nicksydney | wpwrak: another day another WiFi https://www.indiegogo.com/projects/xwifi-a-fingertip-sized-wifi-module-open-source :) | 00:00 |
---|---|---|
DocScrutinizer05 | duh! 19 "free" modules available still (1$ + facebook-like) | 00:18 |
DocScrutinizer05 | too bad, I got no facebook account ;-P | 00:18 |
whitequark | I don't get one thing though, is | 00:20 |
whitequark | er | 00:20 |
whitequark | it doesn't seem to have any kind of security measures | 00:20 |
whitequark | so anyone on your network could do whatever they want | 00:20 |
whitequark | asked them | 00:24 |
DocScrutinizer05 | I think you need to set the password via USB or whatever? | 00:27 |
DocScrutinizer05 | or it has hardcoded password ;-P | 00:27 |
whitequark | if it's just password, then it's shit | 00:28 |
whitequark | anyone with a sniffer on the same network figures it out in a minute | 00:28 |
DocScrutinizer05 | well, I mean that WPA thing | 00:28 |
whitequark | no difference if WPA or dedicated password | 00:29 |
whitequark | suppose you want to control it remotely -- you're screwed | 00:29 |
DocScrutinizer05 | sorry, you lost me | 00:29 |
whitequark | well | 00:29 |
whitequark | if the protocol itself doesn't have any kind of request signing and/or encryption, you can't just port forward it and control it remotely | 00:29 |
whitequark | exposing it to the internet | 00:29 |
whitequark | if the protocol does and the firmware is not written by morons (it probably is, but another story), you can | 00:30 |
DocScrutinizer05 | err, I don't think this is a regular WiFi adapter | 00:30 |
whitequark | what do you mean? | 00:30 |
DocScrutinizer05 | InternetOfThings, it's meant for remote control, not general purpose internet data transfer | 00:31 |
whitequark | sure. so? | 00:31 |
DocScrutinizer05 | So I guess it has a way to set WPA-passphrase and that's it | 00:31 |
whitequark | well, and that's dumb. | 00:31 |
DocScrutinizer05 | WPA2 maybe even | 00:31 |
DocScrutinizer05 | ? | 00:32 |
whitequark | adding request signing with sha256 is like an hour of work and it already makes requests unforgeable | 00:32 |
whitequark | encrypting them is a bit harder, but still can be done | 00:32 |
DocScrutinizer05 | WPA *is* encryption, no? | 00:32 |
whitequark | WPA only matters locally | 00:32 |
whitequark | and also | 00:33 |
whitequark | if the adversary is in your network, you're screwed | 00:33 |
DocScrutinizer05 | sorry, I don't get it | 00:33 |
whitequark | if you forward the traffic to this WiFi thing, you're screwed too | 00:33 |
whitequark | the only case where you're protected by WPA is: 1) the device only EVER communicates inside the network and there is no way to address it from outside 2) ALL of the hosts on the network are trusted | 00:34 |
DocScrutinizer05 | ooh, you want it to have encryption on higher level, not on-the-air | 00:34 |
whitequark | yes. | 00:34 |
DocScrutinizer05 | so you could hook up the thing to arbitrary public hotspots | 00:34 |
whitequark | not necessarily | 00:34 |
whitequark | so I could just say to my home router "expose this IP:port to web" | 00:34 |
whitequark | and still be safe | 00:34 |
DocScrutinizer05 | hmm | 00:34 |
whitequark | this will be a very common use case | 00:35 |
nicksydney | i think probably in terms of security it's not that strong | 00:35 |
whitequark | nicksydney: my point is that adding request signing is *trivial* | 00:35 |
whitequark | adding encryption is less trivial because you need to use a proper mode | 00:35 |
whitequark | i.e. authenticated encryption | 00:35 |
