<jonwil>
Looks like the leaked PowerVR stuff has caused a bit of a kerfuffle with the guy who posted the original git repo (the contents of which seem to have come from a tarball on the LG open source website) being sent legal threats by some guy who claims "I worked on PowerVR for N9 and I think sharing this code is bad"
jonwil has quit [Quit: ChatZilla 0.9.91 [SeaMonkey 2.30/20141013232806]]
<wpwrak>
you can't put the genie back into the bottle :)
che1 has joined #neo900
arcean has joined #neo900
che1 has quit [Remote host closed the connection]
che1 has joined #neo900
mvaenskae has joined #neo900
ShadowJK has quit [Ping timeout: 258 seconds]
ShadowJK has joined #neo900
astr has quit [Ping timeout: 272 seconds]
astr has joined #neo900
SylvieLorxu has joined #neo900
<raccoon_>
tell the guy "thanks for your great work! please put this repo on your resume"
mvaenskae has quit [Ping timeout: 244 seconds]
<bencoh>
:]
<wpwrak>
i wonder in what sense he thinks it's "bad". one bad aspect i can see is that powervr may now become more acceptable for open source, while before us freedom-lovers pretty much knew that a chip that had powervr on it was just as "accelerated" as a dumb frame buffer
<freemangordon>
(in short) read the ful article if interested
<freemangordon>
*full
<SylvieLorxu>
freemangordon: Maybe if he wouldn't treat people so badly ("DO YOU HATE DISABLED PEOPLE?" is not a good argument) he wouldn't be so "mistreated". You reap what you sow
ecloud_wfh is now known as ecloud
ravelo has joined #neo900
<freemangordon>
SylvieLorxu: not that I like kerio's "jokes" that much, but LP's mourn is based on something that never happened. Read the whole article if you didn't already, he attacks Linus and the whole FOSS community, because they "hate" him. Giving kerio's post(put out of the context) as an example. Come on.
<DocScrutinizer05>
actually he is poettering on the "could we collect money to hitr a hitman?" and twists this "joke" into a serious threat for his life. Pathetic
<DocScrutinizer05>
which is an equally bad argument/justification to push poorly designed software down everybody's throat, given in the same (hijacked) "talk" by same LP like the "DO YOU HATE DISABLED PEOPLE?"
<SylvieLorxu>
I like this part myself, going on stage during someone else's presentation, holding bear and talking complete nonsense: https://www.youtube.com/watch?v=_ERAXJj142o#t=53m45s
<freemangordon>
well, looks like he's a celebrity star, what do you expect :)
<DocScrutinizer05>
bth I never thought calling this guy LP since it's way too close to LJP aka Lorn J Potter, a nice guy of qt crew or sth
<DocScrutinizer05>
btw*
<DocScrutinizer05>
Porter?
<freemangordon>
the whole name is too complicated for my laziness to type it :)
<DocScrutinizer05>
"poettering" is easy
<SylvieLorxu>
s/bear/beer/
<SylvieLorxu>
Wow, my typing
ravelo has quit [Ping timeout: 246 seconds]
<SylvieLorxu>
I'd like to see him holding a bear though
<SylvieLorxu>
Could be fun
<DocScrutinizer05>
yeah
<DocScrutinizer05>
a polar bear, pretty please!
<DocScrutinizer05>
the whole talk is epic
<SylvieLorxu>
When he started with "Do you hate disabled people?" I would've just called security
<SylvieLorxu>
Ad hominem is no way to discuss
<DocScrutinizer05>
the original talker makes several very good points, and finally poettering can't stay quiet anymore and has to hijack the stage and defend the crap software with crap arguments
<DocScrutinizer05>
yeah, and for benefit of disabled people, next systemd will absorb linux kernel, so the boot messages could get read out loud to the blind via polypaudio
<SylvieLorxu>
I must say that I've never seen any software getting so much hate as software by Poettering
<DocScrutinizer05>
and all this for the bazillion people out there who asked for it, unlike the bunch of freaks in this room who have no idea about the usual way a linux station is used
<SylvieLorxu>
I wish he'd take the amount of complaints into consideration too
<SylvieLorxu>
Most software don't get so much anger
<DocScrutinizer05>
indeed
<SylvieLorxu>
The simple fact that the anti-systemd movement is so big should tell you something
<SylvieLorxu>
Oh well
<DocScrutinizer05>
not to poettering and the systemd cabal
<SylvieLorxu>
At least, with a lot of work, we can have our own systemd-free distros
<DocScrutinizer05>
not for long anymore
<SylvieLorxu>
Although I'm still not happy having to switch, but seeing his history, I'd rather not run systemd on my servers (9 CVEs already, horrible...)
