DocScrutinizer05 changed the topic of #qi-hardware to: Copyleft hardware - http://qi-hardware.com | hardware hackers join here to discuss Ben NanoNote, atben / atusb 802.15.4 wireless, and other community driven hw projects | public logging at http://en.qi-hardware.com/irclogs and http://irclog.whitequark.org/qi-hardware
Supaplex has joined #qi-hardware
Jim45 has joined #qi-hardware
wej has quit [Ping timeout: 245 seconds]
wej has joined #qi-hardware
yang has quit [Ping timeout: 264 seconds]
yang has joined #qi-hardware
wej has quit [Ping timeout: 260 seconds]
Freemor has left #qi-hardware [#qi-hardware]
panda|x201 has joined #qi-hardware
wej has joined #qi-hardware
wej has quit [Ping timeout: 245 seconds]
wej has joined #qi-hardware
Markvilla has quit [Quit: ["Textual IRC Client: www.textualapp.com"]]
xiangfu has joined #qi-hardware
LunaVorax has quit [Ping timeout: 256 seconds]
arossdot1 has quit [Ping timeout: 240 seconds]
xiangfu has quit [Quit: leaving]
LunaVorax has joined #qi-hardware
cladamw has joined #qi-hardware
LunaVorax has quit [Ping timeout: 256 seconds]
Supaplex has quit [Ping timeout: 245 seconds]
Supaplex has joined #qi-hardware
Supaplex has quit [Ping timeout: 240 seconds]
Supaplex has joined #qi-hardware
cladamw has quit [Quit: Ex-Chat]
LunaVorax has joined #qi-hardware
Jim45 has quit [Ping timeout: 245 seconds]
yang has quit [Ping timeout: 240 seconds]
yang has joined #qi-hardware
panda|x201 has quit [Ping timeout: 276 seconds]
kilae has joined #qi-hardware
roh_ has joined #qi-hardware
kristian1aul has joined #qi-hardware
roh has quit [Ping timeout: 276 seconds]
kristianpaul has quit [Ping timeout: 276 seconds]
Alarm has joined #qi-hardware
lekernel has joined #qi-hardware
wej has quit [Ping timeout: 260 seconds]
FDCX has quit [Remote host closed the connection]
_whitelogger has joined #qi-hardware
_whitelogger has joined #qi-hardware
_whitelogger__ has joined #qi-hardware
_whitelogger__ has joined #qi-hardware
_whitelogger__ has joined #qi-hardware
_whitelogger___ has joined #qi-hardware
_whitelogger___ has joined #qi-hardware
_whitelogger___ has joined #qi-hardware
_whitelogger___ has joined #qi-hardware
_whitelogger__ has quit [Remote host closed the connection]
_whitelogger has quit [Remote host closed the connection]
_whitelogger___ has quit [Remote host closed the connection]
_whitelogger has joined #qi-hardware
_whitelogger has joined #qi-hardware
_whitelogger has quit [Remote host closed the connection]
yang has quit [Ping timeout: 264 seconds]
yang has joined #qi-hardware
wej has joined #qi-hardware
yang has quit [Ping timeout: 264 seconds]
yang has joined #qi-hardware
lekernel has quit [Ping timeout: 245 seconds]
qi-bot has quit [Ping timeout: 256 seconds]
lekernel has joined #qi-hardware
apelete has quit [Ping timeout: 260 seconds]
unclouded has quit [Ping timeout: 260 seconds]
<Fallenou> viric_: I was told NetBSD could run on 16 MB of SDRAM
<yang> Fallenou: maybe even less, as it can run on many vintage machines
yang has quit [Ping timeout: 246 seconds]
5EXAAS4SY has joined #qi-hardware
1JTAAAHYZ has joined #qi-hardware
1JTAAAHYZ has quit [Read error: Connection reset by peer]
arossdotme has joined #qi-hardware
methril has joined #qi-hardware
<mth> larsc: I think the pwm driver isn't ready to be shared between SoCs
<mth> since it is neither fully in control of the registers it uses (unlike the OHCI driver)...
<mth> ...nor is it using fully generalized interfaces (like GPIO)
<mth> pwm is basically gpio + timer, and the timer code is SoC specific at the moment
<mth> although I don't think it has to be, since the registers are the same
<mth> (a few regs and bits within regs were added in the 4770, but all the 4740 regs and bits are still there)
Guest655 has quit [Quit: Changing server...]
