IRC log for #neo900 on 20140929

00:25.21*** join/#neo900 nicksydney (~quassel@130.136-26-211-net25.dynamic.dsl.syd.iprimus.net.au)
03:31.35OksanaWhat is happening with transmission lines? And dome sheets?
03:34.36*** join/#neo900 M13 (~MirandaLS@83.149.35.84)
04:28.18*** join/#neo900 MonkeyofDoom (~MonkeyofD@c-73-8-136-178.hsd1.il.comcast.net)
05:09.57*** join/#neo900 mva (mva@gentoo/contributor/mva)
06:21.52*** join/#neo900 mvaenskae (~mvaenskae@unaffiliated/mvaenskae)
07:07.55*** join/#neo900 kolp (~quassel@55d41c7e.access.ecotel.net)
08:21.22*** join/#neo900 freemangordon_ (~ivo@195.128.224.219)
08:33.06*** join/#neo900 mvaenskae (~mvaenskae@unaffiliated/mvaenskae)
08:55.39*** join/#neo900 modem (~modem@LAubervilliers-656-01-215-246.w217-128.abo.wanadoo.fr)
08:55.43*** join/#neo900 modem (~modem@fsf/member/modem)
09:51.36*** join/#neo900 sixwheeledbeast (~paul@2a01:348:6:860a:3064:8dc0:74c2:8237)
10:33.54*** join/#neo900 hinton (quassel@2a01:7e00::f03c:91ff:fedb:2ffe)
11:21.34*** join/#neo900 enyc (~enyc@muddle.enyc.org.uk)
11:59.47jonwilhttp://talk.maemo.org/showthread.php?t=93938
12:16.10*** join/#neo900 Kabouik (~quassel@147.99.218.243)
12:26.26enycwonders why n900 [not noe900] might like to fgive small rotated display sometimes on bootup... i wonder if there was a known buf with rotation sensors or if its sjust software bug
12:29.31*** join/#neo900 b1101 (~b@108.61.152.243)
12:29.31*** join/#neo900 b1101 (~b@fsf/member/b1101)
12:30.27wpwrak(looking through jonwil's post) /sys/bin/CameraAlgProc ... wow, now that looks evil
12:55.41*** join/#neo900 kerio (kerio@acehack.de)
13:16.47*** join/#neo900 e2718 (~hdesk@p4FD3705D.dip0.t-ipconnect.de)
13:45.35*** join/#neo900 modem (~modem@LAubervilliers-656-01-215-246.w217-128.abo.wanadoo.fr)
13:45.35*** join/#neo900 modem (~modem@fsf/member/modem)
13:49.33*** join/#neo900 mvaenskae (~mvaenskae@unaffiliated/mvaenskae)
14:11.13*** join/#neo900 b1101 (~b@209.222.18.35)
14:11.13*** join/#neo900 b1101 (~b@fsf/member/b1101)
14:45.00*** join/#neo900 e2718 (~hdesk@p4FD3705D.dip0.t-ipconnect.de)
15:08.10*** join/#neo900 Pali (~pali@Maemo/community/contributor/Pali)
15:27.40DocScrutinizer05wpwrak: I guess that's the semi-proprietary image-enhancement software Nokia's camera manufacturer provided for that particular camera
15:29.58DocScrutinizer05re mozilla parcel, I got mistaken by those forms fedex sent me to sign. A call to them turned out that mozilla pays for all expense of this shipping, incl any VAT and import tax (fedex guy - extremely kind one - says "that's usual since they can book the VAT as expense")
15:32.07mvaenskaeDocScrutinizer05: o m g
15:32.13DocScrutinizer05just let's hope mozilla doesn't invoice Neo900 UG in turn, for the VAT and import of 5 dozen totally overpriced N900 and defect BL-5J
15:32.14mvaenskaethat is awesomesauce
15:32.27mvaenskaewell, that would be pretty uncool
15:32.43bencohoO
15:32.50mvaenskaeall batteries dead but the devices are still working?
15:33.01mvaenskaeso you got 60 devices?
15:33.13DocScrutinizer0555, and I haven't seen them yet
15:33.35keriocan i get a couple?
15:34.02DocScrutinizer05kerio: depends. When you're doing development, we for sure will send you a loaner
15:34.15keriobut that requires doing stuff
15:34.48*** join/#neo900 Pali (~pali@Maemo/community/contributor/Pali)
15:35.05mvaenskaeDocScrutinizer05: may i ask what you are going to use them for?
15:35.18DocScrutinizer05for devel loaners
15:35.32DocScrutinizer05and to base a few prototypes on them
15:36.01keriodon't you have like 8 of your n900s for that?
15:36.08mvaenskaehm, devel for the n900? why not devel for the neo900 or is the hardware that similiar?
15:36.20DocScrutinizer05yes, it's that similar
15:36.37DocScrutinizer05that's the main design rule for Neo900
15:36.42keriohold on, is it for fptf?
15:36.48DocScrutinizer05yes
15:36.54DocScrutinizer05among others
15:37.10DocScrutinizer05of course also SHR, replicant, whatever
15:37.12keriohow can you port maemo to the neo900 if you're actually porting it to the n900?
15:37.40DocScrutinizer05kerio: since both are supposed to be hardware compatible to a certain degree
15:37.45keriook actually
15:37.50keriowill fptf's os also work on n900?
15:37.57DocScrutinizer05should
15:37.57keriobecause i'd like a more recent kernel
15:38.07kerioand no busybox
15:38.18bencoh(the last one is unrelated :])
15:38.30bencoh(s/last one/latter/)
15:38.32kerio([:)
15:38.56DocScrutinizer05anyway, just had to report briefly in here. AFK again
15:38.59DocScrutinizer05RL caling
15:42.46*** join/#neo900 JoHnY (~johny@imaginarium.2600.sk)
15:46.19*** join/#neo900 sixwheeledbeast (~paul@2a01:348:6:860a:3064:8dc0:74c2:8237)
15:49.36DocScrutinizer05(<kerio> don't you have like 8 of your n900s)  well, I'm up to 12some on the one hand, on the other hand I already "used" a maybe 5 of them for R&D
15:49.56DocScrutinizer05for proto_V2 we will need another 6 ones
15:50.00kerioi see
16:01.55*** join/#neo900 modem (~modem@LAubervilliers-656-01-215-246.w217-128.abo.wanadoo.fr)
16:01.57freemangordonhmm, maybe REing PA module-nokia-voice etc won't be that hard
16:02.00*** join/#neo900 modem (~modem@fsf/member/modem)
16:02.18freemangordonmost of the stuff in it is same (or similar) to meego modules
16:02.55freemangordonwe will only need to RE xprot, xprot-ambient-temp and eq parts
16:20.27freemangordonDocScrutinizer05: eve seen http://nielsmayer.com/meego/N900-Pulseaudio-Configuration.png ?