<DocScrutinizer05>
yeah
<DocScrutinizer05>
no surprise
<SylvieLorxu>
There's a group of people planning to fork Debian, but I agree, there is was less choice then there should be
<SylvieLorxu>
I mean, sure, systemd has a few nice ideas
<SylvieLorxu>
But the attack surface is really too big
<DocScrutinizer05>
you can't build a spaceship from dogshit
<SylvieLorxu>
And there are some pretty basic design mistakes
<DocScrutinizer05>
nice ideas?
<DocScrutinizer05>
hardly any I know of
<SylvieLorxu>
Well, nothing really original, but some things would be less work with systemd (like hard-to-tamper-with logs)
<SylvieLorxu>
Still, the benefits don't outweigh the costs
<DocScrutinizer05>
parallelism in boot? dead old stuff and doesn't work in systemd any better than in sysV init
<DocScrutinizer05>
LOGS??? BINARY LOGS??
<SylvieLorxu>
Haha
<SylvieLorxu>
That's what I mean with the benefits not outweighting the costs
<Pali>
even worse: binary logs in database without ACID
<DocScrutinizer05>
I'm already going berserk for my mail stored in mysql by Kontact. Guess what I'm going to think about syslogs stored in a database
<Pali>
mysql is better then systemd database
<SylvieLorxu>
The fun part is that log corruption is considered "not a real issues" and should "just be ignored" ;)
<DocScrutinizer05>
Pali: right
<SylvieLorxu>
Yay, regressions
<DocScrutinizer05>
you should drop a PDP-11 on Poettering and the whole systemd cabal
<Pali>
DocScrutinizer05: year ago I was extremly fed up to that mysql and kontact... and I had to fix it = remove akonadi and nepomuk from kontact
<Pali>
systemd idea is like akonadi...
<DocScrutinizer05>
yes, exactly. EXACTLY
<Pali>
both is studid and cause headache
<DocScrutinizer05>
actually even both almost identical concept
<Pali>
systemd is different... they created own database
<DocScrutinizer05>
doesn't make much difference
<Pali>
kde people (re)used some working database
<DocScrutinizer05>
akonadi can get paired with a number of different db too
<Pali>
finally?
nox- has joined #neo900
<Pali>
there were problems with special sql keys and columns...
<Pali>
and special mysql syntax
<DocScrutinizer05>
it's not about own vs existing db, it's about a giant daemon blob using db to store EVERYTHING in it and provide EVERYTHING services to a skin of "apps" that are actually just that: skins
<SylvieLorxu>
That's a serious difference between Poettering and KDE
<SylvieLorxu>
KDE, even though they also sometimes have bad ideas, try their best to support as much as they can
<DocScrutinizer05>
did you know that POP3 and IMAP is done by akonadi and kontact only asks for "please fetch the mail from this account"?
<Pali>
akonadi is like cache for pop and imap
<Pali>
there are clients which downloading data from imap to akonadi
<Pali>
and there are apps which reading those data from akonadi
<DocScrutinizer05>
hey, you can kill -9 kontact and your mail still gets fetched... by akonadi :-(((
<Pali>
yes :-(
<Pali>
idea about abstract interface for email access is ok
<Pali>
link to source code, git repo and packages for ubuntu ^^^
<DocScrutinizer05>
even better: boot KDE and have system settings like "allow indexing of mails" in this dunnowhatshit. akonadi gets started and doesn't just index mails but also fetches new mails, despite not a single mail program/app ever been started by user
<Pali>
for all kdepim/kontact for KDE4 without akonadi and nepomuk
<DocScrutinizer05>
NEPOMUK!! that been the damn name
<Pali>
and that kmail ^^^ support searching in emails without nepomuk
<DocScrutinizer05>
coool
<DocScrutinizer05>
actually it's prolly like: KDE sees "index mails" and starts nepomuk, nepomuk starts akonadi to get access to mails for indexing, and akonadi happily drifts away fetching mails.
