**** BEGIN LOGGING AT Fri Feb 16 03:00:02 2018
Feb 16 05:10:00 <brolin_empey> sixwheeledbeast: I was acknowledging that I had redd your reply to me from 10:14 UTC on Thursday.  Note that I do not know your geographic location.
Feb 16 08:39:50 <zGrr> moin :)
Feb 16 08:53:32 <sixwheeledbeast> brolin_empey: oh, I get you now, UK.
Feb 16 09:07:19 <Juesto> hi
Feb 16 10:21:47 <Wikiwide> Quite exhausted. How do I get routing working in Mappero?
Feb 16 10:23:14 <Wikiwide> I get "Connection terminated unexpectedly", most of the time.
Feb 16 10:23:25 <Wikiwide> Sometimes DBus NoReply
Feb 16 10:27:47 <Wikiwide> Granted, I am using cellular internet, because WiFi signal is weak. Either my N900 is damaged, modem's antenna is damaged, or settings need to be changed.
Feb 16 10:36:58 <Juesto> hi there
Feb 16 10:37:40 <Juesto> Wikiwide: probably just timing outs, not much to do given you're on cellular internet, a weak wifi signal may have speed
Feb 16 10:45:08 <Juesto> Wikiwide: were you ever able to connect to it?
Feb 16 10:45:30 <Juesto> s/it/successfully/
Feb 16 10:45:30 <infobot> Juesto meant: Wikiwide: were you ever able to connect to successfully?
Feb 16 10:45:39 <Juesto> damnit
Feb 16 10:45:41 <Juesto> anyways
Feb 16 11:15:15 <Vajb> update to my "touch screen acting weird when touched from top left corner" Today i noticed that if i "support" the screen by pressing it gently towards body from low left corner, it works as expected. So i guess my problem comes to worn slide mechanism.
Feb 16 11:27:47 <sicelo> 00:29 < DocScrutinizer05> sy: *which* packages depend on KP? .... seems to be KP-related packages, at least according to apt-cache rdepends
Feb 16 11:28:35 <Juesto> interesting
Feb 16 11:28:44 <sy> yeah, i don't think kernel power with uboot 'provides' kernel power
Feb 16 11:28:54 <sy> so packages incorrectly think it's not there
Feb 16 11:28:56 <Juesto> Wizzup: so parazyd got pc builds yet?
Feb 16 11:28:59 <sy> and try to flash it
Feb 16 11:29:08 <sy> then i say no and stuff breaks
Feb 16 11:30:12 <sicelo> sy: what is the package you're referring to as kernel power with uboot?
Feb 16 11:30:16 <Wizzup> Juesto: pc builds? do you mean for a VM?
Feb 16 11:30:34 <Wizzup> http://maedevu.maemo.org/images/virtual-machines/20180207/ amd64 vm images
Feb 16 11:31:05 <sy> sicelo: not at my n900 rn but something like Kernel Power (U-boot package)
Feb 16 11:31:13 <sy> from ham
Feb 16 11:31:29 <sicelo> that's just the zImage/uImage iirc :-)
Feb 16 11:31:57 <sy> yes
Feb 16 11:32:02 <sy> that's all i needed
Feb 16 11:32:14 <sicelo> so yes it doesn't really need to depend on KP, because *it is* KP ;)
Feb 16 11:32:23 <sy> oh
Feb 16 11:32:25 <sy> wait
Feb 16 11:32:28 <sy> i misread
Feb 16 11:32:40 <sy> ignore what i said 
Feb 16 11:32:43 <sy> I'm dumb
Feb 16 11:32:48 <sicelo> at least in my understanding 
Feb 16 11:32:51 <Juesto> its okay :)
Feb 16 11:33:18 <Juesto> Wizzup: well, yeah that's something, how much firmware does it support yet?
Feb 16 11:34:01 <sicelo> i just hope it installs kernel modules if they aren't already there
Feb 16 11:34:43 <Wizzup> Juesto: what do you mean
Feb 16 11:34:52 <Juesto> like, real hardware
Feb 16 11:34:58 <Juesto> and drivers
Feb 16 11:35:13 <Wizzup> it's a generic devuan kernel, so I assume so?
