You're displaying your ignorance on the topic.  GSM is a wideband commercial product not intended to mesh.  GSM cell phones have a practical range of about two kilometers, but that is to a very senitive celltower tranceiver.  If you tried to mesh two handsets, your range is unlikely to be any better that wifi.  Basicly, if you can see the whoites of their eyes or be heard yelling, then it would work.  There are many variables that effect range; including effective radiated power, receiver senitivity & bandwidth, mode of transmission, error correction algos, etc.  Comparing a mesh tech to that of GSM isn't even comparing apples to oranges, it's comparing apples to pencils.  And the goal of such a mesh isn't to connect two cell phones together with a high enough quality to support a voice codec, it's to transmit a data file of half a kilobyte within a few seconds so that device cound forward it again until the transaction finds an internet router.

Alright, how do you come to this conclusion?  Support your position.

Thank you for your unsupported layman's opinion.  As for those of us who actually know a thing or two about radio telemetry & propagation, we've been providing actual facts on this kind of device for longer than most of you in this thread have even known what a bitcoin was.

EDIT: I've literally bounced a 2 watt signal off the F level of the ionosphere and was heard over 200 miles away from my position; and actual line-of-sight distance of at least 350 miles up and down.  Granted, that was with some high quality gear that wouldn't fit in my wallet to save my life; but for someone who's only experiences with low-power digital telemetry involves the wifi scanner app on his smartphone to tell me it-just can't-be-done is offensive.

They don't need to be trusted.  I can do exactly the same thing right now using my openwrt wifi router, simply by blocking non-bitcoin port numbers and letting bitcoin port numbers pass unhindered; or by capture and redirection of all client-side traffic to my local bitcoind.  The whole point of bitcoin's structure is that there don't need to be trusted gateways.  There is little to no malicious activities that any particular gateway node could do to any user.  Can't steal their coins.  Can't expect to double-spend against such a device, since it's as likely as not to be somewhere else when coins are sent to it anyway.  (it's a consumer wallet, not a point-of-sale device.)  Can't see how a sybil attack is likely against such a device, for similar reasons.  Having a full blockchain makes such things impossible, but even just having the block headers, merkle trees & copies of input transactions relevant to the addresses that the wallet has private keys for makes pretty much anything unlikely, or at least "costly" enough of a crime to not make the funds that could be found in a random wallet device worthwhile.

The only malicious activity that I can think of that might be possible, is to feed the wallet device false input transaction data, but even if the client accepts them as true, no one else would so the device simply wouldn't be able to spend those coins that it never really had; which might irritate the owner, but he still couldn't lose anything as a result.

I'd buy one today.

And yes, they could work well, in theory.  And yes, they would need routers in addition to peers, but those would be provided by retailers.  The mesh mechanism is to extend the functional range of such a device, but is inmaterial for a bitcoin device.  A disconnected bitcoin wallet isn't disconnected when you're home, only while you're out using it.  So it can update it's blockchain (assuming it needs it, but it shouldn't) while you are at home or otherwise within wifi range of an open hotspot.  The devices need to be able to broadcast transactions, & hear other transactions, to protect the owner from a possible form of theft by deception.  Namely, if the device didn't broadcast transactions until it could connect to a wifi hotspot, a criminal could pretend to buy the car you were selling on Craigslist, pay you with bitcoin using both your wallet devices, and after you drive away the thug you paid to mug him would destroy his wallet device.  Thus giving the fraudster the chance to overwrite the transaction that he paid with on his own device and keep the funds.  If all bitcoin devices broadcast all transactions immediately, the fraudster in this scenario can't know if the transaction was heard by another wallet device, another person's cell phone, a router connected to the internet or nothing at all.  This lack of certainty removes the potential gain from such a criminal strategy.  It's not really speed of transaction propagation that is critical, but the potential of same.  If bitcoin is ever as popular as Paypal, we can expect that retailers that start accepting bitcoin are going to also invest in a receiver for broadcasted bitcoin transactions, since it's in their own interests to make certain that transactions make it to the Internet as quickly as possible to prevent double spends attacks against themselves.  Although it would be possible to exclude transactions that don't concern the retailer, the additional overhead involved in making such a discriminating router would most often exceed the marginal additional cost of simply forwarding every bitcoin transaction that it hears.  A store the size of WalMart could get away with only one such receiver per store.  A mall might need two or three for all the stores, depending upon it's size and layout.  A bunch of small stores down main street could get together to sponsor one for the whole block, although the costs of such a receiver if you already have Internet access at the store (who doesn't in America anymore?).

Theories aside, there wouldn't need to be a critical mass for such a device to be useful, and in places with retail density, the mesh traffic might actually be counterproductive.  It's when you're far from a retail outlet doing business that such distribution of the transactions are necessary.

An attacker could bias your device from afar, but he'd also need to know precisely when you pushed the 'start' button.   And if you were using a normal radio to do it, you could hear the attempt at screwing with it.  Although it's technically possible, the practical means of influencing the entrophy in this fashion is pretty remote.

Yes, relevant.

How did you get ahold of these so cheap, and how many do you have?  I'm going to bid just so I can play with it, but I might need two.

