The Holy Grail! I wish I could kiss the author of Bitmessage on his face.

[interlagos]

I would like to suggest that they change the name if this will indeed become a successful P2P exchange. Perhaps a more interesting name would be beneficial later on when people wonder what it is.

Thinking about marketing and names I suggest to call it Nipple somewhat similar to GNU

Nipple - Nipple is not a [Ri]pple

EDIT: or more simple version

Not a [R]ipple

[1714920692]

1714920692

[1714920692]

1714920692

[1714920692]

1714920692

[fellowtraveler]

For those asking for hand-holding....

Code:

git clone git://github.com/FellowTraveler/Open-Transactions
cd Open-Transactions

See the docs folder for instructions on installing the pre-requisites.

Once you have those installed, this is usually what I do to build:

Code:

autoreconf -vif -Wall
./configure --with-java

(The reason I configure with java is so I can run the test-GUI.)

Then build the project:

Code:

make
make install

You can also do "make -j2" or -j8 etc to make it build faster, depending on how many CPUs you have (and how much RAM.)

There is also now a tarball installable through apt-get though I don't know if it's the most recent version (better to build it yourself, because then you can always "git pull && make install" to update to the latest version.)

Once you have it installed, I suggest you start by copying over the test data:

Code:

mkdir ~/.ot
cp -R Open-Transactions/sample-data/ot-sample-data/* ~/.ot

Then start up the server:

Code:

otserver

Then you should be able to use the command line tool (in a separate window):

Code:

opentxs help
opentxs list
opentxs stat
opentxs shownyms
opentxs newnym

Etc. You can also make scripts by putting "!/usr/bin/env ot" at the top of the script. (See the scripts/samples folder.)

If you want to try to OT test GUI (Moneychanger):

Code:

java -jar Moneychanger

(You will probably have to tell it the /usr/local/lib folder, and then select an image from your harddrive.)

People ask why they have to select an image. The reason is so that hackers cannot do phishing attacks by tricking you with a fake passphrase dialog. They will have no way of anticipating which image you have selected to appear on that dialog, so it will look different for every user.

I should add that the test GUI is not representative of an actual OT GUI. It's just a visual representation of the API, for testing/developing purposes. To see what an actual GUI would look like, check out the screenshots of the upcoming Monetas iPhone app.

You can get support on the OT IRC channel, #opentransactions at irc.freenode.net

Some people have contacted me asking about investing. I suggest you contact our CEO Johann Gevers: johann@monetas.net

[Kazimir]

Sounds very exciting. But I don't fully understand it yet, I tried to read some material on Open Transactions and BitMessage but still missing the point. Can anyone summarize what it is that OT and BM do, exactly?

OT is a system for federated issuing and transacting of various currencies. It can work with and without blockchain-based currencies.

Bitmessage is a p2p system for messages as well as subscriptions/broadcasts.

In our proposed protocol, OT federated servers will utilize Bitmessage as a discovery layer to enable cross-server escrow, for the purposes of market trading and wiring of funds between OT servers (and in/out of the legacy banking system.)

Thanks a lot man, much appreciated!

As a non-native English speaker, my interpretation of 'federated' is somewhat ambiguous, would that be something like 'based on a universally agreed upon protocol' or something else?

And how do you mean 'discovery' in this context: finding eachother i.e. bringing people (or the services they request and provide) together?

[marcus_of_augustus]

Guys, you do realize that we are building here the first true AI, an agent based one, don't you? If Bitcoin is not a Singularity it is a prerequisite for one.

This particular rabbit hole could be very deep...

Interesting, hadn't thought of that dimension ... but it indeed may take on a 'mind' of its own depending on the bots that will come to inhabit the various asset pair channels ... like an AI springing forth from the emergent behaviour of the collective of bots you mean?


Edit: ... and in this vein I propose the name for the OSS bundle comprised of Bitcoin Bitmessage Open-Transactions be the BITBOT bundle  ... and same name "BitBOT" for any AI that might spring forth 

[FreddyFender]

Guys, you do realize that we are building here the first true AI, an agent based one, don't you? If Bitcoin is not a Singularity it is a prerequisite for one.