Feb 16 11:35:17 <Juesto> linux firmware = driver mega package supported by the kernel
Feb 16 11:35:19 <Juesto> oh fancy
Feb 16 11:35:53 <Juesto> so it should have decent speeds on a laptop theoretically 
Feb 16 11:38:32 <sy> sicelo: so the kernel power package i installed was Linux kernel for power user (boot image for U-boot)
Feb 16 11:39:10 <Juesto> are most of the bugs listed in maemo-leste bugtracker reproducible on original nokia?
Feb 16 11:39:18 <sy> and the package that diuretics l depends on the wrong package is Linux kernel for power user (settings and overclock)
Feb 16 11:39:25 <Wizzup> What do you mean, original nokia?
Feb 16 11:40:36 <KotCzarny> i guess he means stock/cssu
Feb 16 11:40:46 <KotCzarny> ie. fremantle
Feb 16 11:41:25 <Wizzup> I doubt that most of the issues/bugs apply. It's different userspace base (no libhal, but udev, upower, etc), different kernel, different kernel-userspace interfaces
Feb 16 11:44:50 <Juesto> Wizzup: many would appreciate a raw image of the system "hard drive"
Feb 16 11:45:13 <Juesto> for non-vm purposes and uncommon configurations due to the specialized software requirement imposed by those formats
Feb 16 11:47:39 <Wizzup> Juesto: many? :p 
Feb 16 11:47:44 <Wizzup> you're the first to ask about it
Feb 16 11:47:56 <Wizzup> you can also just install devuan ascii, with openrc as option, then add our repo and install hildon-meta
Feb 16 11:55:36 <Juesto> Wizzupwith the image it's easier ;)
Feb 16 11:55:56 <dreamer> Lazyesto
Feb 16 11:55:58 <dreamer> :P
Feb 16 11:56:10 <Juesto> and it has a useful purpose of offering maemo-leste standalone
Feb 16 11:56:17 <Juesto> heh :)
Feb 16 11:56:32 <Juesto> Wizzup: ^
Feb 16 11:56:35 <dreamer> Wizzup: my intention still is to try maemo-leste this weekend ;)
Feb 16 11:57:11 <KotCzarny> wizzup: sdcard image with fremantle boot integration would be nice for non-technical users
Feb 16 11:57:30 <Wizzup> dreamer: cool, hopefully we have the kernel battery bug fixed by then
Feb 16 11:57:49 <Wizzup> KotCzarny: not sure what exactly you mean - how would that work, and finally: want to work on it? ;) 
Feb 16 11:58:05 <KotCzarny> wizzup, dual boot?
Feb 16 11:58:21 <Juesto> booting up leste and maybe install on top of Fremantle to upgrade it?
Feb 16 11:58:28 <KotCzarny> for those who just want to try out without messing (too much) with their devices
Feb 16 11:58:42 <KotCzarny> leste and fremantle will be probably incompatible
Feb 16 11:59:00 <Wizzup> KotCzarny: do you mean a package for fremantle?
Feb 16 11:59:04 <Juesto> so either dual boot or upgrade path
Feb 16 11:59:22 <Wizzup> KotCzarny: we're still pre-alpha, so not a lot of thought has been given to update paths, and I don't rule out people will have to reflash images in the near future
Feb 16 11:59:33 <sy> was anyone able to get a security device working?
Feb 16 11:59:35 <Juesto> or yeah, try to improve Fremantle with backported FOSS blobs
Feb 16 11:59:47 <sy> like yubibey or a card reader
Feb 16 11:59:57 <sy> i just need usb host mode and pcsc
Feb 16 11:59:58 <KotCzarny> wizzup, sd-only os? with the only exception to add an item to uboot menu
Feb 16 11:59:58 <Wizzup> Juesto: that is explicitly not our goal, but I'm sure connman and icd2 can be added to cssu by someone
Feb 16 12:00:12 <Wizzup> KotCzarny: it already is sd-only ; we just need the uboot menu item
Feb 16 12:00:20 <Juesto> alright
Feb 16 12:00:28 <KotCzarny> :)
Feb 16 12:00:43 <Juesto> i know it's beyond the project but it's a good idea ;)
Feb 16 12:00:54 <Wizzup> KotCzarny: you just take the image, xzcat image.img.xz | dd of=/dev/<mmccard>
Feb 16 12:01:16 <Wizzup> KotCzarny: and them from uboot, select 'u boot console' and type 'run sdboot'
Feb 16 12:01:19 <Wizzup> s/them/then/
Feb 16 12:01:19 <infobot> Wizzup meant: KotCzarny: and then from uboot, select 'u boot console' and type 'run sdboot'
Feb 16 12:01:27 <Wizzup> c'est ca
Feb 16 12:02:00 <KotCzarny> then yeah, definitely updating uboot to autodetect or even contain static option to boot (chain boot?) to sdcard 
Feb 16 12:02:53 <KotCzarny> btw. how big is the packed image?