<mth> since the timers are used for non-pwm use cases as well, putting the pwm driver in full control over the registers is not an option
Guest67718 has joined #qi-hardware
Alarm_ has joined #qi-hardware
Alarm has quit [Ping timeout: 256 seconds]
Alarm_ is now known as Alarm
<viric_> Fallenou: I thought of netbsd... but I'm not sure how to deploy to the device easily :)
<Fallenou> is there a NetBSD port of your SoC ?
<viric_> time to check
<viric_> but I've 4MB of flash
<viric_> it's a broadcom small mips
<viric_> system type : bcm63xx/96348GW-11 (0x6348/0x00B0)
<viric_> cpu model : Broadcom BCM6348 V0.7
<viric_> I can't find anything similar in http://netbsd.org/ports/
<Fallenou> hum that's a MIPS32 cpu
<Fallenou> the only thing that matches "broadcom" and "mips" in netbsd ports is : sbmips: http://www.netbsd.org/ports/sbmips/
<Fallenou> but it's not BCM6348 ...
<Fallenou> so I guess to get NetBSD running on your board you need some porting effort ^^
<wpwrak> mth: put the timer code in charge of allocating timers and make pwm depend on it ?
kilae has quit [Quit: ChatZilla 0.9.90 [Firefox 21.0/20130511120803]]
<viric_> Fallenou: complicated.
wej has quit [Ping timeout: 260 seconds]
<Fallenou> viric_: yes I suppose :/
<viric_> I'll stay with an openwrt backfire.
wej has joined #qi-hardware
<mth> wpwrak: that is already the case, but the timer code is under arch/mips/jz4740 at the moment
<mth> ...which won't get compiled when building for jz4770 only
<mth> and having functions with the same name for both SoCs works when compiling for one at a time, but afaik larsc's long term goal is to be able to build one kernel that supports both
<larsc> mth: yes the whole timer code should probably be shared
<larsc> in my opinion the jz4770 should also go into arch/mips/mach-jz4740
xiangfu has joined #qi-hardware
Alarm_ has joined #qi-hardware
Alarm has quit [Ping timeout: 264 seconds]
Alarm_ is now known as Alarm
Freemor has joined #qi-hardware
jow_lapt1p is now known as jow_laptop
LunaVorax has quit [Ping timeout: 256 seconds]
LunaVorax_ has joined #qi-hardware
panda|x201 has joined #qi-hardware
LunaVorax_ has quit [Ping timeout: 256 seconds]
wej has quit [Ping timeout: 248 seconds]
wej has joined #qi-hardware
<wpwrak> sigh. projects is down AGAIN :-(
<larsc> the hoster got hacked pretty badly, maybe that's the cause
<wpwrak> hmm yes, i read about that. well, let's see.
lekernel has quit [Quit: Leaving]
<viric_> what hoster?
<wpwrak> hetzner
<viric_> I don't know it.
<viric_> I saw the word proftpd
<viric_> isn't proftpd the 1st entry point of most hackers since 20 years ago? :)
<viric_> I remember installing some linux 15 years ago, and it warning me about the security of proftpd
<wpwrak> seems they used an exploit in nagios, then manipulated other services, potentially including proftpd
<wpwrak> and yes, ftp security has been sort of a running joke for as long as i can remember
<wpwrak> i was just about to say that a reverse filegive would be nice to help more people avoid ftp. then i saw that you already implemented that :)
<viric_> :)
<wpwrak> now all you need to do is port it to a language people actually use :)
<viric_> people don't have to see the language
<wpwrak> well, they still need to put it on their system before they can use the critter, no ?
<viric_> well, you need the compiler if you want to compile it
<viric_> but there are packagers around the world doing that task for the most.
<viric_> and for windows I give a precompiled binary.
<wpwrak> when i click on "download", all i see it a tar.gz. and apt-get doesn't know about filegive. so ...
<wpwrak> "The site's security certificate is not trusted!" heh :)
<viric_> complain to your distribution managers
<viric_> ;)
<viric_> did anyone package go? if so, it should be fairly easy to build.
<viric_> For me go means 1) easy to type, 2) portable enough and 3) easy to audi
<viric_> t
<wpwrak> there's a "GNU Go" in ubuntu
<viric_> that may be the gccgo
<wpwrak> yup
<viric_> it's a bit harder to build with gccgo. I don't provide a makefile for it.
<viric_> complain one more time to your distribution managers
<viric_> :)
Alarm has quit [Ping timeout: 264 seconds]
<wpwrak> you provide no makefile at all ...