whitequark | AES-GCM, or xsalsa20+poly1305 | 00:36 |
DocScrutinizer05 | you probbaly should implement such stuff on your firewall/router, or on a proxy resp concentrator in your network | 00:36 |
whitequark | 1) that's too much to ask of most users of the device | 00:36 |
whitequark | 2) if your network is insecure, and it is, it doesn't help | 00:37 |
DocScrutinizer05 | most users don't want to control the thing from remote via public internet | 00:37 |
whitequark | of course they do | 00:37 |
whitequark | this is what already happens | 00:37 |
whitequark | source: several full internet scans, published recently | 00:37 |
DocScrutinizer05 | hmm, silly | 00:37 |
whitequark | see point 2 anyway | 00:38 |
whitequark | if your passphrase is "iloveyou" (it is), WPA2 is as good as open | 00:38 |
DocScrutinizer05 | my home automation works different: I have a CCU-1 embedded linux device that has a web frontend on intra(inter)net | 00:38 |
DocScrutinizer05 | this CCU-1 controls all the gadgets | 00:39 |
DocScrutinizer05 | well, for me it's simple since the gadgets all are 866MHz, not any WiFi and no TCP-IP either | 00:40 |
whitequark | sure, I'm just saying that if you have real-world people who are not huge nerds and just want their problem solved | 00:40 |
whitequark | this is how the device should work. | 00:40 |
DocScrutinizer05 | I can see how users buy such WiFi home automation gadgets since they want to communicate to them directly with their iPhone | 00:41 |
DocScrutinizer05 | then exposing them to the public net is a logical next step, though pretty silly | 00:41 |
whitequark | yes yes | 00:41 |
whitequark | exactly what I mean | 00:41 |
whitequark | it may be silly, but you can at least make it possible to have it secure | 00:42 |
DocScrutinizer05 | sure | 00:42 |
whitequark | (still have to not set the password to 123456! but oh well) | 00:42 |
DocScrutinizer05 | but probably those users don't even care ;-P | 00:42 |
DocScrutinizer05 | "duh! why should I worry when some idiot thinks it's funny to switch my living room light on and off?" | 00:43 |
DocScrutinizer05 | "such idiot cannot exist" | 00:43 |
DocScrutinizer05 | ;-) | 00:43 |
whitequark | you should have seen the recent VNC scan of entire web | 00:43 |
DocScrutinizer05 | I seen it | 00:44 |
whitequark | oh, yeah, it's amusing | 00:44 |
DocScrutinizer05 | well, maybe not the most recent | 00:44 |
whitequark | a lot of german municipal systems exposed | 00:44 |
whitequark | sewer control, power stations, etc | 00:44 |
DocScrutinizer05 | yeah, THAT is stupid | 00:44 |
whitequark | the things I seen there are unbelievable | 00:45 |
DocScrutinizer05 | *nod* | 00:45 |
DocScrutinizer05 | maybe I should scan my town's "IP-range" to find a few funny opportunities to save some money or do entertaining stuff ;-) | 00:46 |
mth | I guy I knew in school thought it was funny to cycle through the neighbourhood, pointing a TV remote at random living rooms and see if their TV responded | 00:46 |
mth | since a lot of them used RC5, it worked pretty well | 00:47 |
whitequark | hehehe | 00:47 |
mth | so there is always someone who will prank you like that | 00:47 |
DocScrutinizer05 | yeah, we already improved that prank by using a 15W IR flood light | 00:47 |
mth | even more over the internet, where the chance of getting caught is a lot smaller | 00:47 |
whitequark | DocScrutinizer05: HAHAHA | 00:48 |
whitequark | that's great | 00:48 |
DocScrutinizer05 | IR laser is also useful, for a slightly different "attack scheme" | 00:49 |