Feb 16 12:06:07 <Wizzup> KotCzarny: http://maedevu.maemo.org/images/n900/20180215/
Feb 16 12:06:24 <KotCzarny> hmm, 250M
Feb 16 12:06:26 <Wizzup> KotCzarny: there's also a sunxi image: http://maedevu.maemo.org/images/olimex-lime2/20180209/
Feb 16 12:06:35 <KotCzarny> it might be feasible to even make online installer
Feb 16 12:06:36 <KotCzarny> :)
Feb 16 12:06:48 <Wizzup> (but no mali in there yet)
Feb 16 12:07:09 <Juesto> netinstall?
Feb 16 12:07:30 <Juesto> that could work
Feb 16 12:09:26 <Wizzup> 12:59 < sy> i just need usb host mode and pcsc
Feb 16 12:09:44 <Wizzup> host mode should work eventually/soon enough, there's debugfs, so you can use testmode to force it
Feb 16 12:09:53 <Wizzup> and then just install usbutils, pcsc-lite, ccid and such
Feb 16 12:10:19 <Wizzup> bbiab.
Feb 16 12:12:18 <sy> Wizzup: debugfs?
Feb 16 12:12:39 <Wizzup> mount -t debugfs none /sys/kernel/debug (iirc)
Feb 16 12:12:39 <dreamer> Wizzup: ah, what's the bug?
Feb 16 12:12:46 <CatButts> butts
Feb 16 12:12:51 <dreamer> cats
Feb 16 12:13:00 <CatButts> by our powers combined
Feb 16 12:13:16 <Wizzup> dreamer: instead of reporting "discharging" is reports "not charging", which confuses upower, so it doesn't actually stop with the charging animation - https://marc.info/?l=linux-omap&m=151869726709281&w=2
Feb 16 12:13:27 <Wizzup> also see replies
Feb 16 12:13:34 <CatButts> what's new in the last two months?
Feb 16 12:13:37 <Wizzup> dreamer: btw, this might be better suited for #maemo-leste
Feb 16 12:13:44 <CatButts> my phone has received little love
Feb 16 12:13:56 <Wizzup> dreamer: yet another channel to add to the list :p 
Feb 16 12:14:09 <dreamer> Wizzup: hmmmmmmmm, ok maybe :)
Feb 16 12:15:15 <sy> Wizzup: why does a debug filesystem help with host mode?
Feb 16 12:15:24 <sy> not sure i understand
Feb 16 12:15:57 <Wizzup> sy: n900 doesn't really have 'otg pin', so you need to force the controller to host mode, which is done using debugfs
Feb 16 12:16:06 <Wizzup> there's a lot of info on it
Feb 16 12:16:08 <Wizzup> ~hostmode
Feb 16 12:16:08 <infobot> [hostmode] http://talk.maemo.org/showthread.php?p=652330, or http://www.youtube.com/watch?v=fkCDyUO0sKQ, or see ~hostmode-powered
Feb 16 12:16:21 <sy> oh
Feb 16 12:48:14 <sy> I'm getting key expired errors on some repos
Feb 16 12:49:01 <KotCzarny> ~maemo-repos
Feb 16 12:49:01 <infobot> hmm... maemo-repos is http://wiki.maemo.org/Repository#List_of_Maemo_repositories
Feb 16 12:49:22 <KotCzarny> but it's normal
Feb 16 12:49:25 <KotCzarny> since nokia is no more
Feb 16 12:52:37 <sy> right
Feb 16 12:52:56 <sy> but the errors are on the muarf repos
Feb 16 12:52:59 <sy> oh
Feb 16 12:53:05 <sy> there only mirrors
Feb 16 12:53:12 <sy> is there a way i can ignore this
Feb 16 13:08:36 <KotCzarny> yeah, just turn your eye away
Feb 16 13:08:37 <KotCzarny> ;)
Feb 16 13:19:03 * CatButts wonders if it's possible to cross-compile software for the N900 on windows
Feb 16 13:22:09 <CatButts> I suppose I'd need a specific version of GCC
Feb 16 13:22:22 <CatButts> and probably other complications as well
Feb 16 13:24:27 <CatButts> besides compiling, there's also packaging!