<wpwrak> or a README for that matter
<wpwrak> or any other sort of build instructions for an environment people commonly use
<viric_> doc/home.wiki (so, the web page) has the build instructions
<viric_> at the end
<viric_> I could provide something more gnuish, or a README. I will. :)
<wpwrak> do you intentionally try to keep it out of people's hands ? :)
<viric_> yes ;)
<viric_> I added a README
<wpwrak> is http://freecode.com/projects/filegive the project home ? i don't see a link to the repository, only the tarball download
<viric_> the home page
<viric_> it's the vcs
<viric_> see Timeline links, etc.
<wpwrak> the home page is the vcs ?
<wpwrak> you mean you don't use a vcs but maintain some meta-information manually (and with tarball snapshots)
<wpwrak> basically they way things were done in the 1990es
<wpwrak> well, "often done". cvs and friends predate that, of course
xiangfu has quit [Remote host closed the connection]
LunaVorax has joined #qi-hardware
5EXAAS4SY is now known as FDCX
damo22 has joined #qi-hardware
<roh_> nat-pmp? upnp? bwahahahha
roh_ is now known as roh
<viric_> wpwrak: I use fossil
<viric_> wpwrak: which, by name, sounds a pretty old thing :)
* roh uses xmpp and things like rsync, ssh and ncp for transferring files
<roh> wpwrak: you really aren't in luck with project hosting ;)
<viric_> what is ncp?
<viric_> dankon
<roh> brutal fast. no crypto. for crypto.. combinations on the rest
<viric_> ah, multicast.
<roh> ncp is nice for stuff where one needs bulk data in the lan and is lazy
<viric_> I've been these days trying to understand multicast. :)
<viric_> (through routers, mainly)
<roh> like trowing files into the air, and 'somebody' catching it
<viric_> can multicast traverse internet at all?
<roh> sure. if you want
<viric_> without tunneling?
<roh> in reality, people dont do.
<viric_> May I subscribe to any multicast group and see if someone is sending?
<roh> but it does unicast as well.. so if you can connect to sth. you can use it (need to add an ip tho)
<roh> its a specific one i think, but yes.
<viric_> understood about ncp
<viric_> I mean about multicast in general
<viric_> Is anyone using multicast through internet?
<roh> most carriers filter or do not route it
<viric_> do any route it?
LunaVorax has quit [Ping timeout: 256 seconds]
<viric_> I'd like to know some certain data about multicast. All I get from websites is rumours :)
<wpwrak> roh: (hosting) well, all the new stuff goes to github ...
<roh> dunno.. need to ask. i dont know any consumer carriers which 'just do it' .. i think some of the tv stuff on vdsl is done by mcast.. but thats limited again
<roh> wpwrak: doesnt solve it either. got hacked /downed before.
<wpwrak> roh: well, i don't mind if it's down sometimes. i'm used to that sort of thing. what's troublesome is when it's down most of the time.
<roh> wpwrak: from my pov its just a repo-mirror. development needs more that such kids toys
<wpwrak> roh: well, a nicely working git is a good start. i think you can also have a wiki there, no ? that leaves mailing list, IRC, and generic webspace (files)
<roh> wpwrak: there are 'some' tools.,.. but they re much too basic for serious work from my pov.
<viric_> roh: is that through internet at all?
<roh> and its so slow
<viric_> (tv mcast)
<roh> viric_: depends on what internet is for you? its all ip. so yes.
<viric_> wpwrak: you surrender?
<roh> but its the carriers managed mcast system. there is no worldwide mcast i know of
<viric_> roh: internet is where the public ips are interchanging packets. :)
<roh> worldwide as in 'just routed'
<wpwrak> roh: so you'd rather pick sourceforge, which has pretty much everything you could wish for, in more or less usable shape ?
<viric_> roh: ok, I meant this worldwide thing.
<wpwrak> i've heard even yahoo has many goodies :)
<roh> viric_: dream on. its a web of capitaliest crazies. there is no neutrality for packets
<wpwrak> (mcast) you can always set up tunnels ...
<roh> wpwrak: no. i host services if i want to have them.
<viric_> roh: I didn't talk about neutrality
<viric_> wpwrak: yes, I guess it's what everyone does. tunneling
<roh> wpwrak: no money, no working services.
<wpwrak> roh: i host my own webspace. that's enough admin work for my taste :)
<roh> wpwrak: on all those who do not take money i am not the user. i am the product.
<viric_> roh: cheap or expensive?
<viric_> :)
<roh> most people do not realize that. i do. so i dont use services where i have better replacements myself
<roh> viric_: ?
<viric_> 'the product'
<wpwrak> github doesn't seem too intrusive. they basically want people to get used to their platform, then switch to paid service for things that are non-open
<wpwrak> sounds like a fair enough deal to me
<roh> viric_: doesnt matter if cheap or expensive. you dont pay, you become the product.