This particular rabbit hole could be very deep...

Interesting, hadn't thought of that dimension ... but it indeed may take on a 'mind' of its own depending on the bots that will come to inhabit the various asset pair channels ... like an AI springing forth from the emergent behaviour of the collective of bots you mean?

Herds next to watering holes(BMOT) as opposed to wild jungles(free-for-all exchanges *GOX). WoT is powerful and unknown, just look what Bitcoin did to financial tools in a short time - now imagine contract law, commodities, documents, voting...
We will build the trading bots to protect our wealth that can act in unison or pick and choose which bots best appeal to our current risk/reward. OT is the Wilderbeast Consortium and BM is the trumpet call.

[johnyj]

I don't think a lot of people realise this but localbitcoins.com is actually a centralised server model ... and does anyone know how much data they retain about all the recorded transactions/messaging that passes through their server? ... wonder which jurisdiction/locale the localbitcoins.com server is located in and what their privacy policy is by the way?

Correct, the next step is to make it a distributed P2P application. And there will be some dynamic changing local agencies provide trading with enough volume

The benefit for this approach is: Like in a private auktion, if Alice do a bank transfer to Bob and buy some bitcoin from him, banks don't know the reason behind that transaction, the privacy is kept very well. If you do the transaction through an escrow, then banks will know clearly this money is going into a bitcoin exchange, at least that escrow holds all the information about all the transactions related to exchange

[Piper67]

Just like a gateway in ripple, this is the point of weakness for all the P2P exchange design: That escrow/gateway must be able to communicate with banks through an authorized channel and that channel is controlled by banks

Yes, as long as you have to communicate with legacy bank systems, you will have to follow their rules and regulations KYC etc...
Do not expect this to make proper exchanges obsolete all of the sudden. You need to take the commerce completely out of legacy bank system to not depended on whims of the banks. Or we could just buy a few banks... and have them run some OT servers.

Vladimir, you can be accused of many things, but a failure of the imagination is never one of them.

I just wish someone would give this OT/BM idea as much air time as Ripple is getting.

[johnyj]

* Solution for the Problem with interfacing into the legacy banking system: *

We simply create our own banking system outside of the legacy banking system bypassing old banks all together.

We as individuals create out own mini-p2p-banks with a small or large safe at home to store cash/silver/gold/whatever.
We build trust by having trusted individuals inspect/audit the mini-banks cash holdings and security on a regular basis.

Each mini-p2p-bank builds security and trust scores as independent private trusted auditors inspect their "banks".

It would be smart to store your money at different banks to protect your wealth against a bank crash.
The mini-p2p-banks handle money transports between themselves.

This is not possible, if you take all of your money out of banks and store them locally, banks will all go bankrupt  

[CIYAM]

This is not possible, if you take all of your money out of banks and store them locally, banks will all go bankrupt  

Well actually due to Fractional Reserve banking all banks are "by their very definition" bankrupt.

Like others I am very much looking forward to seeing where this leads.

[lunarboy]

 ..... This is the sound of Inevitability. 

[marcus_of_augustus]

I don't think a lot of people realise this but localbitcoins.com is actually a centralised server model ... and does anyone know how much data they retain about all the recorded transactions/messaging that passes through their server? ... wonder which jurisdiction/locale the localbitcoins.com server is located in and what their privacy policy is by the way?

Correct, the next step is to make it a distributed P2P application. And there will be some dynamic changing local agencies provide trading with enough volume

The benefit for this approach is: Like in a private auktion, if Alice do a bank transfer to Bob and buy some bitcoin from him, banks don't know the reason behind that transaction, the privacy is kept very well. If you do the transaction through an escrow, then banks will know clearly this money is going into a bitcoin exchange, at least that escrow holds all the information about all the transactions related to exchange

The escrow is performed by the smart contracts function on the OT server  ... https://github.com/FellowTraveler/Open-Transactions/tree/master/scripts/smartcontracts/escrow how would the bank have any idea it was not just regular user making a transfer?