<DocScrutinizer05>
consider using such crap in a professional environment. You prolly get fired
<DocScrutinizer05>
after site sysop locking your account
<DocScrutinizer05>
"sorry Mr Reisenweber, we told you 100 times NOT to access the internet from inside the lab intranet"
<Pali>
consider using such crap on your own notebook. You prolly discharge battery in 10 minutes!
<DocScrutinizer05>
that too
<Pali>
and try to use it with 100000+ emails
<Pali>
you will find, download and compile new email client until that allows you to read new emails
<Pali>
it delegates operations to imap server as much as it can
<DocScrutinizer05>
and now think about systemd - it also has DNS resolver and whatnot meanwhile, I heard. Must be an absolute nightmare to configure that shit in a way so it doesn't do same damage to security that akonadi does by inadvertedly polling mails
<DocScrutinizer05>
I try to avoud imap as much as possible. imap is a design error
<DocScrutinizer05>
now find the mail that bashed Poettering for "hey fool, don't you think it's a bit risky to have DNS in init? Do you *really* think _you_ are the God that gets it right from beginning? Please don't do that nonsense!"
<Pali>
DNS cache poisoning
<DocScrutinizer05>
and Poettering's answer "you have no clue, and systemd has no bugs, EVER. STFU!"
<DocScrutinizer05>
kdbus already made me puke
<Pali>
pavelm told that poettering could not be an idiot... idiot cannot break 1/3 of all world computers...
<DocScrutinizer05>
...but now it's more than obvious - systemd *will* conquer the kernel during next 2 years
<DocScrutinizer05>
you mean it needs a evil genius to do so? ;-P
<Pali>
we will see if kdbus will be accepted or not
<Pali>
Greg - just for your information, I will *not* be merging any code from Kay into the kernel until this constant pattern is fixed.
<DocScrutinizer05>
won't
<DocScrutinizer05>
you canot fix a Poettering or any of the other guys in systemd cabal
<Pali>
just quote that Linus post again if there will be next round of kdbus...
<DocScrutinizer05>
they *are* the pattern, they live it
<DocScrutinizer05>
the true problem however is why all this happens. I suspect that's not Poettering and friends, that's a few geniuses in RedHat that made a nice plan how to conquer the world
<kerio>
i thought it was widely known
<DocScrutinizer05>
way up in RedHat hierarchy, some business dudes
arcean has quit [Quit: Konversation terminated!]
<kerio>
if you control an important part of the linux ecosystem you basically become the only real option for support contracts
<DocScrutinizer05>
yes
<DocScrutinizer05>
exactly
<DocScrutinizer05>
RedHat taking linux world as hostage
<kerio>
not as hostage
<kerio>
as a collateral
<DocScrutinizer05>
and for that purpose it's completely irrelevant if systemd is _better_ or whatever, it only matters that systemd is without alternative
<Pali>
linux kernel is still and will be controlled by Linus... and this is big problem for those people in RedHat
<Pali>
there are alternatives to systemd!
<Pali>
but alternatives which working :-)
<DocScrutinizer05>
not really. Same way FOSS kernel wasn't a problem for Nokia, they nevertheless implemented Aegis
<Pali>
but who cares about Nokia kernel? nobody... it died
<DocScrutinizer05>
yes, there was alternative "openmode" (bwahahaha) but it killed your system
<DocScrutinizer05>
same way there's an alternative to running your SLES with systemd: don't use SLES
<Pali>
yes, you can run gentoo or slackware
<kerio>
ok, serious talk time
<kerio>
your obsession with suse needs to stop
<DocScrutinizer05>
meh, make that RedHat then
<DocScrutinizer05>
RHEL
<kerio>
of course there's an alternative to RHEL with systemd
<kerio>
centos with systemd!