16:20.33freemangordon*ever
16:21.11bencohwoot
16:21.32freemangordonhmm?
16:21.52bencohI hadnt, and it's interesting :)
16:22.22freemangordonme neither, just found it while searching for some n900 PA related stuff :)
16:22.44freemangordonnot sure if it is the same on maemo, but should be
16:54.00DocScrutinizer05freemangordon: well, I kinda seem to know that, yeah. Though from other source
16:54.15freemangordongood
16:59.20DocScrutinizer05http://linuxplumbersconf.org/2009/slides/Jyri-Sarha-audio_miniconf_slides.pdf   p.15
17:01.27freemangordonyep, the same
17:08.03*** join/#neo900 paulk-collins (~paulk@gagarine.paulk.fr)
17:11.24*** join/#neo900 Kabouik (~quassel@200.177.138.88.rev.sfr.net)
17:12.41*** join/#neo900 Kabouik_ (~quassel@200.177.138.88.rev.sfr.net)
17:28.33rax<DocScrutinizer> I just sent you an email with some thoughts on external auditing of peripheral sandboxing logic.
17:52.46DocScrutinizer05umm Paul? Yup
17:54.12DocScrutinizer05we however have no sandboxing, we have total separation of address space of any peripherals and the Linux CPU
17:55.09DocScrutinizer05_all_ cimmunication between CPU and any paripheral is under 100% control on protocol level by CPU
17:56.09raxOk, so the CPU has to use memory mapped I/O (or something similar) to get to the memory space the peripherals see?
17:56.33DocScrutinizer05for the CPU (and security concerns) it doesn't make a difference the peripherals are embedded into same case. They as well could sit on next desk, or in Australia and being attached via internet
17:57.01raxThat must make it fun to modify drivers.  :-)
17:57.06DocScrutinizer05no, we talk to peripherals via e.g "RS232"
17:57.31raxWell, hopefully you're faster than an RS-232.  :-)
17:57.35DocScrutinizer05there's no such thing like address space involved at all
17:57.55raxOk, so you have a protocol to a separate CPU that lives with the peripherals?
17:58.17DocScrutinizer05modem is attached via USB
17:58.27DocScrutinizer05WKAN attached via SPI iirc
17:58.34DocScrutinizer05WLAN even
17:58.56DocScrutinizer05BT is UART and some other protocil
17:59.02keriorofl, people actually can't fathom a phone where shit is actually all shared under a closed hypervisor
17:59.08DocScrutinizer05it's always some 2 to 4 wires
17:59.37raxOk, so you have high speed serial to some other board.  4 wires is good.
17:59.42DocScrutinizer05no address bus involved
17:59.56raxSo the CPU board has no advanced peripherals attached to the memory bus at all then.  That's even better.
18:00.12DocScrutinizer05that's what I try to tell you, yes
18:00.22raxOk.  That makes things very easy then.
18:00.27DocScrutinizer05exactly
18:00.48raxI'm used to talking about high performance motherboards where doing that would negatively impact the performance of very high speed peripherals.
18:00.52raxIn this case, it shouldn't be necessary.
18:00.58DocScrutinizer05auditing is: 130 seconds look at http://neo900.org/stuff/werner/web/gta04b7v2-wip-2014092209.pdf
18:01.38kerioUSB can still push enough data for LTE, right?
18:01.40DocScrutinizer05highest bandwidth we need is with modem, thus we got USB2.0 there
18:01.48raxRight.
18:02.04DocScrutinizer05480mbps
18:02.17raxThat's fast enough to make hay.
18:02.23DocScrutinizer05well, actually WLAN might need even more
18:02.42DocScrutinizer05than modem
18:03.16raxAs you can imagine, I'm not one to trust anything that has a state machine, a CPU or DMA.
18:03.16rax:-)
18:03.17DocScrutinizer05but honestly, prolly OMAP3730 isn't fast enough for >>100MBps anyway
18:03.40kerioi want gigE :c
18:04.17raxDo you still want gigE if it gives someone a direct attack vector into your operating system?  :-)
18:04.25DocScrutinizer05it seems sustained write speed to eMMC is ~15MB/s
18:04.33keriorax: as long as the attack vector is fast!
18:05.06DocScrutinizer05read that is, write might drop after a short while
18:05.19kerioi doubt write will even start at that speed
18:06.05raxDo you guys have any estimates on how many N900 bodies are available in the wild?
18:06.09DocScrutinizer05I dunno off top of my head the max BW of SPI
18:06.16DocScrutinizer05err SDIO
18:06.31DocScrutinizer05that's been what WLAN has, I seem to recall
18:06.49DocScrutinizer05prolly in line with eMMC bandwidth
18:09.52DocScrutinizer05NB I don't say the CPU side drivers for interfaces like SDIO, USB etc are secure per se. They are possibly vulnerable to one-off and buffer-overflow attacks like any other protocol. However that's something that can't get possibly solved in hardware, it's responsibility of the driver developers and auditors
18:10.09raxYeah, that's a software problem.
18:10.34raxDoc - is the eMMC removable or hard-wired?
18:10.54DocScrutinizer05eMMC is hardwired. uSD is in tray
18:11.21rax(I'm spending way more than 130 seconds looking at the logic diagram. :-) )
18:11.28DocScrutinizer05we pondered to have a second uSD instead of eMMC but the interface is only 4bit instead of 8bit like eMMC has
18:11.57raxYeah, mem performance would suck.
