Professional TOR network with Bitcoin fees?

[wolverine.ks]

Anyone have any updates on the pay-to-use TOR nodes that use Bitcoin as the currency? I heard the idea a while back, and was wondering if
anyone has been working on it or heard news about it.

Additionally, if this went anywhere, it would seem that it would necessitate lots of really small transactions.
Any thoughts on if they would bump up against the 5430 limit?

Perhaps this would be better suited for a litecoin or a feathercoin?

[bytemaster]

I am actively developing such a system...  the challenge is micropayments, but I think I have it all worked out.

[wolverine.ks]

I'm super excited about that!

keep me posted on that development. and I'll let all my friends know about it.

do you have a website or repository?

[MysteryMiner]

This might be anonymity problem because of correlation and traceability of bitcoins when not handled properly. Imagine 50 people use Tor in the country. But only 7 of them are known to have purchased Bitcoins or have been informed by bitcoin existence. Looking for them all moving their coins while timing the Tor connection might de-anonymize the circuit.

How about someone who cares about internet privacy sponsoring dedicated Tor exit node with bitcoins? I have spare computer around unusable for daily activities because of too weak hardware but I can buy separate anonymous connection and have the exit node running 24/7/365. I need the separate and anonymous connection because running exit node on my own connection created too much hassle some time ago and now it become dangerous. I cannot afford the separate connection by myself right now.

[wolverine.ks]

I realize that more information means more traceability, but is 50 a realistic number to use? Well, perhaps in small countries with not a lot of TOR adoption.

Perhaps if there could be some auto coin washing step in the process...

Perhaps there could be a some sort of premium tor node directory that is subscription based, so you could pay one ip address to access a whole bunch of premium nodes, then the nodes get a portion of the profits based on a percentage of the traffic? If it were done on a monthly basis, and there were enough subscribers, wouldn't that provide enough cover?

or perhaps someone much smarter than me comes up with something completely different and completely awesome....

or perhaps a completely new internet that has anonymizing baked right in...

[MysteryMiner]

I don't want to discourage the development of the project if it is feasible. But the problem with premium node directory list will be sharing of the list (piracy). Also what prevents the exit node for claiming more traffic than it really processed (cheating). There must be some sort of cryptographic link between payment for circuit and source node for that payment. Could it be done anonymously?

Probably some foundation could collect the donations in various forms (including bitcoins) and then forward them to exit node operators. This will benefit everyone using Tor because not everybody could afford paying for premuim circuit. I think Torservers.net is a good starting point.

And there is stopgap solution to deploy Tor exit nodes on botnet computers.

or perhaps a completely new internet that has anonymizing baked right in...

Not going to happen for two reasons. Current internet have too much userbase to migrate to completely new solution. And the government controlfags would dislike network who is not censorable or controllable for them. They will kill the project in infancy.

But anonymity technology on top of current network is feasible, it works for it's purposes.

[wolverine.ks]

is there a way to have the premium nodes have a whitelist? they get a message form the directory host that payments have been made to whitelist certain ip address?

maybe the fees could be flat fees based on the total total bandwidth provided by the pooled nodes?

i think there could be a market for competing directories, similar to the market for competing ISP's

directories could claim certain bandwidths and certain prices.

government controlfags

oooh, im gonna steal this one....

[MysteryMiner]

Just got an idea. What if the exit node receives bitcoins anonymously, then the last layer of the Tor encrypted data is encrypted using the Bitcoin private key used to sign transaction of bitcoin payment? The exit node then can decrypt to see that the data come from same entity who signed the bitcoin tx. In such way the IP of person using the premium service is not revealed. The directory of nodes describing their services and prices might be added to standard Tor directory. Something similar might be used for relay nodes. The payments must be made in two unrelated transactions to avoid linking of identities if nodes are cooperating in deanonymization attack.

This of course means there might be need for adjustment of encryption keys in Tor software and close integration of Tor/Vidalia and Bitcoin-Qt. Probably Bitcoin-Qt might be unchanged but Vidalia and Tor will communicate to Bitcoin-Qt like Armory does.

This probably opens more network privacy and anonymity problems but this might be good start.