Feb 16 13:25:42 <KotCzarny> sure, install virtualbox and use vb image
Feb 16 13:27:18 <CatButts> I mean, having tools directly on windows
Feb 16 13:27:29 <CatButts> at least the crosscompile
Feb 16 13:27:33 <KotCzarny> useless
Feb 16 13:28:00 <CatButts> not even the crosscompile?
Feb 16 13:34:27 <CatButts> guess I will have to be stubborn and help myself, KotCzarny :p
Feb 16 13:34:37 <sy> can i flash kernel power then u boot?
Feb 16 13:34:54 <sy> or will it break things since u boot comes with a kernel
Feb 16 13:35:14 <CatButts> I think you need to install special KP for u-boot
Feb 16 13:35:37 <CatButts> u-boot will then load that KP
Feb 16 13:49:59 <sy> ugh
Feb 16 13:50:31 <sy> but packages depend on the kp that i can't use with uboot
Feb 16 14:06:54 <sy> Wizzup: confused about how to install pcsc-lite and ccid
Feb 16 14:08:04 <Wizzup> my comments were not for fremantle
Feb 16 14:08:12 <Wizzup> I suggest you don't even try to do it
Feb 16 14:08:30 <Wizzup> everything there  is too old to work with the keys (gnupg, pcsc-lite, ccid, etc)
Feb 16 14:08:31 <sy> oh shit
Feb 16 14:09:06 <sy> Wizzup: you meant leste?
Feb 16 14:09:26 <Wizzup> I was describing a hypothetical situation with recent kernel and distro, yes
Feb 16 14:09:41 <Wizzup> I wasn't recommending to do it now, but you can try to post host mode tools/scripts
Feb 16 14:09:50 <Wizzup> s/post/port/
Feb 16 14:09:50 <infobot> Wizzup meant: I wasn't recommending to do it now, but you can try to port host mode tools/scripts
Feb 16 14:10:08 <sy> right
Feb 16 14:11:23 <sy> could h-e-n work on leste?
Feb 16 14:11:44 <Wizzup> yes.
Feb 16 14:11:55 <Wizzup> but you will have to port it
Feb 16 15:19:42 * CatButts thinkens
Feb 16 15:22:06 <CatButts> first thing I should ask myself
Feb 16 15:23:08 <CatButts> is what dificulties would I encounter with using GCC versions other than the one used in Maemo SDK for N900
Feb 16 15:23:41 <CatButts> I want to make an SDK-in-a-zip type thing
Feb 16 15:24:38 <CatButts> windows and and linux have very different mentalities to how software should be installed/placed
Feb 16 15:25:13 <CatButts> windows has a monolithical "slap everything into a single folder" thing going
Feb 16 15:25:53 <CatButts> linux is "put application parts into corresponding places"
Feb 16 15:26:18 <CatButts> in the OS' folders
Feb 16 15:27:27 <CatButts> there's exists a minimal ext2 image with maemo SDK(circa 50mb)
Feb 16 15:27:50 <CatButts> for running directly on the N900 via chroot
Feb 16 15:28:19 <CatButts> no doubt, that has the required headers/sources/libs
Feb 16 15:28:50 <CatButts> and I am hoping to marry that with the appropiate crosscompiler
Feb 16 15:29:37 <CatButts> (for windows host)
Feb 16 15:29:57 <CatButts> for starters, I want to compile "Hello World" and run it on the device
Feb 16 15:33:15 <CatButts> anyone wish to hint me on some possible difficulties regarding old libs/kernel and gcc versions?