18:11.57DocScrutinizer05so for sake of available bandwidth we want at least one eMMC attached
18:12.26raxLooking at this hurts my head because I'm more familiar with thinking about memory bandwidth in the 10's of Gbps.
18:12.51DocScrutinizer05that's not embedded though ;-)
18:13.32raxWhat's the maximum memory size of the uSD?
18:14.14DocScrutinizer05you asked about final sales price. Please refer to http://neo900.org/faq#cost
18:14.46DocScrutinizer05(max uSD size) I seem to have seen people using 64GB cards in N900 without problems
18:15.15DocScrutinizer0532GB are "guaranteed"
18:15.19raxOk.
18:15.23raxThat's enough.  :-)
18:17.01raxI really like what you guys are doing.  The landscape for phone technology is depressing.
18:17.19bencohso true
18:17.22DocScrutinizer05I think there should be some headroom in price to make some money from a secure OS, selling the combo at 1000+
18:17.31raxThe problem(s) are from top to bottom and you guys are working at the bottom.  I'm working at the top.  :-)
18:17.55raxI've had conversations with people and there is plenty of demand for a very high priced secure device.
18:18.05DocScrutinizer05(secure OS) actually this is a point on my strategic long term business plan
18:18.37raxHowever...you may not make many friends in high places.
18:18.46DocScrutinizer05;-P
18:18.59raxRemoving the hardware attack vectors goes a long way toward making linux viable to be truly secure.
18:19.01DocScrutinizer05not like I made many before in my long life ;-)
18:19.07rax:-)
18:19.36DocScrutinizer05(Removing the hardware attack vectors) that's the goal
18:20.07raxThe world is evolving to the point where there are larger and larger factions of buyers who will pay more and more for a secure hardware device.
18:20.15DocScrutinizer05make a device that's decent enough on hw level to allow a decent OS on top
18:20.24raxQuestion - not sure if you know the answer...(I know very little about the cell protocols.)
18:20.52DocScrutinizer05I know a bit, not much though
18:20.54raxWhen audio data is sent between handsets through the network, is the packet data massaged at all by the network or can it be packetized raw data?
18:21.29DocScrutinizer05usually gets mangled by audio-enhancement in exchange
18:21.38raxThat's what I suspected.
18:22.06rax"Audio enhancement" = "Break communications except in audio form"
18:22.16DocScrutinizer05yup, basically
18:22.29DocScrutinizer05unless you use CSD data connection
18:22.40DocScrutinizer05or modem-over-GSM
18:22.59DocScrutinizer05which wil net out at ~1000bits/s
18:23.07raxYeah.
18:23.14raxIt is clear what has been set up.  :-)
18:23.17DocScrutinizer05allegedly somebody made almost 2000 iirc
18:23.20raxSomething that is ... not clear.
18:23.46raxYeah, but as soon as you do (de)mod, you're stepping on all sorts of rules.
18:24.05DocScrutinizer05ask modem
18:24.10raxDon't want to get on the wrong side of laws.  :-)
18:24.13DocScrutinizer05user: modem
18:24.46DocScrutinizer05we had a discussion about all that a few weeks ago
18:25.09raxFCC rules state (I believe) that transmitting anything encrypted with the intention of obscuring the message is a no-no.
18:25.25raxThat's in the US, anyway.  I'm sure all major states treat this the same way.
18:25.38DocScrutinizer05and I just today finally closed the browser window of that kickstarter dongle thingie that encrypts voice over any arbitrary GSM connection
18:26.04DocScrutinizer05no, we don't have such noncense here
18:26.09raxAnyone doing that gets on a short list real quick.
18:26.23DocScrutinizer05afaik, at least
18:26.28raxYeah.
18:26.32DocScrutinizer05inly applies to HAM radio
18:26.53raxThe FCC encryption rules only apply to HAM?
18:26.56DocScrutinizer05encryption via any sort of phonecall is absolutely legal
18:27.07DocScrutinizer05the German rules do
18:27.16raxI like the German rules.  :-)
18:27.16DocScrutinizer05NFC FCC
18:27.31rax(Although I know almost 0.0% of what they are.)
18:27.36DocScrutinizer05France forbids encryption completely
18:27.47DocScrutinizer05or at least they did
18:28.34DocScrutinizer05Grm,an government sponsored an university research project to encrypt voice calls
18:28.34raxDo you guys have plans to make your own cases if you get this first phase of your business model functioning and profitable?
18:29.11DocScrutinizer05depends. When we hit 5digits order volume, we have to
18:29.33raxOh yeah, do you have any idea how many N900's exist to be bought for these?
18:29.36DocScrutinizer05for now we're struggling inmidst the 3 digit range
18:29.53raxI saw the latest info - you've got something like 350 preorders?
18:29.55DocScrutinizer05we need spare parts, no N900
18:30.08DocScrutinizer05the N900 option is for customers doing a DIY swap
18:30.17raxYou need your own bodies.  The problem is the startup cost of that, I imagine.
18:30.26DocScrutinizer05yes
18:30.48DocScrutinizer05it starts with palstic case, and keymat. then comes digitizer, then LCD
18:30.54raxHow much of the cost is built into the N900 cases?  Does that include the display?
18:31.36DocScrutinizer05we hope to get complete set of spare parts (incl LCD) for max 150EUR
18:32.32DocScrutinizer05still work in progress, pretty much. However we need to start sourcing those spare parts RSN
18:32.36raxIf this gets off the ground and you can get the word out to the high-end market, there is a market for this that isn't very price sensitive.
18:32.55DocScrutinizer05yeah, I hope for that a lot
18:33.33DocScrutinizer05major companies getting a 1000++ device for each of their 1500 managers, no utopy
18:34.09DocScrutinizer05even security sensitive NGOs
18:34.19DocScrutinizer05political parties
18:34.22raxYeah. I've had conversations with high-placed people in bank security that said "everyone I know would pay >$1000 for a device [like that]"
18:34.42DocScrutinizer05:nod:  I'm aware of all that
18:34.43raxEvery executive at any $10M+ company would pay >$1000 for it.