* kerio
flees
<Pali>
:D
<DocScrutinizer05>
doesn't matter, the fact is: ripping out systemd will wreck the system and you're better off just switching to another distro
<kerio>
eeh
<kerio>
apt-get install sysv-init
<kerio>
i wonder how well it'll work
<DocScrutinizer05>
the problem froma more distant POV is: many sysadmins will think "rather than using a pseido-windows based on windows-me-too stuff invented by a few silly nerds, I rather just switch to the real thing and be a servant of M$ rather than a servant of Mr Poettering"
<DocScrutinizer05>
so basically what RedHat does is "let's nuke the linux thing and scavenge all the rocks that did _not_ come down in M$ land. They will all be ours then"
<Wizzup>
openrc is a nice alternative
<DocScrutinizer05>
Poettering pretty bluntly uttered that linux should get rid of this "stay compatible with *nix" paradigm since that allegedly stops innovation. He also clearly stated that the whole cabal isn't interested at all in embedded etc, those user groubs can get lost in a dark place or switch to BSD or whatever. Linux is for workstations and nothing else, according to Poettering
<Wizzup>
he also stated he doesn't think his printer is a file
<Wizzup>
because it's a printer
<DocScrutinizer05>
sorry, that's a statement of an idiot
* DocScrutinizer05
beats poettering with a paper printout of PLAN9
<Pali>
why shouldnt poettering create and use its own shits on his own computers... and nowhere else???
<DocScrutinizer05>
that's the question. Actually poettering already _did_ a fork away from true linux with such statements
MonkeyofDoom has quit [Ping timeout: 258 seconds]
<DocScrutinizer05>
The question is: should we allow him to take the "linux" trademark and run away with it?
<DocScrutinizer05>
or actually "GNU" ;-P
<DocScrutinizer05>
which makes me wonder what's RMS' take on all that
<DocScrutinizer05>
prolly the future is dark when we ask RMS to decide
<DocScrutinizer05>
I wonder who claims ownership of init. Clearly not the kernel guys. So Linus is out. GNU aka RMS? Prolly neither.
<kerio>
ownership of init?
<DocScrutinizer05>
yeah, you know... GNU-linux
<DocScrutinizer05>
linux is the core (kernel, plus what else?)
<DocScrutinizer05>
GNU is... well GNU
<DocScrutinizer05>
but who owns init?
<DocScrutinizer05>
seems it will be Gnome-init in a few years ;-P
<DocScrutinizer05>
so RedHat claimed it to be theirs now
<kerio>
i don't think they call it init
<kerio>
they call it systemd
<DocScrutinizer05>
haha, it's still the init process, do a "top"
<kerio>
no shit
<kerio>
ls can be from coreutils or from bsd
louisdk has joined #neo900
<kerio>
or busybox
<kerio>
it's still ls
<DocScrutinizer05>
yeah, busybox prolly has a symlink "/sbin/init" too ;-P
<kerio>
sure
<kerio>
it can do init
<kerio>
without runlevels
<kerio>
rofl
<DocScrutinizer05>
does it do sysV, systemd, upstart... what else?
<kerio>
it parses inittab
<kerio>
it's like sysv but without runlevels
<DocScrutinizer05>
sounds like sysV
<kerio>
yeah it's basically a simplified sysv
<DocScrutinizer05>
sorry, I'm a tad dizzy today. Prolly shouldn't even engage in such involved discussions
<kerio>
i wonder if they'll update it to make it simplified systemd :P
<kerio>
busybox dbus
<DocScrutinizer05>
WAAAAHHHH!
MonkeyofDoom has joined #neo900
<DocScrutinizer05>
did I tell you that I never really trusted dbus and still think it's a can of bugs and worms?
<kerio>
it's ok, now it's gonna be in the kernel
<DocScrutinizer05>
ooh good¡
<kerio>
well
<kerio>
i mean
<kerio>
let's put it this way
<kerio>
now it's going to be maintained by the kernel maintainers
<DocScrutinizer05>
hey, that's actually a bright thing
<DocScrutinizer05>
HEHE, the one to rule them all
<DocScrutinizer05>
when Poettering thinks he can conquer kernel, kernel guys should shanghai kdbus and force Poettering stuff out of dbus ;-P
<DocScrutinizer05>
I mean, what's all the poettering cruft without control over dbus? ZILCH ;-P
<Oksana>
Three Rings for the Elven-kings under the sky, Seven for the Dwarf-lords in their halls of stone, Nine for Mortal Men doomed to die, One for the Dark Lord on his dark throne In the Land of Mordor where the Shadows lie. One Ring to rule them all, One Ring to find them, One Ring to bring them all and in the darkness bind them In the Land of Mordor where the Shadows lie.