whitequark | you're evil | 00:49 |
whitequark | :] | 00:49 |
DocScrutinizer05 | I know ;-) | 00:50 |
ray__ | hi there can i use librewrt with tp-mr3420? wouter [20:10] <ray__> want sure new to this | 01:14 |
whitequark | DocScrutinizer05: any chance you know a good guide for building class E amplifiers? | 01:23 |
whitequark | a book maybe | 01:23 |
DocScrutinizer05 | class E? | 01:23 |
whitequark | class E. | 01:23 |
DocScrutinizer05 | I admit I don't even know class E | 01:23 |
whitequark | keyed amplifier with LC matching network | 01:23 |
ray__ | there can i use librewrt with tp-mr3420? wouter [20:10] <ray__> want sure new to this | 01:24 |
DocScrutinizer05 | o.O | 01:24 |
whitequark | I've been told it's a good idea to use a class E one for my PSU | 01:24 |
ray__ | whats oo | 01:24 |
whitequark | less power dissipation, simpler | 01:24 |
ray__ | white did you see my q | 01:24 |
DocScrutinizer05 | we see your q, we don't know an answer when we don't answer | 01:25 |
DocScrutinizer05 | whitequark: that sounds like class-D to me | 01:25 |
ray__ | k | 01:25 |
whitequark | class-D is PWM, class-E is halfsine though | 01:26 |
whitequark | and I think there are other differences | 01:26 |
whitequark | https://en.wikipedia.org/wiki/Amplifier#Class_E has a good overview | 01:27 |
whitequark | The class-E/F amplifier is a highly efficient switching power amplifier, typically used at such high frequencies that the switching time becomes comparable to the duty time. As said in the class-D amplifier, the transistor is connected via a serial LC circuit to the load, and connected via a large L (inductor) to the supply voltage. | 01:27 |
whitequark | indeed, this is exactly my case | 01:27 |
ray__ | what happens if change subnet? | 01:27 |
DocScrutinizer05 | whitequark: yepo, I see | 01:27 |
ray__ | from the default | 01:27 |
whitequark | hm, maybe my Horowitz and Hill has it | 01:28 |
ray__ | what happens when i change default subnetmas? | 01:28 |
whitequark | no, doesn't cover PAs at all | 01:29 |
ray__ | white did you see my q about subnetmas | 01:31 |
ray__ | ? | 01:31 |
whitequark | ray__: I have no idea what are you asking at all | 01:31 |
ray__ | white sorry im austic | 01:32 |
whitequark | please read http://www.catb.org/esr/faqs/smart-questions.html | 01:32 |
ray__ | white what if i change from befault subnet mas to sothing eles what will it do? | 01:32 |
DocScrutinizer05 | aaah, you're thinking about your 100W 12kV thing | 01:32 |
whitequark | DocScrutinizer05: sure | 01:33 |
whitequark | oh, found http://people.physics.anu.edu.au/~dxt103/class-e/ | 01:34 |
ray__ | white did you understand my q that time | 01:34 |
whitequark | no | 01:35 |
ray__ | white lets try tyhis one more time | 01:36 |
DocScrutinizer05 | whitequark: there you are :-) | 01:36 |
DocScrutinizer05 | hail anu.edu.au | 01:36 |
ray__ | white what happnes if change the default subnet mask? | 01:36 |
DocScrutinizer05 | ray__: what's your topic? | 01:37 |
ray__ | want to knwo what dose changeing the default subnet mask dose? | 01:37 |
DocScrutinizer05 | nothing good | 01:37 |
ray__ | im just woundering | 01:37 |
DocScrutinizer05 | I'm pretty sure you're not supposed to change it | 01:38 |
ray__ | mine tell me i can DocScrutinizer05 | 01:40 |
DocScrutinizer05 | hehe! >>the proof of the pudding is in the eating<< | 01:53 |
DocScrutinizer05 | whitequark: no wonder I never heard of 2class E amp" before. In my book this classifies more for a TX oscillator than what I understand is an amplifier | 01:57 |