18:35.07raxThe trick is to sell to the people who cause the problem.  :-)
18:35.10DocScrutinizer05I do the hw, need somebody to take care about the software and infra side
18:35.28raxIs the display output a frame buffer or protocol?
18:35.35DocScrutinizer05fb
18:35.40raxWhat's the pixel or memory bandwidth to the display?
18:35.48raxOk...frame buffer.
18:35.55raxNice....any idea of the bandwidth possible?
18:36.05DocScrutinizer05errr, the display is 800*480, with dunno max 60fps or somesuch
18:36.08raxAlso, what's the display resolution? 800x480 or so
18:36.33raxThat's around 100M/s.
18:36.39raxNot bad.
18:36.49raxBetter than some terrible laptops.
18:36.56raxFor the display size, it is really good.
18:37.02DocScrutinizer05well, it's probably a 5/6/5 RGB
18:37.15raxOk, so 50M/s.
18:37.20DocScrutinizer05:nod:
18:37.23raxStill.  For that frame size, that's quite respectable.
18:37.34raxDo you know how much memory bandwidth the CPU has to the frame buffer?
18:37.43kerioDocScrutinizer05: we can just get estel_ to make some aluminium cases :P
18:37.48rax(I do graphics software and specialize in UI design.)
18:37.58DocScrutinizer05err, full RAM bus bw, I guess
18:38.20keriomh, wasn't there a HSV mode or something?
18:38.20DocScrutinizer05kerio: OHYEAAAH!
18:38.54kerioor something like that, the reason why the empty fullscreen movies have greenish letterboxing
18:39.04kerio*the fullscreen movies
18:39.11kerioor the empty screen in fullscreen movie playing
18:39.13raxCan it do RGB 24-bit?
18:39.19keriois so fucking good at explaining himself
18:39.20rax(Not that it is critical.)
18:39.30DocScrutinizer05rax: OMAP video design is ... intricate. There's a GPU, a DSP, a 4 or some independant framebuffers...
18:39.59raxOk.  Is all of the control of the GPU open?
18:40.00DocScrutinizer05rax: (24bit) iirc it could
18:40.20raxWhat do I type to do private messages?  (I assume that's what I'm getting...)
18:40.25rax(I not IRC gud.)
18:40.26DocScrutinizer05(GPU) only 2D, 3D drivers are closed source
18:40.36keriorax: this is just a channel message with a highlight
18:40.51keriousually most clients will do tab-completion of usernames
18:40.58DocScrutinizer05/query <nickname>
18:40.59kerioand add some sort of suffix, if at the beginning
18:41.24raxAh, ok.
18:41.50DocScrutinizer05rax: however see what keriosaid about highlight of lines with own nick in them
18:42.01kerioprivate conversations on IRC are... well
18:42.04kerionot frowned upon
18:42.07keriojust useless
18:42.19DocScrutinizer05but not exactly appreciated either
18:42.22kerioyeah
18:42.30raxOk.  Makes sense.  I figured I was probably being an idiot.
18:42.33kerioespecially if it's something that might also benefit someone else
18:44.09raxI haven't been on IRC in YEARS.
18:44.09DocScrutinizer05aluminium case is funny but useless. No RF will come out of it
18:44.15raxhahaha
18:44.29raxMake it out of beryllium.  That won't get you noticed or anything.
18:44.49DocScrutinizer05diamond
18:45.08DocScrutinizer05idly wonders if diamond is shielding RF
18:45.22raxSpeaking of cases, the conversations I had with people who were very interested in security were about a hypothetical device that could be large and clunky.  They aren't very concerned with that.
18:45.22DocScrutinizer05graphene does
18:45.29kerioDocScrutinizer05: hey, the iphone 6 is made of aluminium
18:45.31raxThey want security.  They're going to have a normal phone for normal stuff.
18:46.03DocScrutinizer05kerio: yeah, and iPhone5 had the "you hold it wrong" prank
18:46.11keriohey
18:46.15DocScrutinizer05since the antenna been integrated into the aluminium frame
18:46.21kerioif i hold the n900 wrong when i'm at my holiday house, i get no reception
18:46.26raxIf you really want to get paranoid about the attack vectors, put the CPU board in an isolated copper mesh - faraday it.
18:46.49DocScrutinizer05meh, cast in lead
18:47.06DocScrutinizer05or steel
18:47.14raxBut that only helps if the CPU itself can be triggered, which is highly unlikely.
18:47.23keriowhat happens if you have a conductor that goes through a faraday cage?
18:47.31raxIt is an antenna.
18:47.34DocScrutinizer05sorry! to recharge you need to flex away the small end
18:48.04keriolike
18:48.08kerioa laptop inside a faraday cage
18:48.15keriowith an ethernet cable that goes through the cable
18:48.21kerio*through the cage
18:48.37keriowould it actually work?
18:48.40DocScrutinizer05lightning will kill your PC
18:48.43kerioD:
18:48.50keriook what about a real answer now
18:48.59DocScrutinizer05that been a real answer
18:49.04DocScrutinizer05a cable is a leak#
18:49.11keriobut it's surrounded by plastic n shit
18:49.13raxYou can power the PC on the inside of the cage with a non-conducting mechanical power transfer of some sort, like a shaft with an alternator.
18:49.39raxDesigning isolation spaces isn't trivial.
18:49.39kerio(the ethernet cable)
18:49.45raxOptical.
18:50.05raxBut as soon as you have ethernet, you have an ethernet controller and that gets sticky. USB 2 and below are good because they're isolated from the CPU.
18:50.09DocScrutinizer05optical is a leak only for *very* short wave electromagnetic waves, yes
18:50.29raxLike, like leaks in the terabit range, which shouldn't be a problem.  :-)
18:51.25raxIf you really want to go hog-wild, you can build a nested faraday cage with an anechoic DMZ complete with high amplitude noise.  :-)
18:51.40raxOf course, then you get on short lists again.  :-)
18:51.45DocScrutinizer05but it seems we're getting slightly off topic now. We're not building the pentagon conference room
18:51.53raxTrue.
18:51.55raxI digress.
18:52.53raxHere's a legitimate question (maybe.)