wazrus has joined #neo900
<DocScrutinizer05>
Oksana: thanks for helping me out :-D
<Oksana>
Good morning :-)
<DocScrutinizer05>
kinda. Same to you
<DocScrutinizer05>
before you ask:
<DocScrutinizer05>
news: Werner and me have defined a first draft of hackerbus
<Oksana>
How is it going with Snaptron? Gerry? 1GB RAM? :-) Hackerbus... Now I can look up how difficult it is to attach lightning detector, or (to back-cover) mouse buttons and mouse wheel and mouse optical sensor, or something else.
<DocScrutinizer05>
the above news were comprehensive
<DocScrutinizer05>
this is only the pinout. Using 1.27mm pitch post connector receptacle
<DocScrutinizer05>
and as usual: Subject to change without further notice
<Oksana>
I was looking up LCD panels. 3.5 inch LCD panel with aspect ratio of 5:3 and resolution higher than 800x480 is impossible to find. Why? Didn't they develop 400+ ppi technologies?
<Oksana>
Hackerbus: USB: I can see ground, minus and plus. What about red wire +5V?
<DocScrutinizer05>
VBUS?
<Oksana>
Thanks! Okay, so now we have USB-to-modem inside device, USB-micro for external usage, and USB on hackerbus. Anything else I missed?
wicket64 has joined #neo900
<DocScrutinizer05>
the USB on hackerbus is a duplicate of the regular USB-OTG
<DocScrutinizer05>
all pins shall be reasonably protected against electrical abuse
<DocScrutinizer05>
(e.g. we of course provide full ESD and even a OVP planned to protect against damage up to +-12V)
<DocScrutinizer05>
we're still pondering if the logic levels are the general 1.8V GPIO level seen on all GPIO of SoC, or if we add level shifters (prolly the latter) to allow connecting any logic from 1.8V up to 5V
<DocScrutinizer05>
several of the pins have a solder option to reassign them. Alternative signals include audio-in/out, USB ID-pin, I2C bus #2, you-name-it
<DocScrutinizer05>
*maybe* we can even get modem-USB D+/- on the laternative signal group
<DocScrutinizer05>
VBAT_RAW is what the name suggests: direct line to Batt plus, for charging and high current discharge
<DocScrutinizer05>
VBAT_SWITCHED is also what name suggests, it is Vbatt ülus switched by system to be off when system is powered down
wazrus has quit [Quit: Lost terminal]
<DocScrutinizer05>
s/ü/p/
<DocScrutinizer05>
the GPIOs will get some logicaly assigned primary purpose like "WAKE_CPU", "MODEM_RING" etc, but those are just by convention to provide a sort of a "standard". You're free to use each GPIO any way you like
<DocScrutinizer05>
at least two of them GPIOs are capable to trigger SoC_IRQ/WAKE
<wpwrak>
tracking name changes ...
<DocScrutinizer05>
UART is #3, which on N900 is on the test pads under battery, has console, and in Neo900 is also used for IrDA/CIR
<DocScrutinizer05>
wpwrak: huh? sorry I didn't mean to introduce name changes
<wpwrak>
already done ;-) besides, that we they're easier to keep apart
<wpwrak>
now all names but 1V8 and 3V3 are even valid C identifiers
<DocScrutinizer05>
sounds good
<wpwrak>
for the gpios i'd want the real names on the CPU, but that'll have to wait until they're assigned. so "a" and so on for now
<DocScrutinizer05>
ooh, the 1V8 and 3V3 prolly need some explanation: those are GPIO with alternative power supply controlled in software
<DocScrutinizer05>
so you can either use the pin as GPIO or switch it to power supply
<DocScrutinizer05>
if anytrhing, then those are most likely to change
<DocScrutinizer05>
maybe nobody is interested in a 1V8 supply rail