whitequark | yeah | 01:58 |
whitequark | I was surprised too | 01:58 |
DocScrutinizer05 | would be pretty hard to make a audio amp out of this ;-) | 01:58 |
whitequark | pfff, who needs audio | 01:59 |
DocScrutinizer05 | hehe | 01:59 |
DocScrutinizer05 | actually I wondered how the heck they would modulate that thing | 01:59 |
whitequark | AM | 01:59 |
DocScrutinizer05 | 100% preferably | 01:59 |
whitequark | you just change the supply voltage of the amplifier | 01:59 |
DocScrutinizer05 | which is what they did, on a morse rate of 0.1 char/minute | 02:00 |
whitequark | like, use a controlled boost converter for that | 02:00 |
whitequark | at least that is what I will do | 02:00 |
DocScrutinizer05 | :nod: | 02:00 |
ray__ | any one want some green apple cookie? | 02:02 |
DocScrutinizer05 | bot? | 02:05 |
ray__ | nope | 02:05 |
ray__ | trying to be nice 2 | 02:05 |
ray__ | 2+2= 4 | 02:05 |
DocScrutinizer05 | when that's a turing test then you didn't pass it yet ;-) | 02:07 |
ray__ | i know marry popion movie | 02:10 |
DocScrutinizer05 | :-D | 02:10 |
DocScrutinizer05 | alas I don't | 02:10 |
ray__ | DocScrutinizer05: you a bot? | 02:10 |
DocScrutinizer05 | maybe ;-) | 02:10 |
ray__ | DocScrutinizer05: so you are? | 02:10 |
DocScrutinizer05 | some think I am, yes | 02:10 |
DocScrutinizer05 | ;-) | 02:10 |
ray__ | DocScrutinizer05: im for real | 02:10 |
ray__ | DocScrutinizer05: school starts soon | 02:11 |
DocScrutinizer05 | I'll proceed to the shower and see if any shorts are caused ;-) | 02:11 |
Action: DocScrutinizer05 waves | 02:12 | |
DocScrutinizer05 | austic? | 02:12 |
DocScrutinizer05 | hmm | 02:12 |
DocScrutinizer05 | we shouldn't be hostile to him, eh? | 02:13 |
whitequark | DocScrutinizer05: btw, a less braindead version of that poettering thing | 02:15 |
whitequark | http://nixos.org/ | 02:15 |
whitequark | http://nixos.org/nixos/about.html outlines the benefits | 02:17 |
DocScrutinizer05 | mhm | 02:23 |
whitequark | wpwrak: remind me, which is the good rigol today? | 03:32 |
whitequark | 1102E? | 03:34 |
whitequark | 1074Z? | 03:39 |
whitequark | the *B series I can find here is either sold out or has very little RAM | 03:41 |
whitequark | 1074Z seems nicest so far. lot of RAM, big screen, four channels | 03:44 |
eintopf | hi | 07:01 |
wpwrak | whitequark: 1000Z is nice. 4 chan, low cost, hackable (if you think of buying all the options individually, then you'd be better off with directly buying a higher-end scope), | 07:18 |
wpwrak | whitequark: drawbacks: 1) sample rate gets mercilessly divided down to 250 MSa/s. 2) intensity grading not as nice as in the better rigols. | 07:20 |
wpwrak | but if you can live with these limitations, then it's a good choice. if you want fancier, you could get a 2000. but they come only with 2 channels and hacking them is much harder. | 07:21 |
wpwrak | 3rd option is siglent. new kid on the block in that range. their sds2000 looks quite promising but still had a number of ugly quirks in the firmware (plus a stupidly placed knob). a bit cheaper than the rigol 2000, less unbundling, and goes up to 4 analog channels. | 07:23 |
eintopf | wpwrak: when I have money, I will ask you about an oscilloscope and then I will buy it. I am 100% sure you are the expert about low cost oscilloscopes in home environment. | 07:30 |
eintopf | wpwrak: do you think I can build an own oscilloscope with an high ADC sampler and some serial wire connection to pc. PC runs an application which plot data from serial. | 07:31 |
eintopf | this will be slow, but it's a possible solution to build one? | 07:32 |
