feb 08 15:58:37 (I dont use the libvte-based thing) Feb 08 15:58:46 so irssi on device (with perl support)? Feb 08 15:59:02 im running irssi on n900 with simple script to vibrate when mentioned Feb 08 15:59:04 osso-xterm isn't really usable (and libvte is pretty bad / slow) Feb 08 15:59:25 investigation project: why does dbus-send fail for vibra patterns when no --print-reply ? Feb 08 15:59:55 first step: study dbus-monitor Feb 08 15:59:55 hey, maybe jonwil want's it :) Feb 08 16:00:21 Sicelo: yes. Found on forum Feb 08 16:00:37 maybe bencoh was the one who made it? Feb 08 16:00:41 i know that *one* :p i built it ;) Feb 08 16:01:00 ah it was u Feb 08 16:01:01 aaaah wait wait wait, it's a flaw in mce I bet - or rather a feature: any pattern (LED or vibra) terminates when dbus sender disappears from bus Feb 08 16:01:29 well i knew it was either Sicelo or bencoh and odly enough i failed at 50/50 guess.. Feb 08 16:01:44 at least that would be a semi-reasonable explanation, however for LED patterns it evidently doesn't apply Feb 08 16:02:01 Vajb: interestingly enough, i don't run any scripts on it now :p Feb 08 16:02:15 :p Feb 08 16:02:23 mine is almost perfect Feb 08 16:02:47 i just need to figure way to make it not vibrate when window is active Feb 08 16:03:20 and i wanna figure it out by myself thx ;) Feb 08 16:03:53 share afterwards. no idea Feb 08 16:03:57 please paste the dbus-send cmdline here Feb 08 16:05:17 "dbus-send --system --print-reply --dest=com.nokia.mce /com/nokia/mce/request com.nokia.mce.request.req_vibrator_pattern_activate string:PatternChatAndEmail >/dev/null" Feb 08 16:05:27 as is in my script Feb 08 16:05:51 ta Feb 08 16:05:57 in script that i copied somewhere and added just that string* Feb 08 16:11:04 wtf is wrong with dbus(-monitor)? http://paste.opensuse.org/81641486 I don't see the dbus-send message? Feb 08 16:11:23 monitor system? Feb 08 16:11:50 I do, no? Feb 08 16:11:59 indeed Feb 08 16:12:32 does dbus-monitor not see most of the stuff when started as user? Feb 08 16:12:42 dunno Feb 08 16:15:41 no, same under root Feb 08 16:15:49 okay. Lets see if it works say my nick now Feb 08 16:15:57 Vajb: Feb 08 16:16:02 ok Feb 08 16:16:17 now again in few seconds Feb 08 16:17:55 anytime u want :) Feb 08 16:18:18 Vajb: Feb 08 16:18:37 hmm now i broke it Feb 08 16:18:44 dbus is thoroughly fsckdup Feb 08 16:18:53 :D back to drawing board Feb 08 16:19:25 dbus-monitor doesn't show messages sent by dbus-send, mce doesn't vibrate when sender doesn't --print-reply... Feb 08 16:23:11 man dbus-monitor: >>The message bus configuration may keep dbus-monitor from seeing all messages, especially if you run the monitor as a non-root user.<< haha, even when run as root Feb 08 16:24:56 freedesktop.org, what else would you expect Feb 08 16:27:01 FLOS https://pappp.net/?p=969 Feb 08 16:29:18 https://bugzilla.redhat.com/show_bug.cgi?id=477964 Feb 08 16:29:19 04Bug 477964: was not found. Feb 08 16:32:12 redhat == nsa Feb 08 16:32:21 https://bugzilla.redhat.com/show_bug.cgi?id=477964#c11 Feb 08 16:32:23 04Bug 477964: was not found. Feb 08 16:32:38 k made some changes. Could someone say my nick twice with 5second sleep inbetween? Feb 08 16:32:47 vajb 1 Feb 08 16:32:48 2 Feb 08 16:32:49 3 Feb 08 16:32:50 4 Feb 08 16:32:52 5 Feb 08 16:32:54 vajb 5 Feb 08 16:33:20 hmm now it works opposite as meant lol Feb 08 16:33:38 thx KotCzarny :) Feb 08 16:34:05 and >>The problem with using --print-reply is that the call will block waiting for the reply from dbus.<< (comment 17) Feb 08 16:34:57 inz: problem is, there is no world, just vibrating superposition of the void Feb 08 16:35:30 riiiight Feb 08 16:35:48 this reminds me of a problem buried deep inside dbus we faced back in FSO when switching from python to vala. Seems the message buffer gets destroyed in dbus-send *before* the message got copied to the destination buffer, or sth like that Feb 08 16:36:34 this wasn't a dbus-send bug but generic dbus-lib mega bug Feb 08 16:37:39 and would explain a lot of randomly occurring erratic behavior in all dbus systems Feb 08 16:39:30 hmm, how about hacking vbell into vibra in x ? Feb 08 16:41:33 you dont really need to hack X, just hildon-desktop Feb 08 16:42:25 right, wm Feb 08 16:42:29 but it wouldn't work with osso-xterm anyway, since it doesn't handle vbell Feb 08 16:42:44 add vbell support to xterm too Feb 08 16:42:49 so you'd need to either move to another term emulator, or patch it Feb 08 16:43:16 https://bugs.freedesktop.org/show_bug.cgi?id=896 Feb 08 16:43:17 04Bug 896: missing instructions in point 4 of installing scratchbox in gregale/INSTALL.txt Feb 08 16:43:21 I think it was discussed here some time ago (with sicelo maybe) Feb 08 16:43:28 anyway, ctrl-g should either beep on audio or do vibra Feb 08 16:47:02 >>For method calls, we simply wait for the reply, even if we don't print it.