[johnyj]

I don't think a lot of people realise this but localbitcoins.com is actually a centralised server model ... and does anyone know how much data they retain about all the recorded transactions/messaging that passes through their server? ... wonder which jurisdiction/locale the localbitcoins.com server is located in and what their privacy policy is by the way?

Correct, the next step is to make it a distributed P2P application. And there will be some dynamic changing local agencies provide trading with enough volume

The benefit for this approach is: Like in a private auktion, if Alice do a bank transfer to Bob and buy some bitcoin from him, banks don't know the reason behind that transaction, the privacy is kept very well. If you do the transaction through an escrow, then banks will know clearly this money is going into a bitcoin exchange, at least that escrow holds all the information about all the transactions related to exchange

The escrow is performed by the smart contracts function on the OT server  ... https://github.com/FellowTraveler/Open-Transactions/tree/master/scripts/smartcontracts/escrow how would the bank have any idea it was not just regular user making a transfer?

That OT server must be authorised by banks to access their SEPA transaction data, without that authorization it can only prove a bitcoin transaction, not a fiat money transaction

[marcus_of_augustus]

..... This is the sound of Inevitability. 

http://www.youtube.com/watch?v=x5m1A7zoIcc

Hear that Mr. Andersen ....

[caveden]

Any colored-coin issuer can fully-obey KYC / AML.

Are you sure about it?
Something tells me that the simple fact that a colored coin issuer can't control where his coins go to already makes it impossible for them to implement such restrictions on most jurisdictions.
Wasn't that the very reason ecash was dropped by banks?

But suppose it's legal. If I'm getting this right (and that's a big if, I confess I need to study this much more), in this new scenario MtGox-like institutions would emit colored-coins which could then be exchanged against normal bitcoins in a p2p, hard-to-track way. But to get fiat in and out, you'd have to go to your fiat-holder and then everything would be tracked as it is today. You'd still need to trust your fiat-holder not to commit fraud or disappear with your money, but, for sure, that'd be a harder coup to play than disappearing with your bitcoins.

So, we would no longer need to worry with "trade engine lags" and alike, and we would no longer need to store our BTC in a shared wallet. Any other big advantage?

Question: Wouldn't a p2p marketplace for escrows which do not hold your fiat be even better than that? The escrows only need to intervene when there's a dispute. But traders are still responsible for transferring funds to one-another, as in OTC. BTC-funds would be locked in 2-of-3 addresses, waiting for the fiat transfer to complete. No "bank-account to hold them all". If you have an API to access your bank account (think merchants) you can even automate everything, as the look-up for escrows could be based on deterministic criteria.

[marcus_of_augustus]

I don't think a lot of people realise this but localbitcoins.com is actually a centralised server model ... and does anyone know how much data they retain about all the recorded transactions/messaging that passes through their server? ... wonder which jurisdiction/locale the localbitcoins.com server is located in and what their privacy policy is by the way?

Correct, the next step is to make it a distributed P2P application. And there will be some dynamic changing local agencies provide trading with enough volume

The benefit for this approach is: Like in a private auktion, if Alice do a bank transfer to Bob and buy some bitcoin from him, banks don't know the reason behind that transaction, the privacy is kept very well. If you do the transaction through an escrow, then banks will know clearly this money is going into a bitcoin exchange, at least that escrow holds all the information about all the transactions related to exchange

The escrow is performed by the smart contracts function on the OT server  ... https://github.com/FellowTraveler/Open-Transactions/tree/master/scripts/smartcontracts/escrow how would the bank have any idea it was not just regular user making a transfer?

That OT server must be authorised by banks to access their SEPA transaction data, without that authorization it can only prove a bitcoin transaction, not a fiat money transaction

You changed from "banks will know clearly this money is going into a bitcoin exchange" to "OT server must be authorised by banks to access their SEPA transaction data" ... which are obviously two different issues. Interfaces with legacy banking system are problematic but they can't stop individuals making transfers or their money would become more useless than it already is ... the more barriers they put up to personal transfers the less valuable their money becomes ...

[Luckybit]

Uh... I don't see the distributed marketplace. (As most here, I'm still trying to wrap my head around it.)