wpwrak | serial = usb, yes :) | 07:32 |
wpwrak | ah, build a complete device | 07:32 |
eintopf | ftdi chip | 07:32 |
wpwrak | well yes, but it'll be a toy, not really useful | 07:32 |
eintopf | wpwrak: this would be a post on hackaday! | 07:33 |
wpwrak | naw, there's already a ton of "arduinoscopes" and such junk | 07:33 |
eintopf | grml :(, somebody stole my idea. I doesn't know this. | 07:33 |
eintopf | and arduinologicanalyzers, too? | 07:34 |
wpwrak | i'm sure there are some as well | 07:34 |
wpwrak | i built one with the ben :) actually somewhat useful - does a few dozen MHz | 07:34 |
eintopf | another idea was to buy some arm with fpga and built an open usb debugger with that. | 07:34 |
eintopf | but I can't programm a fpga. The usb debugger should work with usbmon. | 07:35 |
eintopf | and many companies are angry to me, who sells some kind of this hardware for 2000 eu | 07:35 |
eintopf | eur* | 07:36 |
wpwrak | you could consider it a learning opportunity :) | 07:36 |
wpwrak | and no, you won't be able to sell something like this at such a high price :) | 07:36 |
wpwrak | fpga-based LAs do of course exist, too. and they're relatively inexpensive. | 07:36 |
eintopf | so all my ideas are already developed | 07:38 |
eintopf | okay, an android app/iphone app with gpg encryption to load something in the cloud | 07:38 |
eintopf | in connection with an anelok | 07:38 |
eintopf | so we have already some popular members like rihanna | 07:39 |
eintopf | :P | 07:39 |
eintopf | but then you need to put some swarozky stones on the anelok! | 07:39 |
eintopf | wpwrak: (selling) it was not my goal to selling such device. Goal is too have a open source community driven stable device for something like that. ;) | 07:40 |
eintopf | sniffing usb devices | 07:41 |
eintopf | s/swarozky/swarovski | 07:44 |
wpwrak | (glass) i'm sure she can afford real diamonds ;) | 07:46 |
wpwrak | (open) there's also an open critter of that kind. also, there's bitscope | 07:46 |
wpwrak | what they all have in common: they're not very good. e.g., the fpga-based LAs often just use the fpga memory. so they can't store long sequences. | 07:47 |
eintopf | a doctorand of my university works with hash algorithmn (tested SHA-3 candidates) he use xillinx fpga's | 07:49 |
eintopf | and the linux IDE | 07:49 |
eintopf | he always are very angry about the IDE | 07:49 |
eintopf | always broken | 07:49 |
eintopf | error messages like windows | 07:49 |
eintopf | s/are/is/ | 07:50 |
eintopf | wpwrak: (low memory) but when you have a high bus, like PCI-E and connected FPGA, you can use the memory from arm processor or something else | 07:51 |
eintopf | and I mean dma activated etc... you can already buy some solutions like this | 07:51 |
wpwrak | (sythesis tools) everybody hates them. have you seen this ? http://captiongenerator.com/4460/Hitler-manages-a-FPGA-design-company | 07:55 |
wpwrak | those fpga-LAs normally go USB. yes, if you made a board that goes into a pc, you could do something a lot nicer | 07:56 |
wpwrak | but also a lot harder to get right | 07:56 |
kyak | wpwrak: amazing ! :) | 07:57 |
eintopf | that's not funny when you know german language | 08:07 |
eintopf | okay... a little bit | 08:07 |
wpwrak | turn off the sound :) | 08:08 |
eintopf | yea, but the most funny part is the angry reaction speaking | 08:12 |
eintopf | and this isn't funny if you know what he is talking about. | 08:13 |
eintopf | In the 00er years of the internet there was something similar like this. "Fanta world news" | 08:13 |