18:53.07raxThe display controller...does it have a single display output and can it drive larger displays?
18:53.30raxI know this is out there...I'm thinking of whether or not it is practical (possible) to attach larger displays to the device.
18:53.39DocScrutinizer05rax: consider your mail as answered ;-)
19:03.11DocScrutinizer05sorry, afk, busy
19:03.17raxIt seems like there should be a high-priced market for what you're doing.
19:03.20raxkk
19:14.07DocScrutinizer05dong Neo900 for the linux hackers and maemo lovers now, basically exploiting my own work. Later on a special edition (possibly with different housing) for the security market, accompanied by a carefully tailored-to-meet-requirements OS and infra provided on secure servers
19:14.22DocScrutinizer05this will cost premium
19:14.41raxOk, that sounds good.
19:15.16DocScrutinizer05blackphone, a hoax
19:15.18DocScrutinizer05;-)
19:15.31raxBasically.
19:15.35DocScrutinizer05sandboxing? why? don't need that ;-P
19:15.53raxThey're just taking advantage of a market without deep solutions to the scariest problems.
19:17.07DocScrutinizer05yes
19:17.38DocScrutinizer05blackphone based on crappy standard android phone architecture, vulnerable just like any samsung
19:18.20raxYeah.  You may have noticed that I'm not very impressed with linux security.  It is too large to audit and based on programming practices that are too difficult to secure.
19:18.29DocScrutinizer05you can't base a secure OS on a crappy vulnerable hardware
19:18.34raxNot that I think linux is a bad thing - it isn't.  It is just a beast.
19:18.55raxExactly.  Without hardware, you have nothing.  Once you get that, you need secure software that you can trust.
19:19.11raxLinux is close.  I'm just uber strict because I can be in my situation.
19:20.04kerioopenbsd :3
19:20.41raxMultix.  :-)
19:38.45*** join/#neo900 xes (~xes@unaffiliated/xes)
19:49.48PaliDocScrutinizer05: windows machines are vulnerable too :-) http://imgur.com/cTD5b5D
19:50.34DocScrutinizer05those who are moderately concerned about vulnerabilities in their phone, but have no clear idea of the basic design and concepts, still hope for full audit of all baseband, so they could "certify" there's no pending threat from baseband sharing memory with APE CPU. This is a pipe dream, since a) will never happen since baseband doesn't get disclosed, and b) it's simply impossible to security-audit a software of that complexity, even when
19:50.35DocScrutinizer05you had access to full sourcecode
19:52.11DocScrutinizer05Pali: niiiice ;-P
19:59.13Palihttp://www.heise.de/security/news/foren/S-Was-ist-eigentlich-mit-Windows/forum-286128/msg-25869241/read/
20:01.42*** join/#neo900 DougLashz (~dal@gateway/tor-sasl/douglashz)
20:02.11DougLashzDo donations help donors reserve devices?
20:04.47raxYes.  DocScrutinizer - I have just donated 1000 EU.
20:05.37raxDoc - how do you feel about others producing compatible devices?
20:06.21raxDoug - yes, DocScrutinizer has told me that donating 100 EU or more puts you on the list of people who get credit toward a device.
20:07.00DougLashzCool, I would like to donate next month :)
20:07.06DougLashzas in Oct
20:08.20raxAs you can see, I've just done that.  :-)
20:08.29raxThis project appears to be well worth getting behind.
20:08.58DougLashzDefinitely looks like it
20:10.17raxOf course, as stated on the neo900.org site, there are no guarantees.
20:10.24raxBut hey...  :-)
20:10.37DougLashzyeah
20:10.49raxAt least someone is trying to make a securable computing device.
20:10.58DougLashzyeah :)
20:12.16raxThis thing has TV out?  I see Mic/TV (TS5A22366) with TVOUT...  (Darn...must read data sheet.)
20:25.52raxIt'll be interesting to see if the TVOUT can be a separate frame buffer than the LCD.
20:31.31keriorax: it can
20:31.38keriobuiltin function
20:31.48raxOh my.  That's evoking loving emotions.
20:31.51keriowhen playing a video, you get the video on the tv and the overlay n shit on the n900 screen
20:32.10keriowell, that's true for the n900
20:32.13raxThat makes it much easier to use the device as a desktop device when 'docked.'
20:32.16kerionot sure if the neo900 does the same
20:32.20kerioand ha ha ha hell no
20:32.26raxIs it the same family of CPU?
20:32.31kerio640x480 at best
20:32.37keriocomposite video
20:32.53raxI come from the days when that wasn't half bad.  :-)
20:36.41kerioit's definetely an improvement over an abacus
20:36.47keriobut it's not good enough for a modern desktop usage
20:37.07raxIt actually has it's uses for presentations and video conferencing.
20:37.27kerioyeah, i guess so
20:37.29raxI'd prefer HD 1080p, but hey, I'll sacrifice that for security.
20:37.32kerioimages and big text
20:37.44*** join/#neo900 norly (~norly@enpas.org)
20:41.49raxYes, or nicely antialiased smaller text.  :-)
20:44.13keriocomposite cable
20:44.26raxYeah.  Smeary.
20:44.26kerioit's less antialias and more blur
20:44.33kerioyeah not even blur
20:46.42DocScrutinizer05(<kerio> not sure if the neo900 does the same) of course it does exactly same. Plus VGA output still not being off the table, though this is only possible - due to SOC limitations - mutually exclusively to LCD display
20:47.11raxOk, so TVOUT is an option but disables the LCD?
20:47.20keriono no, tvout won't disable the lcd
20:47.34DocScrutinizer05(<rax> Is it the same family of CPU?) yes, N900 has OMAP3530, we use DM3730
20:47.52DocScrutinizer05VGA will
20:47.57raxI'm looking at the data sheet for the DM3730, but there's a lot to dig through to figure out what it is actually doing.
20:48.02raxAh, ok.
20:48.08keriostill worth it
20:48.12raxVGA isn't terrible - straight VGA or SVGA?
20:48.14DocScrutinizer05in case you use a resolution different to what LCD supports
20:48.24keriooh, otherwise you can do mirroring?