[bytemaster]

My approach only use bitcoin backed anonymous cash (ie: Open Transactions).
Each node in the chain is paid in turn... with a 'credit basis' so you pay for 10MB at a time.
Any node that 'cheats' is obvious and people stop routing to it.
Every node in the network can operate 'at profit' while keeping the transactions private.

Any 'trusted node' can serve as a 'bank' for the other nodes and the bank would manage anonymous accounts funded via bitcoin.

The challenge is to scale up the processing / exchanging of digital cash so that it doesn't become a bottleneck. 

When it is all said and done people can 'mine' bitcoins simply by running a node on the network and routing traffic.  Each node will set its own price / bandwidth and the market will drive performance.

The real challenge is not giving up your identity via latency. 

I have old code from related projects up on github, but nothing ready.

(An early version of the program.. riddled with bugs and not worth anyone downloading right now)
https://github.com/bytemaster/tornet

I am just posting the link so people know I am a real, capable software developer... I just need time away from my day job to get this done.    I would love to work on it full time if we could find funding for such a project.

[joesmoe2012]

I have a couple fast servers that I'd be happy to beta test with. When you need a few extra places to help test from, let me know.

[bytemaster]

Ok, this sounds great!   Outside interest/team work can really help motivate me to get this done.  So here is the deal, pester me about this and I will attempt to deliver.   I do best under pressure, so lets set a target deadline for initial capability say 1 month.

Let me give some of my design ideas to get feedback.   I could start with TOR or I2P but unless all nodes are operating for a profit you will have some major performance issues.   Besides, I think TOR and I2P both have some positive design aspects.   In the end, I think it will be faster to create a tor-like system rather than integrating with TOR.

1) It will be UDP based to enable easy traversal of firewalls and enable 'streaming' data that doesn't require order.
2) It will use the UDT protocol to enable streaming of TCP data over UDP
3) Each node will have a 2048 public key that will identify the node.
4) The public key is used to prevent man-in-the middle and to negotiate a blowfish key.
      - the blowfish key is used because it is 'fast' and secure enough and we want to
        minimize the latency in the network.
5) version 1 will implement a simple HTTP proxy interface on the client side that will allow
    any browser to connect.  Later versions will use SOCKS.
6) on the 'exit side' each node can either set up its own http proxy that all 'exit' tcp connections will connect to.

How you pay for service:
1) Initially I will set up a bitcoin bank that will issue anonymous blinded cache tokens (Open Transactions), the client will allow you to deposit bitcoin and withdraw cash tokens.  These tokens will be a small denomination, perhaps 1uBTC.   You can also deposit cash tokens and withdraw to a btc address.
2)  This bitcoin bank will also host a public database of all known nodes in the system along with their IP/public key.  This database will be used by the client to plan routes.
3) Nodes and clients will alternate being the 'creditor' such that initially the client will deposit 1uBTC and have right to 1 unit of data transfer.  The client will be allowed up to 2 units of data transfer before being cut off with a -1 uBTC balance.  To continue the client would then deposit 2uBTC.  This back/forth exchange should cause the 'average balance on close connection' to be 0 and should encourage 'long lived' connections (10 minutes or so) which should have only a small percentage under/over payment.   
       
When you open a connection you plan a route through N nodes and progressively build out the channel.  As you do so you deposit 54uBTC at each node and are credited with a fixed amount of data.   Once the client will automatically deposit more as your balance runs low.   

Now if you leave your node on the network all the time, you will earn more BTC than you use and you can have a 'fast' and 'free' experience.   If you need to download a lot of data then you may deposit BTC.   In theory, the price you pay for the service should approach the cost of bandwidth used.  High-speed nodes may charge more than DSL nodes.   

Initially there will only be one 'bank', but that can easily be decentralized.

An alternative 'micropayment' system would be to have nodes calculate a 'proof of work' toward a mining pool.  Unfortunately, this system creates a delay and will probably require end users to have ASIC to generate enough value to pay for their bandwidth usage in real time.

[qxzn]

This of course means there might be need for adjustment of encryption keys in Tor software and close integration of Tor/Vidalia and Bitcoin-Qt. Probably Bitcoin-Qt might be unchanged but Vidalia and Tor will communicate to Bitcoin-Qt like Armory does.