eintopf | some japanese news and you can put own subtitle on it | 08:14 |
eintopf | that was funny! | 08:15 |
eintopf | but I never know what they really talked about | 08:16 |
eintopf | wpwrak: do you can also speak some japanese? | 08:16 |
wpwrak | nope, none of the asian languages | 08:22 |
eintopf | wpwrak: but you decrypt sometimes asian datasheets? | 09:08 |
wpwrak | with a lot of luck ... and a lot of english :) | 09:11 |
whitequark | wpwrak: yeah I think 1074z | 12:43 |
kyak | wpwrak: http://hackaday.com/2014/09/02/developed-on-hackaday-chromefirefox-appsextensions-developers-needed | 13:02 |
kyak | dunno maybe you've seen | 13:02 |
wpwrak | kyak: the apps extension ? no. but it's a logical piece. mooltipass ? sure. has a bit different goals than anelok, though. last time i checked, it was designed around a smart card. | 14:57 |
Action: DocScrutinizer05 is undecided if likes "moolti" | 17:46 | |
DocScrutinizer05 | sounds like Jovovich in "the 5th element" | 17:47 |
DocScrutinizer05 | http://scifun.files.wordpress.com/2010/03/multipas.jpg | 17:49 |
DocScrutinizer05 | https://www.youtube.com/watch?v=9jWGbvemTag | 17:50 |
wpwrak | yeah, that's where they "stole" the name. let's hope it won't get them into legal trouble. | 18:15 |
DocScrutinizer05 | let's hope it's not as pathetic as the original muhltipass | 18:16 |
DocScrutinizer05 | Jovovich is hot, but the movie is more like a huge parody | 18:17 |
DocScrutinizer05 | I'd be reluctant to use it for a branding | 18:17 |
Action: whitequark pokes ltspice | 18:18 | |
whitequark | ltspice seems pretty cool, except it apparently hasn't evolved since about uhh | 18:18 |
whitequark | 1994? | 18:18 |
DocScrutinizer05 | watch out, I heard it pokes back | 18:18 |
whitequark | either the GUI or the engine | 18:19 |
DocScrutinizer05 | N-either ? | 18:19 |
DocScrutinizer05 | nor | 18:19 |
Action: DocScrutinizer05 glances at quics | 18:20 | |
DocScrutinizer05 | err qucs | 18:20 |
whitequark | either GUI or engine have not evolved, or, neither GUI nor engine have evolved | 18:20 |
whitequark | I would expect better understanding of boolean logic from you :> | 18:21 |
DocScrutinizer05 | exactly, the first is a logic XOR, the second a logic NAND | 18:21 |
wpwrak | whitequark: now you know the exact age of the world, according to "young earth" creationists :) | 18:22 |
whitequark | DocScrutinizer05: oh | 18:22 |
DocScrutinizer05 | s/NAND/NOR/ | 18:22 |
whitequark | indeed | 18:22 |
whitequark | foiled by my own deeds! | 18:22 |
whitequark | wpwrak: the world was created at 00:00 Jan 1 1980 | 18:23 |
wpwrak | aka D(OS)-day ;-) | 18:27 |
whitequark | it's Unix epoch, no? | 18:27 |
whitequark | or was it 1970? | 18:27 |
whitequark | right, 1970 | 18:27 |
DocScrutinizer05 | actually | 18:28 |
DocScrutinizer05 | jr@saturn:~> date -d@0 | 18:28 |
DocScrutinizer05 | Do 1. Jan 01:00:00 CET 1970 | 18:28 |
DocScrutinizer05 | hmmmm | 18:30 |
DocScrutinizer05 | jr@saturn:~> TZ='America/Los_Angeles' date -d@0 | 18:30 |
DocScrutinizer05 | Mi 31. Dez 16:00:00 PST 1969 | 18:30 |
Action: DocScrutinizer05 starts musing | 18:31 | |
DocScrutinizer05 | the world got created all the same time? | 18:31 |
DocScrutinizer05 | and when same time, then absolute, or same local time? | 18:31 |
whitequark | -d specifies offset including timezone | 18:32 |
whitequark | not Unix time, in fact | 18:33 |
whitequark | (because Unix time is at UTC) | 18:33 |
--- Wed Sep 3 2014 | 00:00 |
Generated by irclog2html.py 2.9.2 by Marius Gedminas - find it at mg.pov.lt!