<< https://bugs.freedesktop.org/show_bug.cgi?id=896#c26 Feb 08 16:47:06 04Bug 896: missing instructions in point 4 of installing scratchbox in gregale/INSTALL.txt Feb 08 16:47:19 povbot: shut up! Feb 08 16:47:19 DocScrutinizer05: Error: "shut" is not a valid command. Feb 08 16:47:25 :] Feb 08 16:51:39 https://lists.freedesktop.org/archives/dbus/2008-March/009526.html >> At which point the badness happens: The daemon (apparently) immediately closes the connection, oblivious to the fact that there is critical data in it's socket buffer as-yet un-read.<< Feb 08 17:05:35 Vajb: many thanks for bringing up this bug :-) Feb 08 17:06:01 fixing it might improve whole maemo stability a lot Feb 08 17:06:17 oh that's cool :) Feb 08 17:11:19 for a first stopgat (though only for dbus-send) we could overlay the dbus-send binary with a script wrapper that adds "--print-reply" and ">/dev/null/ if there's no "--print-reply" in the options handed to the command Feb 08 17:11:52 DocScrutinizer05: unless maemo stuff has been built around the bug :] Feb 08 17:12:00 I'm pretty sure there are a bazillion other binaries that do a dbus-send() and then exit Feb 08 17:12:25 bencoh: can't be, the bug 'evaluates to" a no-op Feb 08 17:13:21 worst case some stuff slows down a bit Feb 08 17:43:27 a little miscellanea: Feb 08 17:43:30 https://www.cl.cam.ac.uk/~mas90/misc/maemo-5-word-completion/ Feb 08 17:52:35 gotta love scripting. 3 hours and only thing that improved is that irssi drops dbus reply. So basically i got nowhere from where bencoh left off. Feb 08 17:52:53 what is your problem? Feb 08 17:53:06 well i know several ways about not to make it work Feb 08 17:53:10 :) Feb 08 17:53:29 you have to write/think rules something has to do Feb 08 17:53:36 then start implementing it as a script Feb 08 17:53:45 KotCzarny: i wanted to modify my vibrate script so it would only vibrate if my nick is mentioned and im not viewing channel Feb 08 17:54:22 is there a current channel variable? Feb 08 17:54:46 >>It is noteworthy that if word completion has any effect, it is to increase the time taken to enter text<< ;-P I always been aware of that Feb 08 17:55:30 KotCzarny: i thought Irssi::active_window(); would take care of that part Feb 08 17:55:42 did it? Feb 08 17:55:43 or i don't know what u asked :D Feb 08 17:55:50 nah Feb 08 17:56:00 what it returns? Feb 08 17:56:10 with all my addings it just stopt vibrating at all Feb 08 17:56:15 i guess word completion much more useful for vkbd. There it's quite fast (at least for me) Feb 08 17:58:00 vajb: Feb 08 17:58:01 https://raw.githubusercontent.com/irssi/irssi/master/docs/perl.txt Feb 08 17:58:41 closest i got was it did nothing when channel was active and did "can't use string (#chan: nick who called me) as a HASH ref while "strict refs" in use at /path to script and line pointer to line i added" Feb 08 17:59:07 Vajb: if you're "viewing" the channel, with your current proposal you may miss messages. Maemo has only one active window at a time, so if you've got something else on foreground, your 'viewed' irssi window may even scroll past .. just my 2 cents Feb 08 18:00:17 UI::Window->{active} <-- Active window item Feb 08 18:01:25 yes, indeed. that's absolutely non-trivial Feb 08 18:01:47 you also want to deal with locked screen etc Feb 08 18:02:36 generally I'd consider it less hassle to have notifications even when I just stare at the trigger happening Feb 08 18:03:18 otherwise it would end with using frontcam to detect a pair of eyes focused on screen ;-) Feb 08 18:03:29 :) Feb 08 18:04:17 hmm u r right Feb 08 18:04:42 oh well now it vibrates till my usermode is away Feb 08 18:04:49 :) Feb 08 18:05:13 without printing dbus reply :) which is great Feb 08 18:05:25 but i might have a vulnerability there Feb 08 18:05:32 :> Feb 08 18:05:57 when Sicelo009N said "viewing" i got reply Feb 08 18:06:07 9 dbus-send: Data item "viewing" is badly formed Feb 08 18:06:27 scratch 9. Is part of time stamp Feb 08 