Feb 16 15:41:31 <pkill9> Guix aims to solve the problems caused by dumping different parts of programs into one folder, it's working quite nicely 
Feb 16 15:42:21 <pkill9> basically each software package gets it's own directory tree, even if within that directory tree it uses folders like /lib, /share, etc
Feb 16 15:42:58 <CatButts> ah, that's neat
Feb 16 15:43:09 <CatButts> but I am running windows
Feb 16 15:43:16 <pkill9> ah
Feb 16 15:43:32 <CatButts> and for windows I stubbornly want a toolkit
Feb 16 15:43:37 <CatButts> without VM images
Feb 16 18:11:29 <Juesto> CatButts: nope there isnt but nokia's ubuntu lucid vmware, it's for best that way
Feb 16 18:12:00 <Juesto> although i think something was made for Windows but no idea actually
Feb 16 18:12:13 <CatButts> I could attempt the same
Feb 16 18:13:28 <CatButts> at least, as far as collecting the right libs and headers/sources, the proper gcc version for them and making a working binary
Feb 16 18:13:45 <Juesto> you have it on that vm, CatButts
Feb 16 18:14:07 <Juesto> why waste time figuring out
Feb 16 18:15:04 <CatButts> so I rip those out of VM and supply crosscompiler myself
Feb 16 18:15:24 <CatButts> and make a windows doohickey!
Feb 16 18:15:46 <Juesto> again im not sure if there's anything for Windows
Feb 16 18:15:58 <APic> lol
Feb 16 18:16:14 <Juesto> CatButts: pointless anyway, cygwin and mingw
Feb 16 18:18:08 <CatButts> yes, but I'd have to have right versions of headers/libs to match the outdated stuff of Fremantle
Feb 16 18:40:03 <Juesto> use the vm
Feb 16 18:40:13 <Juesto> that one matches your needs
Feb 16 18:46:27 <CatButts> use the VM, Luke!
Feb 16 18:47:34 <APic> lol
Feb 16 18:48:08 <CatButts> Juesto, I stubbornly want to use tools directly on windows, without VM intermediary
Feb 16 18:50:56 <Juesto> then, you wont code
Feb 16 18:51:01 <Juesto> Google is your friend
Feb 16 18:51:07 <Juesto> idc
Feb 16 18:52:32 <Juesto> CatButts: All you can do in windows is make apps with qt creator
Feb 16 18:52:47 <CatButts> I don't like Qt
Feb 16 18:53:03 <Juesto> Then bad news for you
Feb 16 18:54:16 <CatButts> surely it's not impossible to compile helloworld.c on windows and have it run on the n900
Feb 16 18:56:28 <pkill9> just switch to linux
Feb 16 18:56:41 <CatButts> no
Feb 16 18:57:30 <Juesto> pkill9: ignore him, he said it twice
Feb 16 18:58:16 <CatButts> who ignore who and for what reasons, Juesto
Feb 16 18:58:26 <CatButts> *ignores
Feb 16 19:00:57 <DocScrutinizer05> sicelo: sy: yes, "Linux kernel for power user (settings and overclock)" obviously MUST NOT depend on KP directly, since KP comes in two packages: *-flasher (lives on NAND) and *-uboot-image (as a file)
Feb 16 19:01:28 <Juesto> why the separation?
Feb 16 19:01:54 <sy> some might not want to use power as a kernel
Feb 16 19:02:03 <sy> not all the time*
Feb 16 19:06:52 <DocScrutinizer05> Juesto: uBoot and kernel live in same NAND partition, and powerkernel leaves no space for uBoot
Feb 16 19:08:04 <DocScrutinizer05> so uBoot means you have stock kernel on NAND and when you want KP you need the KP-uboot-zimage or whatever the name
Feb 16 19:09:33 <Juesto> ah i see
Feb 16 19:09:50 <Juesto> so it is better to carry uboot separately if installing powerkernel?