20:49.02DocScrutinizer05sure, why shouldn't have VGA the LCD signal same resolution and framerate? no problem
20:49.25kerioyour phrasing strange very
20:49.36kerioanyway, 800x480 is such a weird resolution
20:50.31DocScrutinizer05but the video interface we connect to LCD is the only high bandwidth video interface on OMAP3. So we either use it for 800*480 5/6/5, or we use it for whatever OMAP can do (1024*768?) and LCD goes mad or gets shut down during that
20:50.39raxIt is actually 16:10, so it isn't all that weird.  :-)
20:51.47DocScrutinizer05you still can use TVOUT and VGA on separate framebuffers ;-)
20:52.03raxIs there a good reason to use 5/6/5 if the device is capable of 24bpp or 32bpp?
20:52.14rax(Doc: Excellent re: separate frame buffers and VGA.)
20:52.32DocScrutinizer05I think the display is only capable of 5/6/5 but I'm not sure at all about it
20:52.38rax(Doc: Will the device have VGA outputs that aren't just the Mic/TV out?)
20:52.48raxAh, ok.  The display needs it.
20:55.12DocScrutinizer05we have a AV connector aka headset jack that provides common usual CVBS aka composite video output, either PAL or NTSC
20:55.37DocScrutinizer05we evaluate our options to place a micro-connector somewhere to offer VGA
20:55.48raxExcellent.  :_)
20:55.50rax:-)
20:55.59keriostylus slot!
20:56.21DocScrutinizer05The former is mandatory, to be feature complete with N900, the latter is an opportunity we evaluate, not finalized yet
20:57.02DocScrutinizer05might become HDMI or displayport or whatever, as well. Nothing set in stone yet
20:57.11DocScrutinizer05maybe we cannot do it at all
20:57.14raxYeah.
20:57.17DocScrutinizer05we'll see
20:57.30raxIt isn't critical at this point.  Future devices can have different displays and display output options.
20:58.11raxuHDMI would be SWEET.
20:59.48DocScrutinizer05yes, of course. Future devices may have OMAP5, or some completely new SOC (though I don't know of any valid alternative yet. Dang chip industry!)
21:00.08kerioDocScrutinizer05: quad-core i7
21:00.14DocScrutinizer05OMAP5 iirc has THREE vide interfaces, one of them being HDMI
21:00.44DocScrutinizer05OMAP3 has 2, one of them being composite video
21:01.23DocScrutinizer05rax: thanks for donation! :-)
21:01.28raxOk.  That makes more sense.
21:02.44DocScrutinizer05once Pyra is available, we easily can test our OS on that one, and if they work, we can base next phone on OMAP5
21:04.08DocScrutinizer05(sidenote: Pyra is kind of "sister project" to Neo900. They inherited quite some stuff from us, and we will benefit from GDC's experience with NEW OAMP SOC in Pyra
21:04.10DocScrutinizer05)
21:04.13raxThere should be a market for a slightly larger device that acts as both a secure communications device and a low-end notepad for presentations and videoconferencing.
21:04.35rax(And other things, like black-box VPN.)
21:04.56DocScrutinizer05that would be the next generation NeoX00 then
21:05.42DocScrutinizer05Pyra still is massively gaming console centric
21:06.05raxVery interesting.  Is there anything online about it?
21:06.18raxThat's a really good sister project for the enthusiasts to get behind.
21:06.20DocScrutinizer05I wouldn't want to use it as PDA or phone, even if it was able to establish phonecalls - which it most likely won't be
21:06.30DocScrutinizer05~pyra
21:06.36DocScrutinizer05hmmm
21:07.09raxYou can also tunnel point-to-point 'calls' through wifi.
21:07.29DocScrutinizer05http://www.dragonbox-pyra.com/
21:07.34raxOf course, I'm talking about applications where the participants really want to know they're secure.
21:07.41raxThey're willing to do 'non-normal' things to get it.
21:07.46raxYeah, I'm reading that site now.
21:07.57DocScrutinizer05~pyra is http://www.dragonbox-pyra.com/
21:07.57infobotokay, DocScrutinizer05
21:08.29keriorax: can't you just use skype for that? :P
21:08.44DocScrutinizer05http://www.dragonbox-pyra.com/specs.html
21:10.33DocScrutinizer05((inherited)) e.g. >>It can also do a fake multi-touch so you can use it for gestures like pinch-to-zoom or rotate.<<
21:11.26DocScrutinizer05if it been not obvious from my above blabbering: Pyra and Neo900 are both developed by Nik/GDC
21:11.42raxRight.
21:13.21DougLashzDocScrutinizer05: So does a 100 Euro donation earn me credit toward a possible phone? :)
21:13.30DougLashzDocSrutinizer05: Is there anything I need to know about donating before donating?
21:13.51DougLashzDocScrutinizer05: I read the donation stuff on the website but it seemed possibly out of date
21:14.09DocScrutinizer05DougLashz: yes, for a board. For the case Neo900 case parts we soon need to start true preordering with separate payment
21:15.05DocScrutinizer05of course we also highly appreciate "donations" of >>100EUR
21:15.38DocScrutinizer05but 100 already makes sure we will build a NeoN board with your name on it (unless a meteorite hits my house)
21:15.51raxThe device will cost more than 100 EU...  By giving more than 100 EU, you increase the odds that it will actually happen and also aren't losing anything in the long run (assuming they get produced.)
21:16.01DocScrutinizer05yes :-)
21:16.34DocScrutinizer05you actually get an amazing 2% interest per *month* on your donation amount
21:16.49raxSo you only lose on nominal inflation!  ;-)
21:17.19DocScrutinizer05is <2% recently in Germany, afaik
21:17.34DocScrutinizer05per anno
21:17.39DougLashzHaha okay cool :)
21:17.55raxAh, sorry - 2% MONTHLY return.  Big difference.
21:18.02DougLashzI will donate a little bit over 100 euros next time a client decides to get around to paying an invoice, which should/must be soon
21:18.09DocScrutinizer05we're not drowning in funds, so higher donations are highly welcome
21:18.32DougLashzyes, that's why I'll donate over 100 euros
21:18.45DougLashzjust not sure what I can do right now
21:18.53raxThat's why I donated 1000.  I'll likely end up with several of these.