I realized today that I left out an important step in my previous description of the simple biased gossip protocol.  The header of any packet must contain a simple hash of the data, so that (lacking forward error correction) a receiving node can know when the message it heard was corrupted, and simply discard it.  Again, I don't think error correction would be worthwhile because 1) although AM is a noisy mode, UHF tends to not be that bad, at least concerning natural noise and 2) it's not actually necessary that a transaction be heard and forwarded by other devices, it's just beneficial.  The repeated transmission at intervals takes care of the redundancy of data issue without error correction.  I'm guessing it would, anyway.

Well, that's exactly why I mentioned an am receiver tuned to static.  Pipe that into the stereo mic-in jack of a small computer, mash the resulting bitstream up with some hashing algos, and you've got a pretty decent RNG hardware on the cheap. 

If it's a dash7 device, it uses 433 mhz.

It doesn't hold a blockchain, so it doesn't actually have to update in that manner.  Yes, it would require some kind of supporting infrastructure, such as a specialized internet router that acts as a Stratum server.

It's more of a motorbike than a car....

http://www.hover-bike.com/

Actually, there is a new type of capacitive touch screen that is actually a solar panel itself, so that you can let your phone charge by simply placing it on the table face up and letting the screen time out.

http://ht.ly/6yUTG

Here's the site of a manufacturer of a similar device, but not the same device.

I saw this the other day.  It looks to me like someone has been reading my other threads...

https://bitcointalk.org/index.php?topic=75729

https://bitcointalk.org/index.php?topic=8540.0

https://bitcointalk.org/index.php?topic=53567.0

https://bitcointalk.org/index.php?topic=1256.0

https://bitcointalk.org/index.php?topic=1256.0

https://bitcointalk.org/index.php?topic=2386.0

https://bitcointalk.org/index.php?topic=2039.0

And managed to find some hardware to do it with.  I've been trying to find the manufacturer of the card device in order to get more details on the thing, but so far it seems to be vaporgear.

Although this is a cool project, and I'm sure that I'll participate when the time comes, it's not really relevant to the topic problem.  However, this group is bound to be a great resource regarding the FCC process should a licensed band prove necessary.

This is true, and exactly why NFC tech is being included into some phones these days for Google Wallet and other similar credit card substitutions.  I don't know that I would want to depend upon NFC being in every hardware device, however.  I'd much rather start with a single radio device that can manage to do everything that it needs to with that one radio first, if that's possible.

Discussion about how to use bitcoin to circumvent taxes or government finance laws is a verboten topic on this forum.  If you can manage making your bitcoin transactions anonymous, you can manage to get to the SilkRoad also.  I'm sure that if you want to have this conversation there, many will be accomodating, although don't assume that you aren't talking to a cop there either.

Mind?  No I don't.  But the US federal government isn't a monolithic structure, in order to spend the kind of money & time to do the deed, some faction is going to have to get enough of a budget to do it.  The only department with enough of a budget now would be either the military or the CIA.  Neither has much interest in monetary policies or "crimes".  The attempt to approve such a budget for any dept that would actually care would send up some real red flags within the government itself.  I'm not saying that it couldn't be done, if anyone can the US government can.  I'm saying that the US government has easier attack vectors to pursue first, such as simply making this all illegal and a huge propaganda campaign.

The 433 mhz band is an international ISM band, just like 2.45 Ghz that wifi uses, so it's legal at less than half a watt most places and no one cares much about the rest since they still work just fine.


That would work, assuming that the 'heartbeat' function was turned on.  However, I'm concerned about a man-in-the-middle type attacker, who places a hacked device hidden near a POS system and randomly attempts to pretend to be the POS register, or just uses a beam antenna from across the street.  Most of the time it wouldn't work anyway, but if it works at all it's a problem.  There would have to be some kind of challenge & response crypto standard, say to display a four digit pin on both the POS system checkout screen and your device so that you can look at them and make certain that they match.  This wouldn't matter for the messaging system, because messages could be encrypted to the receiver's public key, which you aquired some time ago in a 'bump' kind of fashion when tapping the two (gps known and in radio contact) devices together.  If public key encryption is too much for the device (probably not, if it's got to do the bitcoin encryption anyway) the devices could trade a set of random data to be used as one-time-pads.  Actually, for texting this would be better, for public key encryption has the habit of making the messages much longer than any simple text message could be, which is one reason that SMS isn't encrypted normally.  Longer than necessary messages would be a killer for the app.  The messaging functions of the device would simply have to collect entrophy data, and then package that into a one-time-pad, trade that data in near proximity (this is where NFC shines) and encode messages with a 1/1 byte ratio, then add a header that told the receiving device where the message starts in thier pad in case there were unreceived messages, then change all used pad data to zeros.  Any device that used the same nym wouldn't be able to see the message.  Plus we would have to assume that there would be nodes that saved all traffic, so privacy would require encryption anyway.  The device would have to have enough flash memory to keep a several kb of pad data per saved contact.  5kb for the sending pad, 5kb for the receiving pad and you're good for several hundred texts before you have to get close again.  The app could warn you when your pads were getting low.


I think you should do a hack to get that transceiver hooked up to a USB output together with a reasonable aerial. The datasheet says it takes serial input. That way developers with breadboarding skills can make a couple and start trying them out. You can crowdsource the writing of the mesh network software ! :-)
[/quote]