Feb 16 19:10:27 <DocScrutinizer05> "Linux kernel for power user (settings and overclock)" needs to depend on a *property* "KP" that's provided by both. Or actually it doesn't need to depend on anything like that, rather check *AT RUNTIME* if everything as expected regarding kernel
Feb 16 19:10:28 <Juesto> s/installing powerkernel/powerkernel is installed/
Feb 16 19:10:28 <infobot> Juesto meant: so it is better to carry uboot separately if powerkernel is installed?
Feb 16 19:11:36 <DocScrutinizer05> when uboot is installed you need the KP-uboot-zimage
Feb 16 19:17:21 <DocScrutinizer05> when you had KP(-on-NAND) and then install uBott (with stock kernel) I wonder if the device will even bootloop since kernel modules might not match kernel
Feb 16 20:12:06 <Juesto> how well would the system work with kernel updates?
Feb 16 20:12:15 <Juesto> especially when uboot is separate and using power kernel
Feb 16 20:37:16 <sicelo> Juesto: some of it does not some care, but then u0boot is not a package for everyone
Feb 16 23:26:12 <DocScrutinizer51> kernel updates work as usual, just you need to update a zimage file instead of flashing a binary to a NAND partition
Feb 16 23:27:21 <DocScrutinizer51> the maemo stock kernel packed together with uboot as siamese twin file for flashing to NAND will never see any updates, I guess
Feb 16 23:28:10 <DocScrutinizer51> what you update is e.g. the uboot powerkernel image file
Feb 16 23:41:54 <CatButts> ooh, keepassx updated
Feb 16 23:42:10 <CatButts> I hope it's a pleasant surprise and a usable GUI :P
Feb 16 23:44:04 <CatButts> it is
Feb 16 23:44:22 <CatButts> they unfuckled the topbar menu
Feb 16 23:44:36 <CatButts> and made it usable on the n900
Feb 16 23:44:39 <CatButts> kudos to that
Feb 16 23:47:46 <pkill9> Wizzup: is it possible to tell the kernel to only charge up to 80% battery in maemo-leste?
Feb 16 23:48:02 <pkill9> if possible it would be good to have that option
Feb 16 23:48:17 <pkill9> nice CatButts
Feb 16 23:50:31 <CatButts> https://cdn.discordapp.com/attachments/348544204072288257/380981105232314368/screenshot00.png previous state :P
Feb 16 23:59:27 <sixwheeledbeast> updated
Feb 17 00:03:32 <sixwheeledbeast> Oh. 1 feb it was pushed to repo, still only v1 db's tho.
Feb 17 00:07:58 <pkill9> what does it look liek now CatButts?
Feb 17 01:10:08 <DocScrutinizer05> pkill9: that's a question of charging hardware more than a question of kernel. condonable but severe very common misconception
Feb 17 01:10:34 <DocScrutinizer05> not everything in a device is software
Feb 17 01:12:01 <pkill9> ok
Feb 17 01:13:37 <DocScrutinizer05> it's actually a controversial consideration if charging and battery management is a genuine kernel task at all, even when a particular hardware allows or requires software taking care about battery management
Feb 17 01:14:42 <DocScrutinizer05> anyway in N900 you can't really do charge-80%
Feb 17 01:15:26 <DocScrutinizer05> well, you could but that's a very hacky thing then, and for sure has no place in kernel
Feb 17 01:16:47 <DocScrutinizer05> btw charge-80 is way less effective than discharge-30
Feb 17 01:16:58 <DocScrutinizer05> regarding battery live span
Feb 17 01:18:14 <DocScrutinizer05> IOW don't deplete your battery further than 30, ideally 50% SOC, ever
Feb 17 01:19:12 <DocScrutinizer05> one 80->0 cycle is as damaging to battery as maybe 5 100->30 cycles
Feb 17 01:23:20 <pkill9> oh ok
Feb 17 01:23:25 <DocScrutinizer05> and unlike charge-80, the system actually could do shutdown at discharge-30, you even could configure that. Not a kernel issue either
Feb 17 01:23:35 <pkill9> nice
Feb 17 01:23:47 <pkill9> what about charge 800>30
Feb 17 01:23:54 <pkill9> 80->30 *
Feb 17 01:24:01 <pkill9> would that be the best?