18:06:50 heck, you shouldn't send data from IRC via dbus anyway Feb 08 18:07:05 i guess not Feb 08 18:08:04 your dbus-send script line is flawed Feb 08 18:08:21 it must not have any variables Feb 08 18:08:27 DocScrutinizer05: it actually works properly with Conversation Feb 08 18:08:46 and I think the relevant part is already in hildon-desktop Feb 08 18:08:48 sorry? Feb 08 18:09:04 proper notification handling vs active window and/or lockscreen Feb 08 18:09:41 the relevant part is has_focus() Feb 08 18:09:52 or is_visible() or sth like that Feb 08 18:10:43 the usual thing you frequently use to not do nonsensical power consuming display updates when the display isn't visible at all Feb 08 18:10:43 all you'd need for it to work properly is have hildon-desktop/mb2 generate a notification on urgency hint event Feb 08 18:12:33 anyway I'm maximum scared with the buggy state of dbus at large, on maemo Feb 08 18:12:59 hmm i changed the thing i suppose did that thing. So try to say my nick and "test" Feb 08 18:13:01 this is probably one of the most severe bugs we ever found Feb 08 18:13:12 vajb test Feb 08 18:13:14 2 Feb 08 18:13:15 3 Feb 08 18:13:16 4 Feb 08 18:13:17 5 Feb 08 18:13:21 vajb test2 Feb 08 18:13:40 and test with "" Feb 08 18:13:47 vajb "" ? Feb 08 18:14:12 i mean (vajb "test"( Feb 08 18:14:16 err Feb 08 18:14:26 vajb "test" ? Feb 08 18:14:32 yeah Feb 08 18:14:45 didn't print anything so i guess it is safe now Feb 08 18:15:05 i removed $message part Feb 08 18:41:21 btw. Was the current state of purple facebook plugin for maemo, solved not gonna happen? Feb 08 18:41:41 i remember reading something about need of newer glibc Feb 08 18:43:49 for pidgin in maemo to be exact Feb 08 19:21:54 Vajb_: new *glib*, not glibc Feb 08 19:22:03 and, yeah, it looked like a real mess Feb 08 19:22:27 (deps hell) Feb 08 21:23:19 is it just me, or on tmo, after replying to a private message, i cannot see my reply Feb 08 21:24:54 hmm, you have to go "Advanced" when replying. sucks that a checkbox for this is not available on 'default' reply box/page Feb 08 22:51:44 Vajb_: it's no bug, it's an error in wiki or wherever you got the vibrator dbus-send command from. It should have --type=method_call and there's actually no other way than redirecting output >/dev/null Feb 08 22:55:34 bah, still no closer to figuring out why so many https sites dont work now that I have updated the root CA certificate store Feb 08 22:56:47 isn't there a cmdline tool to evaluate the certificate chain of an arbitrary site? Feb 08 23:01:30 ssllabs.com and sslanalyzer.comodoca.com Feb 08 23:19:27 At this point my guess is that NSS in Maemo is too old to handle certain certificates in the new root CA store for some reason Feb 08 23:19:35 looks like online tools to evaluate a forgein or own server Feb 09 00:28:33 unrelated nice stuff: http://khenriks.github.io/mp3fs/ http://sysadminsjourney.com/content/2008/12/11/convert-flac-mp3-fly-mp3fs/ Feb 09 00:30:33 fstab: /home/jr/Musik /home/jr/mp3cnv fuse.mp3fs ro,users,bitrate=192 0 0 #still needs `mount ~/mp3cnv` by user, when root does it the whole thing is completely untouchable in userspace, prolly thanks to systemd and dang cgroups and stuff Feb 09 00:33:00 if the whole thing goes south (it does sometimes, when there are broken .flac files), you need `fusermount -u ~/mp3cnv` since umount blows chunks Feb 09 01:47:59 I want to assign and close some bug reports on bugs.maemo.org. Who can grant me permission to do it? Feb 09 01:48:38 Does somebody has such permission to assign/close/... bugs? Feb 09 01:54:53 Find assigned/closed bugs and look up who assigned-closed them? Feb 09 02:25:41 I'd send a mail to council, that's clearly a question for them to answer Feb 09 02:28:08 last bugmaster I seem to remember was Andre_Klapper(?) - council might have assigned the role to somebody else meanwhile Feb 09 02:30:47 I guess there are a few guys who can close bugs, however only council could assign the permissions to you to do so Feb 09 02:32:09 s/assign/grant/ Feb 09 02:32:10 DocScrutinizer05 meant: I guess there are a few guys who can close bugs, however only council could grant the permissions to you to do so Feb 09 02:38:08 http://wiki.maemo.org/Task:Document_maemo.org_processes Feb 09 02:39:17 http://wiki.maemo.org/Maemo.org_team Feb 09 02:39:42 http://wiki.maemo.org/Maemo.org_team#Bugmaster even **** ENDING LOGGING AT Tue Feb 09 02:59:59 2016