You would still need to
1. Obtain some kind of IOU (DGC)
2. Trade it at a particular place. (OT server)

From my understanding of FellowTraveler's initial post, BM is used for matching Bids and Asks by creating an exchange protocol on top of BM that interested parties can subscribe to.

The only problem is this doesn't sound very user friendly. How can it all be turned into a simple web app?

[BeeCoin]

I can also reveal that Adam Back believes he has solved the problem of homomorphic amounts (so the OT server can't see any of the amounts, on any of the transactions it processes.)
I will be integrating credlib into OT soon and also working with Adam on adding his homomorphic code to OT as well.

Wow... this is big!!

[Anenome5]

Last night, randy-waterhouse and I were experimenting with Bitmessage. (*smooch*!!)

Bitmessage is a p2p messaging (and broadcast / subscription) protocol, based on the Bitcoin protocol.

It uses its own blockchain, but the chain only stores the last 2 or 3 days worth of messages. (It's assumed they were delivered within that time, where they are then safely stored on the recipient's inbox.)

Combining the above Bitmessage capabilities--which we already proved out experimentally--with Open-Transactions, makes possible fully-decentralized p2p markets, as well as p2p escrow across OT federated servers, easy p2p and server-to-server wiring of funds and conversion of currencies, both within OT and also between OT and the conventional banking system.

Furthermore, this is possible with little-to-no changes inside OT itself, and will not require the issuing of credit, nor will it require any pre-mined currency.

How does it work?

-----------------------------------------------------------

A few concepts...

--- First, keep in mind the concept that Bitcoins and Colored Coins (either/both) could be issued onto an OT server, without having to trust the server itself, through the use the multi-sig "voting pools" on the blockchain itself. I've already extensively discussed this on this board, and here's an article on how it's done:  http://bitcoin.stackexchange.com/a/834/309

--- Second, keep in mind that using Colored Coins instead of Bitcoins is advantageous in certain circumstances, as it allows users to buy/sell those colored coins (for the purposes of transmitting other currencies) without incurring any capital gains tax liability. (I'm not a lawyer and that's not legal advice. The basic gist is, if you buy a colored coin for $100, and sell it for $100, there is obviously no gain or loss.)

-----------------------------------------------------------

T H E   H O L Y   G R A I L

Enter Bitmessage! (Which solves discovery across federated OT servers.)

As I said, randy-waterhouse and I already TESTED Bitmessage last night to prove experimentally that this is possible (and it worked.)

-----------------------------------------------------------

Using Bitmessage with OT to effect server-to-server wiring of funds: http://pastebin.com/NjQgDarx

--- The wiring protocol is all about Alice trying to discover Bob so she can move her money from one server to another (and Bob trying to discover Alice so he can make a profit by moving money from one server to another.)

-----------------------------------------------------------

Using Bitmessage with OT to effect escrow-based conversion of currencies across OT federated servers:  http://pastebin.com/S1W5guAQ

--- The currency conversion protocol is about Alice and Bob being able to choose a server they can agree to meet on so they can trade one currency for another inside OT. (For cases where they aren't already trading on the same OT server.)

-----------------------------------------------------------

Using Bitmessage with OT and SEPA so that Alice can p2p send any currency which Bob receives as Euros in his Euro account: http://pastebin.com/SsLrxVP6

--- The SEPA transfer protocol is about Alice being able to send Silver Grams, which Bob receives as Euros in his Euro bank account. It's also about Jorg earning a profit in silver grams, by sending a SEPA transfer to Bob on Alice's behalf.

-----------------------------------------------------------

We already knew that OT offered quite a few benefits to Bitcoin: http://bitcoin.stackexchange.com/a/2710/309

But now, combined with Bitmessage, Open-Transactions becomes a juggernaut!

The above protocols can be implemented inside OT wallet GUIs, such that they are automated and transparent to users.

May a million currencies bloom!

Very interesting! Great work you're doing. I've got to read up on all this; I think this work might benefit something I've been working on myself

[Meatpile]

Bitmessage sounds a bit unscalable, the white paper says you are attempting to decrypt every single message that comes in over the wire with all of your local keys... That would quickly get wasteful and cpu intensive

[mmeijeri]

Read the paper, it's addressed.