Anonymous Internet Banking Project

[grondilu]

This is my secret project I'm currently working on.  A full internet anonymous/pseudonymous banking service.

Basically it will allow anyone to own financial assets anonymously (or pseudonymously) and to trade them.

It will be free and open source, because the more implementations, the better.

It will be centralized.  Yes, centralized.  Not P2P.  Because a financial asset is not a money.  You NEED TO trust someone who promesses to pay revenues from this asset.  So until an other genius like Satoshi invents a dividend-paying cryptoasset, we got to stick to the reputation/trust/responsabiliy model.

Every asset account will be indentified by a cryptographic key.  Several types will be accepted.  GnuPG, DSA, ECDSA.  Any request to accept any other kind of key will be studied.     Public keys can be large, so we'll use hashes to designate accounts.  Again, no restriction on the type of hashes.  But only "good" hash function.   RMD160, sha256, sha124,...  all those hases from the same public key will be accepted to identify any asset.

Asset transfer will be possible by signing a document made of the asset transfered amount, the destination public key (or its hash) and the current bitcoin block hash as a time indication.

All transactions will be public, so that everyone can check signatures and balances.  The full transaction history will be available via HTTP.

The server will be hidden on the Web using Tor.

The bank will offer a clearing/escrow/compensation service for trading assets.  Crossed order books will be maintained for all possible asset pairs.  Real-time fast trading should be able via XMPP/IRC/Jabber, whatever.


Bitcoin would make all this much easier.  I doubt I could do that all alone, but I very much would like to live long enouhg to see this becoming reality.

[FreddyFender]

Did you place Open-Transactions on the backend or did you create all the code from scratch? What percentage complete are you? I noticed you posted on the OT posting and figured you put 2&2 together, eh?

[Vinnie]

This is my secret project I'm currently working on.  A full internet anonymous/pseudonymous banking service.

Basically it will allow anyone to own financial assets anonymously (or pseudonymously) and to trade them.

Will these assets be traditional securities? In other words, will the bank (or bank owner/operator) have title to them according to the rest of the world?

[Cryptoman]

The last time I tried Tor hidden services, it was very slow.  Have you considered anonymously registering your domain and then hosting it on the regular internet (mirrored, of course)?  You could do a continual transaction backup through Tor to a server you have possession of.

At any rate, I think it's a great project, and I wish you the best of luck. 

[grondilu]

Did you place Open-Transactions on the backend or did you create all the code from scratch? What percentage complete are you? I noticed you posted on the OT posting and figured you put 2&2 together, eh?

Oh indeed I forgot about that.  I haven't already looked into OT enough.

It's possible that it is what I am trying to do indeed.

But for some reasons I want to start an other implementation from scratch.

[Nefario]

Did you place Open-Transactions on the backend or did you create all the code from scratch? What percentage complete are you? I noticed you posted on the OT posting and figured you put 2&2 together, eh?

Oh indeed I forgot about that.  I haven't already looked into OT enough.

It's possible that it is what I am trying to do indeed.

But for some reasons I want to start an other implementation from scratch.

I have sent you a pm about this project, it is something we can work together on, I have resources that I can use for this project. Anyone else that is interested (by contributing technical skills, resources such as computer or network, or with donations or "investments") pm me.

[grondilu]

I have sent you a pm about this project, it is something we can work together on, I have resources that I can use for this project. Anyone else that is interested (by contributing technical skills, resources such as computer or network, or with donations or "investments") pm me.

Yeah I'll definitely look into it.

It requires OpenSSL version 1.0.0, which is quite a high version number.  I'm on debian Sid and I only have version : 0.9.8o-4.

The 1.0.0c is on debian experimental, though.  I'll have to install it.

[Nefario]

I have sent you a pm about this project, it is something we can work together on, I have resources that I can use for this project. Anyone else that is interested (by contributing technical skills, resources such as computer or network, or with donations or "investments") pm me.

Yeah I'll definitely look into it.

It requires OpenSSL version 1.0.0, which is quite a high version number.  I'm on debian Sid and I only have version : 0.9.8o-4.

The 1.0.0c is on debian experimental, though.  I'll have to install it.

Don't worry, I had already pm'd you. I have installed 1.0.0c on the machine. Ubuntu 10.10 so you should be familiar.

[grondilu]

Don't worry, I had already pm'd you. I have installed 1.0.0c on the machine. Ubuntu 10.10 so you should be familiar.

I'm on IRC right now if you want to talk.

[ShadowOfHarbringer]

This is my secret project I'm currently working on.  A full internet anonymous/pseudonymous banking service.