21:18.55DougLashzI want to help :)
21:19.07DocScrutinizer05thanks! :-D
21:20.45DocScrutinizer05for a ballpark figure: recently ongoing prot_V2 development and production costs Neo900 UG roundabout 25kEUR
21:21.34DocScrutinizer05might get cheaper when things pan out unexpectedly fine
21:22.44DocScrutinizer05e.g. last week Nik reported that he didn't use up the quoted amount of working hours we assumed for schematics finalization, yet the schematics are almost done.
21:23.28DocScrutinizer05let's see what this week brings us
21:23.34rax:-)
21:23.51DocScrutinizer05I hope for publishing a draft of proto_V2 schematics
21:23.58raxDoc - you're the software guy, right?  Nik is the EE?
21:24.34DocScrutinizer05nah, I'm the product manager, genuine EE . Nik is running a hardware house, and doing EE and sw
21:24.51raxOk.
21:25.02DocScrutinizer05if anybody is our sw guy then that's clearly werner almesberger aka wpwrak
21:25.24raxOk.  Just curious.
21:25.26DocScrutinizer05certified linux kernel hacker
21:25.37DocScrutinizer05even has a wiki page ;-)
21:25.39raxI'm a software guy; always interested in knowing hardware guys because I can't do what you guys do.
21:26.29DocScrutinizer05https://en.wikipedia.org/wiki/Werner_Almesberger
21:29.31DocScrutinizer05basically we're all parts of the old OpenMoko team and affiliates
21:34.11raxRight.  Is the uSD a MicroSDXC interface?
21:34.33raxOr MicroSDHC?
21:36.17*** join/#neo900 kolp_ (~quassel@55d40159.access.ecotel.net)
22:15.26DocScrutinizer05HC, OMAP3 controller doesn't support XC
22:18.20raxkk
22:48.02OksanaWhat is xprot-ambient-temp ? :) Unless a meteorite hits your house :)
23:07.15freemangordonWTF is wrong with this guys ?!?
23:07.28Oksana?
23:07.42freemangordondoc knows :)
23:07.55freemangordonOksana: sorry, you don't want to know
23:08.50freemangordonoh, re your question...
23:09.09freemangordonI didn't dig into it, I can only guess
23:09.44freemangordonthat xprot is either lowering the speakers volume with temp getting high..
23:10.09freemangordonor there is a threshold
23:10.32freemangordonafter which speakers are turned off
23:11.01OksanaInteresting... Thank you!
23:11.32freemangordonOksana: any PA knowledge on your side?
23:11.48DocScrutinizer05hey freemangordon! :-)
23:12.02FIQhm
23:12.02freemangordonbecause I am a dumb guy when it comes to PA and alsa
23:12.12freemangordonDocScrutinizer05: hi!
23:12.22DocScrutinizer05freemangordon: I dunno what's wrong with those guys. Probably woody's mental disorder is congestious
23:12.52freemangordonI can;t belive that, woody was a very reasonable guy :(
23:12.57FIQso this guy in repair shop fixes lcd on n900 for me for $30 and said he might be able to fix the USB for a small additional cost (+$20)
23:13.03FIQthat would be convenient
23:13.09DocScrutinizer05yeah, until some 9orso months ago
23:13.30FIQgetting n900 fully working (besides BT) for $50 is nice
23:14.20freemangordonhonestly, I can't believe it, I still remember when we argued with "meego guys" that in 2 years there will be no meego, but maemo will be alive and kicking still :(
23:14.43FIQwell, and speakers, but I literally never used them
23:14.44freemangordonme and woody that is
23:14.47DocScrutinizer05freemangordon: I can help out a tiny bit with ALSA, nfc about PolypAudio
23:14.50FIQhence why I never bothered fixing them
23:14.51FIQOh well
23:15.06FIQlike, it took me a week
23:15.15FIQuntil I realized they were broken in first place
23:15.15freemangordonFIQ: take your chances and have your n900 repaired at no cost :)
23:15.16FIQlol
23:15.37FIQfreemangordon, what do you mean?
23:15.51freemangordonDocScrutinizer05: I know your opinion on PA, but as of now, we have no options
23:15.59DocScrutinizer05sure
23:16.24DocScrutinizer05I just say I have literally Not the Faintest Clue
23:16.54DocScrutinizer05all I faintly know of PA is the borked ALSA compatibility layer
23:16.55FIQfreemangordon, well
23:17.01freemangordonFIQ: the amount you shared seems as a low cost for a repair, thus " take your chances"
23:17.04FIQmaemo is a shell of its former glory
23:17.12FIQbut it didn't end up as bad as meego lol
23:17.17FIQwhich is literally dead
23:17.27FIQfreemangordon, ah, fair enough
23:17.37FIQwell to be honest I forgot about my speakers :P
23:17.42FIQwould have asked otherwise
23:17.55DocScrutinizer05FIQ: speakers are easy to swap
23:18.00FIQand I never cared about bt, never ever used it in my 4 years of owning it
23:18.08DocScrutinizer05FIQ: no soldering
23:18.27FIQyeah I heard it was trivial to fix
23:18.33FIQyou need replacement parts though
23:18.42DocScrutinizer05yeah, not that hard to get
23:18.47FIQok
23:18.53DocScrutinizer05seems they are used in other Nokia phones as well
23:19.08FIQhe didn't have the lcd part either but he had no issues buying a replacement part
23:19.17FIQso I guess it would be easy
23:19.19FIQfor him to fix
23:19.22FIQif I brought it up
23:19.26DocScrutinizer05glances at that bag with 4 spare speaker transducers
23:20.04DocScrutinizer05guess I ordered them at ownta back when
23:22.27DocScrutinizer05we are using similar ones but with a lower(?) Ohms-value on Neo900
23:22.45FIQok
23:24.01DocScrutinizer05a quick googling only shown me earpiece transducers ("speaker") that fit to N900 and several other Nokia phones. Seems the original speakers IHF are not that easy to source
23:24.54DocScrutinizer05freemangordon: you now guess it why I'm not on #maemo* channels anymore since weeks?