Feb 17 01:24:30 <pkill9> if so, maybe just configure an alert sound when it reaches 80% charge
Feb 17 01:24:43 <DocScrutinizer05> as mentioned, it's not really supported by hardware to stop charging at 60%. you need nasty software hacks to make the phone stop charging before end-of-cahrge got detected
Feb 17 01:24:50 <pkill9> like my dumbphone beeps every now and then when it's fully charged and still plugged in
Feb 17 01:24:57 <DocScrutinizer05> s/60/80/ whatever
Feb 17 01:25:22 <pkill9> no i mean, is it best for battery longevity to stop charging at 80% and start charging at 30%
Feb 17 01:25:52 <pkill9> just manually disconnecting from power when it reaches 80% charge
Feb 17 01:26:16 <DocScrutinizer05> yes, that would be best for battery. well almost. better is 50<->85
Feb 17 01:26:53 <pkill9> shame battery tech sucks on smartphones
Feb 17 01:26:54 <DocScrutinizer05> disconnecting from battery however means you start another discharge cycle
Feb 17 01:27:08 <DocScrutinizer05> err disconnecting from charger
Feb 17 01:28:16 <DocScrutinizer05> while N900 would either start recharging when battery falls to 76 or 80% and stops when battery charge current falls below a threshold of iirc 16mA
Feb 17 01:29:11 <DocScrutinizer05> and a 100->75->100 cycle is easier to the battery than a 80->30->80 cycle
Feb 17 01:29:35 <DocScrutinizer05> even assuming you would re-connect the charger as soon as battery falls to 30%
Feb 17 01:30:30 <pkill9> ok
Feb 17 01:30:31 <DocScrutinizer05> odds are you forget it and battery discharges to EndOfCharge voltage, that does damage to the battery
Feb 17 01:32:14 <DocScrutinizer05> as a matter of fact, my N900 are udling at charger and charged to 100% at 96% of the time, hardly ever discharging further than 50%, and the batteries survived that in pretty good condition for 5 and more years
Feb 17 01:35:32 <DocScrutinizer05> this minute:  http://paste.opensuse.org/4214364
Feb 17 01:36:05 <DocScrutinizer05> usually it has even lower floating charge current, but I used it off-charger for a 10 minutes some gours ago
Feb 17 01:36:22 <DocScrutinizer05> hours*
Feb 17 01:38:15 <DocScrutinizer05> http://paste.opensuse.org/82182452
Feb 17 01:38:35 <DocScrutinizer05> might also be the Polarcell that behaves a tad different
Feb 17 01:44:04 <DocScrutinizer05> Neo900 charger design is superior. That's why you could start up a Neo900 without any battery at all inside, or with completely depleted battery. No bootloop deadlock for flashing ;-)
Feb 17 01:44:37 <DocScrutinizer05> ~flatbat
Feb 17 01:44:45 <DocScrutinizer05> ~flatbat-recover
Feb 17 01:44:55 <DocScrutinizer05> damn
Feb 17 01:45:08 <DocScrutinizer05> ~listkeys flatbat
Feb 17 01:45:09 <infobot> Factoid search of 'flatbat' by key (2): flatbatrecover #DEL# ;; flatbatrecover.
Feb 17 01:45:17 <DocScrutinizer05> ~flatbatrecover
Feb 17 01:45:17 <infobot> Remove battery for 1 minute. Insert battery. Plug powered ***NOKIA WALLCHARGER*** to device. Watch steady amber. Let sit and charge. Do NOT try to boot. After 30 min, you got either a) a booted up N900, b) flashing amber which means you can boot, c) steady amber shut off -> start over again with ~flatbatrecover while already searching for a new battery. CAVEAT! Only works when ~rootfs OK (no ~bootloop)
Feb 17 01:46:00 <DocScrutinizer05> ~bootloop
Feb 17 01:46:00 <infobot> i guess bootloop is when your device has broken rootfilesystem, so during reboot it fails on some service startup or kernel module load and thus reboots. This *drains* battery! And you can't reflash to stop bootloop when battery is drained. Recharge your battery by other means before reflashing. E.g. using ~rescueOS. Or external charger or BL-5J compatible other device.
**** ENDING LOGGING AT Sat Feb 17 03:00:00 2018