Basically it will allow anyone to own financial assets anonymously (or pseudonymously) and to trade them.

It will be free and open source, because the more implementations, the better.

It will be centralized.  Yes, centralized.  Not P2P.  Because a financial asset is not a money.  You NEED TO trust someone who promesses to pay revenues from this asset.  So until an other genius like Satoshi invents a dividend-paying cryptoasset, we got to stick to the reputation/trust/responsabiliy model.

Every asset account will be indentified by a cryptographic key.  Several types will be accepted.  GnuPG, DSA, ECDSA.  Any request to accept any other kind of key will be studied.     Public keys can be large, so we'll use hashes to designate accounts.  Again, no restriction on the type of hashes.  But only "good" hash function.   RMD160, sha256, sha124,...  all those hases from the same public key will be accepted to identify any asset.

Asset transfer will be possible by signing a document made of the asset transfered amount, the destination public key (or its hash) and the current bitcoin block hash as a time indication.

All transactions will be public, so that everyone can check signatures and balances.  The full transaction history will be available via HTTP.

The server will be hidden on the Web using Tor.

The bank will offer a clearing/escrow/compensation service for trading assets.  Crossed order books will be maintained for all possible asset pairs.  Real-time fast trading should be able via XMPP/IRC/Jabber, whatever.


Bitcoin would make all this much easier.  I doubt I could do that all alone, but I very much would like to live long enouhg to see this becoming reality.

Haven't i heard this somewhere ?
What is the difference between this and loom ?

https://loom.cc/

[Timo Y]

It will be centralized.  Yes, centralized.  Not P2P.  Because a financial asset is not a money.  You NEED TO trust someone who promesses to pay revenues from this asset.  

Just because you need to trust someone doesn't mean it MUST be centralised.

For example: The ripple monetary system relies on nothing BUT trust in order to work, yet it is fully decentralised.  

In some ways it's Bitcoin's opposite, since Bitcoin was designed to work without any need of trust whatsoever.

The web of trust makes it possible to lend money to a complete stranger who is connected to you via people you trust, and people they trust.  Since they collectively "vouch" for that stranger, you can have a degree of certainty that the stranger will fulfil their promise to you.

This kind of system is more trustworthy and resilient than what you are proposing, because it has no single point of failure, and because it's unlikely that you lose all your assets at once.

I personally would not trust my wealth to an anonymous website for any kind of extended time period (say 5 years) even if I knew the owner was competent and honest. Too many things could go wrong. Death, disease, bankruptcy, data loss, hackers, government crackdowns...

My traditional bank is a safer bet. Here in Switzerland at least, if my bank fails, or screws up, it has an agreement with other banks (not the government, mind you) that they will bail out its customers.  Some kind of insurance by a third party is the least I would expect from an internet bank.

my 0.02BTC...  good luck anyhow.
 

[grondilu]

Haven't i heard this somewhere ?
What is the difference between this and loom ?

https://loom.cc/

Yeah right, I forgot about loom too.  Beat me.

Let's just say I want my own implementation

And I want it in bash.

[grondilu]

Well, until I find an hosting alternative, I can open an hidden service using my personal computer.    It is not always connected to internet, though.

http://3p3y4pqhrgpxoqw6.onion/brokerage.cgi

This is my personnal laptop, so don't try anything funny, ok ?

[ShadowOfHarbringer]

http://www.kimsufi.pl/ks/

Super-cheap dedicated hosting.
One core/250GB/2GB/100Mbit starts from about 20 EUR / Month.

And yes, it is quite stable/reliable, I'm already using it for more than a year.

Here is the french page - it's very alike in matter of pricing & selection:
http://www.kimsufi.com/ks/

There is an UK site too, but it is little more expensive.

[kiba]

Shouldn't you guys look for hosting services that you can pay in bitcoin? It support the bitcoin economy plus you don't have to go through the extra step of converting bitcoin to euro or dollars.

[fellowtraveler]

This is my secret project I'm currently working on.  A full internet anonymous/pseudonymous banking service.

Basically it will allow anyone to own financial assets anonymously (or pseudonymously) and to trade them.

It will be free and open source, because the more implementations, the better.

It will be centralized.  Yes, centralized.  Not P2P.  Because a financial asset is not a money.  You NEED TO trust someone who promesses to pay revenues from this asset.  So until an other genius like Satoshi invents a dividend-paying cryptoasset, we got to stick to the reputation/trust/responsabiliy model.