23:26.14DocScrutinizer05doesn't matter, we don't need the maemo.org servers and we don't need the HiFo or the MCeV or whatever
23:27.27DocScrutinizer05what we need is doing a proper backup of all repos, tmo, wiki, for the time when even wardare can't keep stuff running anymore thanks to too many fools rushing over it
23:27.42DocScrutinizer05warfare*
23:33.06freemangordonDocScrutinizer05: It shouldn't be like that, youknow
23:33.20DocScrutinizer05freemangordon: those guys act like rabid dogs because they know THEY messed it up, and the only way the see for fixing it is bashing council into ignoring own rules and surrender to their new brave world governed(SIC!) by a GA and a 3-members-BoD appointing that GA
23:34.45freemangordondon't believe that. I am sure they do it 'cause they think it will be better. and you (and it seems me) are stopping them :)
23:36.18DocScrutinizer05freemangordon: I know it shouldn't be like this. But when a 500 fighters with kalaschnikov raid a town of 60000 and those 60000 are all either in holiday or don't bother to even open their window and shout at those fighters, what could the administration do?
23:36.21freemangordonhowever, I won; argue with their highnesses anymore, lets see how it will turn out, I'll concider if i will continue my participation once it is clear what is going on
23:37.53DocScrutinizer05what we see is pretty much a takeover like in Donezk
23:38.39freemangordonit just makes no sense to me to waste time in conversations like that http://talk.maemo.org/showpost.php?p=1441058&postcount=60
23:38.53freemangordonshould I really care?
23:39.05DocScrutinizer05on the bright side there's little left over for those guys to ruin.
23:39.22DocScrutinizer05I gave up
23:40.25freemangordonwell, you are some kind rough on the edges, i thought that maybe i can start some productive talk. I must admit I was wrong
23:42.06DocScrutinizer05I became rough at the edges since i'm suffering this sort of fighting since roundabout 9 months now
23:42.34freemangordonanyway :)
23:42.46freemangordon(I meant my own experience)
23:42.56freemangordonyou know - thumb, SR, etc
23:44.00*** join/#neo900 wicket64 (~wicket@gateway/tor-sasl/wicket64)
23:44.11DocScrutinizer05I tried to build a bridge for them, in http://talk.maemo.org/showthread.php?p=1440932#post1440932 >>Generally nothing would be wrong with MCeV<<  -  they ignore it
23:45.30freemangordonwell, you are "the enemy", I don;t balieve anyone from the "right guys" will try to grok whatever you say, right now
23:45.30DocScrutinizer05(thumb, SR) I just tried to make clear we need proper evaluation for it. Never been against it, though you probably sometimes thought I was.
23:45.44DocScrutinizer05yes
23:45.56DocScrutinizer05futile
23:46.22freemangordonnaah, I know I am better than you, so i've just tried to convince you that you are wrong :P (thumb, sr)
23:46.39DocScrutinizer05:-)
23:47.03freemangordonthe point is that i've never taken it personally
23:47.09DocScrutinizer05exactly
23:47.24freemangordoneven the "you're idiot" from your side
23:47.32freemangordonthus I wonder...
23:47.37DocScrutinizer05sorry for that :-/
23:47.38xes:)
23:47.45DocScrutinizer05hi xes!!
23:48.08xeshi DocScrutinizer05 , hi freemangordon :)
23:48.12freemangordonfor sure I am not, and ther is no a random guy over the IRC who can judge it :D
23:48.17freemangordonhi xes!
23:48.34DocScrutinizer05how's life with our most kind sysop ever?
23:49.31xesi would suggest you a two weeks of fixes&updates over bash/apache/xen and you would avoid everything that would become "personal"
23:50.12freemangordonxes: wanna re nm-nav-provider? I will fix and upgrade whatever you tell me :)
23:50.26freemangordonin the meanwhile :P
23:50.48xesno thanks..
23:50.56freemangordonI knew it
23:51.44freemangordonoh, BTW, if i missed to announce: https://gitorious.org/community-ssu/nokiamaps-navigation-provider/source/5142e28be6e3c61ae962ca282c94890c33a3f7be:
23:52.12DocScrutinizer05xes: do you think we might provide a special edition RTL-SDR rady to plug in to Neo900 hackerbus?
23:52.22DocScrutinizer05ready*
23:52.39DocScrutinizer05or maybe s/might/should/
23:52.44freemangordonDocScrutinizer05: wait, isn't that specie USB?
23:52.55DocScrutinizer05yep, has USB
23:53.08DocScrutinizer05just no USB jack
23:53.15freemangordonwhy a "special edition" than?
23:53.40DocScrutinizer05one without USB jack and without housing, and maybe special antenna whip?
23:53.50freemangordonah, I see
23:53.54DocScrutinizer05s/jack/plug/
23:54.25xesDocScrutinizer05: since it's external it would be enough a stable usb port with enough current
23:54.33DocScrutinizer05I think we could buy a 25 of those dongles and manually rework them
23:55.40DocScrutinizer05xes: sure, that works 100%, no doubt. I thought about some "integrated" solution that wouldn't need to plug a lengthy dongle to the USB external port
23:55.55DocScrutinizer05raher have it internal under mugen cover
23:56.21*** join/#neo900 jonwil (~jonwil@27-33-80-219.tpgi.com.au)
23:56.52DocScrutinizer05((enough current)) the charger chip we currently plan to use can provide a bombing 1000mA VBUS
23:56.59DocScrutinizer05;-)
23:57.10xesintegrated inside would be very interesting, but you have to plan a proper shield and antenna path
23:57.21DocScrutinizer05aaah, good point
23:58.07DocScrutinizer05antenna I thought of actually having a 5cm whip with a small antenna connector at end of it
23:58.55DocScrutinizer05or have a pogopin contacting to a telescope antenna you slide into stylus-bay? ;-)
23:59.12xesprobably this one is the smallest available: http://goughlui.com/2013/07/07/adsb-range-testing-with-the-rtl2832ur820t-tuner/
23:59.59DocScrutinizer05ta!

Generated by irclog2html.pl Modified by Tim Riker to work with infobot.