<roh> no warranties for anything.
<viric_> roh: I meant if you, the product, are cheap or expensive ;)
<roh> viric_: i meant that in general. if you use github YOU are THEIR product.
<roh> they sell you to their investors.
<viric_> guess your price
<viric_> :)
<roh> in what way differs.
<viric_> I'm just joking
<viric_> I end up doing that too, hosting all I can
<roh> viric_: its not about amount. i know these bullshit-calculations from oloh... duuuh.. we're not 14 anymore.
<viric_> But I don't do it the hard way (setting up ruby on rails or anything fancy). I keep my feet on ground :)
<roh> hosting isnt the issue at all. or maintainting shit. its people realizing that there isnt anything for free.
<viric_> roh: you could play being more clever than they
<roh> it can be accounted flat. but thats as good as it gets.
<viric_> tricking.
<viric_> some people like that.
<viric_> I'm lazy for that.
<viric_> roh: what do you think of mega?
<roh> just another bullshitter
<viric_> haha
<viric_> What's bad there? not good enough ciphering?
<roh> well.. i am not against things like github. but they are too big, and thus primary target.
<roh> viric_: their service isnt anything i have any use for at all
<viric_> cooling down...
<viric_> roh: hm similar happens to me. But for some people, they may find it quite useful
<roh> github has a ok service. but well.. its not free at all.
<viric_> it sounds like an easy place where to have at least one backup copy
<roh> if you are a company.. either pay somebody for hosting/maintaining shit or you pay github
<roh> viric_: bwahahaha
<viric_> bwaaaahahaha :D
<roh> backups are something on a physical disk you have in your hand.
<roh> which you lock in a safe. not bullshit in some cloud.
<roh> raid is not backup also
<viric_> I feel like I should do the turing test with you
<viric_> ;)
<roh> you should believe in less marketing.
<viric_> some play that game well.
<roh> everybody cooks with water. and even some of the most visible out there have dirty water in their pots
<wpwrak> there are very few guarantees in this world :)
<viric_> the rules are set, you can tricky play them
<roh> marketing always lies
<roh> well.. i gotta run.. bbl
<viric_> the backup being *only* something on a physical disk you have in your hand.... sounds like it's not a very good backup :)
<wpwrak> well, you can pay someone to pay you something if for some reason they fail to provide the service, but ...
<wpwrak> physical disks often fail to work after spending too much time in storage
<viric_> For me, self-sufficency is the option giving me most tranquility
<viric_> But of course, that means giving up on features. :)
<wpwrak> the best approach would probably be to find half a dozen non-cost file hosters, replicate the data to all of them, and move it around when one or some of them disappear
<viric_> I built 'meogpg', did you see?
<wpwrak> in the end, it's risk management, not roh's idealistic zero-risk situation
<viric_> that's related to my backups and so...
<viric_> wpwrak: eys
<viric_> yes
<viric_> I'm not using it yet, but I wanted to use that giving data to a group of friends
<wpwrak> e.g., if he goes bankrupt, his gadgets could simply be seized. then he has nothing. meanwhile, they can strip me naked and then shoot me, but my github projects will still live on
<viric_> yes, not a bad management.
<viric_> wpwrak: I plan to use the meogpg against the 'shoot me' case :)
<wpwrak> (meopgp) hmm, does that work ?
<viric_> meogpg
<viric_> I didn't have the oportunity to test it *on the field* ;)
<wpwrak> name=`sed 1q </dev/random` # :)
<viric_> multiple eyepairs only - gnupg
<wpwrak> i mean the xor sets. so you basically generate X random numbers and one that's xorsum(key, random numbers)
<viric_> wpwrak: there is a "How does it work" h2
<viric_> what is xorsum?
<wpwrak> yes, but i wonder if what's written there is correct. intuitively, it seems wrong
<wpwrak> xorsum(a, b, c) = a XOR b XOR c (i just made the term up)
<viric_> ah ok
<viric_> key = a xor b xor c xor d ....
<viric_> a, b, c, d ... random numbers long as the key
<wpwrak> except for the last one
<viric_> what last one?
<damo22> xorsum(...) = foldl(xor,...) ?
<wpwrak> well, depend on whether you pick the key or whether it's the result of your random numbers
<wpwrak> if the key is given, one of the numbers must be non-random
<viric_> *the key* is the result of xoring the random numbers
<viric_> no key given
<wpwrak> so you generate the key. fine. then it's basically
<wpwrak> do { R[] = random_numbers(); key = xor(R); } while (key_not_strong_enough(key));
<viric_> :) I really have problems writing public explanations on websites. hehe.