Every asset account will be indentified by a cryptographic key.  Several types will be accepted.  GnuPG, DSA, ECDSA.  Any request to accept any other kind of key will be studied.     Public keys can be large, so we'll use hashes to designate accounts.  Again, no restriction on the type of hashes.  But only "good" hash function.   RMD160, sha256, sha124,...  all those hases from the same public key will be accepted to identify any asset.

Asset transfer will be possible by signing a document made of the asset transfered amount, the destination public key (or its hash) and the current bitcoin block hash as a time indication.

All transactions will be public, so that everyone can check signatures and balances.  The full transaction history will be available via HTTP.

The server will be hidden on the Web using Tor.

The bank will offer a clearing/escrow/compensation service for trading assets.  Crossed order books will be maintained for all possible asset pairs.  Real-time fast trading should be able via XMPP/IRC/Jabber, whatever.

Bitcoin would make all this much easier.  I doubt I could do that all alone, but I very much would like to live long enouhg to see this becoming reality.

I definitely suggest you take some time to play around with Open Transactions, and read over the wiki.

Email me if you have any questions.

-FT

[grondilu]

I definitely suggest you take some time to play around with Open Transactions, and read over the wiki.

Email me if you have any questions.

If you don't mind I'll ask you my questions here.  There's no reason why it should not be public.

I don't like OT very much.  Seems quite complicated right now.  But I will take a deeper look into it when it implements stock markets.  According to the main page, it should come soon, right ?

[fellowtraveler]

I definitely suggest you take some time to play around with Open Transactions, and read over the wiki.

Email me if you have any questions.

If you don't mind I'll ask you my questions here.  There's no reason why it should not be public.

I don't like OT very much.  Seems quite complicated right now.  But I will take a deeper look into it when it implements stock markets.  According to the main page, it should come soon, right ?

Suggest you check out the Use Cases for a list of the basic client-side functionality:
https://github.com/FellowTraveler/Open-Transactions/wiki/Use-Cases

That article also shows which API functions to call for implementing each case.

To play with the functionality without coding, try the command-line test client.
Instructions:
https://github.com/FellowTraveler/Open-Transactions/wiki/Test-Wallet-commands

OT already supports markets. (That is, trading 2 asset types against each other by making an
offer onto a market, where it may trade against other offers from other users.) To read more
about OT markets, here:
https://github.com/FellowTraveler/Open-Transactions/wiki/Markets
Note that each trade involves 4 accounts. If we are trading dollars for wheat, then there
are my dollar and wheat accounts, and there are your dollar and wheat accounts. Total of 4.

What OT does NOT yet support is stocks. (A stock being an asset type that may occasionally
pay dividends, as set by the issuer, which are paid out in a different asset type.) Once stocks are
added to OT, you will be able to trade them on markets like any other asset type.
You may find value in this article about future potential instruments on OT:
https://github.com/FellowTraveler/Open-Transactions/wiki/Future-Direction

From what you have written, OT seems like what you are describing, and having written it,
I can assure you that playing around with OT for a few hours will be much easier and simpler
for you, than actually writing it from scratch was for me.  I have endeavored to protect you
from that complexity as much as possible with a simple interface (the OT API and classes.)

FYI, here's a project to create a perl wrapper for the OT API:
https://github.com/DsPearson/tempest

There is a web (HTML) GUI layered over that, but soon the author will release a command-line
version as well. (A real UNIX style command line, unlike the current OT command-line test wallet.)
You may find that such a command line is also well-suited to your needs.

Good luck!

-FT

[Nefario]

fellowtraveler:

I will be using OT in my project (using ruby on rails), I've been playing around with the demo client and server alittle and think it's just right.

However, when trying to get the bindings for ruby to build I was having some difficulty. I had ruby 1.9.2 built, and for openssl had it link to 1.0.0c openssl. But OT wouldn't build the ruby bindings.

Any tips, I am not a c cpp person, just a lazy and slow scripter.

Also a question about markets, this is one area I didn't play around with. What happens to the spread when a transaction occurs?

Actually I'm very happy you are on the bitcoin forums, you certainly know who will be the early adopters. Also any time frame for stocks and contracts that pay dividends will be supported ?

[grondilu]

Anyway I have already written a bash CGI that does what I wanted.

Basically the owner of an asset is identified by a bitcoin address.   Thanks to dirtyfilthy, it is possible to extract a bitcoin private key from the wallet, and thus to sign a transaction using openssl.  Using a bitcoin address as an identifier makes it much easier for stocks (dividends just have to be paid on this bitcoin address).

Unfortunately I have to wait until I get a better web hosting solution to show it to the world.

The source code is already available on github though :

http://github.com/grondilu/bitcoin-bash-tools