Or maybe one could introduce Tor credits which can be tightly integrated into Tor software, while at the same time having a freely floating rate of exchange with bitcoin.

[mmeijeri]

Any updates on this?

[bytemaster]

I have put in about 30 hours of development effort.  Turned parts of Open Trx into a library for anonymous cash, learned / integrated elliptic curve cryptography and made progress on the protocol layer.

Then I came up with an amazing way to enable fiat currencies to be turned into fully backed crypto-currencies that enable a true P2P exchange + USD accounts that actually pay 'interest' while being anonymous and entirely decentralized. 

Now I am torn, I want to see both things happen!

[mmeijeri]

I think the distributed exchange thing is well covered by Ripple and OT, though more redundancy is always welcome. I don't see anything happening on the TOR front though, and that may become crucial if statist forces try to clamp down on cryptocurrencies. I think you may be able to make a bigger difference there.

[bytemaster]

I am well aware of what OT and Ripple have and trust me, my solution is both simpler and an order of magnitude more robust.   Granted, we do need a tor like system as well! 

I posted a white-paper on another thread you may want to check out:

https://bitcointalk.org/index.php?topic=213588.msg2238954#msg2238954

A quick summary is this:

1) all balances (fiat or otherwise) pay interest
2) all fiat balances are created by someone borrowing against a crypto-currency used as collateral. 
3) payments from the collateral are distributed to those who hold a balance in that fiat currency.
4) people borrow against their crypto-currency because they expect it to go up in value relative to fiat AND the crypto-currency also pays dividends from transaction fees.
5) once you have a fiat balance it can be traded as easily as a bitcoin balance.
6) fiat balances are always 100% redeemable at face value (excluding transaction fees) so long as the underlying crypto-currency maintains any amount of value.
7) the effective interest rate on fiat balances goes UP when the value of the crypto currency goes up.

Given my approach we can now eliminate the centralized exchange.  Allow individuals acting as 'local-bitcoin operators' or posting on craigslist to 'deposit' and 'withdraw' $USD or EUR from other individuals without having exchange risk but only a potential 'transaction fee'.    There would be more people wanting to deposit $USD because their $USD balances pay interest and with more people wanting to deposit that means more people wanting to withdraw which equals more opportunities for local exchanges.   People could use the system without ever owning an asset denominated in a crypto-currency to trade $USD among themselves, etc.

[joesmoe2012]

tor first please.

[bytemaster]

I'll tell you what, if you can raise more funding for the TOR network then it will get done first.   As it stands I have several orders of magnitude more investment money behind BitShares.   

Now if you want me to switch back to the TOR project, then I can provide the details of my design and let someone else implement it or you can attempt to collect the 10 BTC bounty to get me to abandon BitShares and I will come back to this.

*note*  I still want to see this happen as I think it will be critical!   But I have also recently learned of a new p2p network that is soon to be released (from one of the BitShare backers) that *might* make my work on this redundant.

[ondratra]

I am actively developing such a system...  the challenge is micropayments, but I think I have it all worked out.

The microtransactions could be solved by withdrawing them from Bitcoin network like when putting to cold storage and then sent to a paid Tor node in form of string, couldn't it?


And I have one question for everyone - we are tlaking about paying for Tor, but what is legality of running a Tor exit node? What if user do something illegal in your country (while representing YOU by your IP)?? The police might knock on your door anytime! Did anyone some research about law and Tor? I would be very interesting to learn something

[bytemaster]

You cannot use bitcoin directly to pay for it.  It would be too slow and not anonymous enough.   The minimum transaction fee would be greater than the cost of a 10 minute connection.

Instead what I have proposed is to pick several "trusted" nodes to serve as a bank for blinded cash tokens.  You would have Bitcoin deposit ~ $1.00 at a time and then simply pay by sending the
tokens across the network to each hop.

The trick is to make sure that the nodes do not reveal the circuit by 'cashing in as you build it', they would have to make requests to the bank at regular intervals and 'trust' the tokens in the mean time.

If someone is getting paid for an 'exit node', the 'exit node' itself could use one or more proxies / VPNs.

There is also the potential of white-listing safe sites.