<wpwrak> depending on the crypto algorithm, key_not_strong_enough() may be { return 0; }
<viric_> what would make the key 'not strong enough'?
<viric_> well, it's a gpg symmetric key
<wpwrak> well, some algorithms have weak keys. certain bit patterns that are "easy" to crack
<viric_> you configure the symmetric ciphering for gnupg as you wish.
<wpwrak> now ... the question is whether you can distribute the random numbers such that
<wpwrak> a) < N people never have all the numbers
<wpwrak> b) >= N people always have all the numbers
<viric_> that's the point of the program
<viric_> (about combinatorics)
<viric_> I didn't use much else.
<wpwrak> that's the theoretical basis i was looking for. thanks ! :) reading ...
<viric_> (it's linked at the bottom of the page)
<wpwrak> yeah, now that i scanned the page before complaining that you should link to the paper, i saw the link :)
<viric_> you don't read in diagonal, you read in vertical
<viric_> ;)
<viric_> wpwrak: if you need more complains to be satisfied, you could tell me that there is no README
<viric_> haha
<viric_> roh: following that ncp thing I found opentracker. nice. :) http://erdgeist.org/arts/software/opentracker/
<wpwrak> hmm yes, the algorithm seems plausible. that's a nice one. i though you'd need something like the redundant coding they consider(ed?) for multicast for this.
<viric_> hm I didn't understand the last sentence
<viric_> ah, well, I could send a copy to every friend, or publish it in my website with a link, if I trust the crypto. :)
<viric_> "Download all my secrets"
<roh> viric_: ah.. didnt know it?
<viric_> no
<viric_> due to not knowing that, I've only used dht bittorrent. :)
<wpwrak> (coding) like the FEC algorithm described in II.A of ftp://ftp.cis.upenn.edu/pub/securitylab/fec/pblr.pdf
<viric_> well, gpg has error detection
<viric_> but for correction.. that can be spread everywhere and it's fine.
<viric_> I mean multiple copies give FEC. :)
<viric_> I tried to make all simple.
<wpwrak> that FEC algorithm is about m-out-of-n receptions to reconstruct the original data. more complicated than yours (well, stamfest's) though.
<viric_> ah ok.
<viric_> yes, stampfest idea was very simple
<viric_> I was impressed how could he make it so complicated (new binary file format, PKCS#*, ...)
<viric_> both a console and a gui program with a common c++ lib...
<viric_> roh: what do you think about flattr?
<wpwrak> (making it complicated) yeah ;-)
<roh> viric_: nice experiment. too bad it uses centralized currency
<viric_> well, I don't like the bitcoin approach of burning as many Watts as possible either.
<viric_> Specially without demurrage and all that.
<roh> thats one thing. it dont like being trackable without reason
<roh> so i really like cash
<viric_> well, yes. But if I've to expect to get money in cash for what I publish...
<viric_> It's very hard to receive some cents with flattr, imagine with cash :)
<roh> i rather see people me paying cash for services than via flattr
<viric_> right, that's for services.
<viric_> I agree.
<roh> usualy i do bank transfers for big amounts/stuff which has paperwork anyhow, the rest is cash
<viric_> nice :)
<damo22> selling services for cash? sounds a bit seedy
<roh> damo22: i write bills
<damo22> :P
Supaplex has quit [Ping timeout: 255 seconds]
Supaplex has joined #qi-hardware
arossdotme has quit [Ping timeout: 240 seconds]
lekernel has joined #qi-hardware
LunaVorax has joined #qi-hardware
Supaplex has quit [Ping timeout: 240 seconds]
Supaplex has joined #qi-hardware
arossdot1 has joined #qi-hardware
wej has quit [Ping timeout: 245 seconds]
wej has joined #qi-hardware
rz2k has joined #qi-hardware
Supaplex has quit [Ping timeout: 252 seconds]
Supaplex has joined #qi-hardware
rz2k has quit []
lekernel has quit [Quit: Leaving]
newcup has quit [Ping timeout: 245 seconds]
wej_ has joined #qi-hardware
zear has quit [Ping timeout: 240 seconds]
zear has joined #qi-hardware
wej has quit [Ping timeout: 245 seconds]
Guest67718 is now known as Coyo
Coyo has quit [Changing host]
Coyo has joined #qi-hardware
arossdot1 has quit [Ping timeout: 264 seconds]
emeb has joined #qi-hardware