Thanks cjp and blueadapt, your ideas look really cool. I will get into them and get back. Thanks!

Thanks a lot vancsj.

Complexity is the hallmark of an unfinished idea, which this one certainly is. Hope, a discussion will allow me to spot the elements that can be taken away.

Thanks Bitsalame, you could be correct. It is only by being wrong that I have a hope of eventually being right.

It would be good to have a concrete analysis from you, especially since the proposal does away with a Clearing House for securities trading. There is a difference between Settlement and Clearing House.

Thanks for the kind words BillBags. I understand your perspective; I am a large holder of bitcoins and events like the current price decline don't worry me at all.

That said, I think there are 2 parallel movements about to happen: One is Bitcoin, and the other is the rise of a new type of Fiat banking. Call it crypto-banking if you must. Ripple kickstarted this movement and Stellar is a strong follow up. I am doing a thought experiment to explore if there is a different way of building Ripple. It would be great to have technical views from the excellent Bitcoin folk to help with the thought experiment.

Ripple is as much a challenge to the current banks as is Bitcoin. That is because it allows small startup banks to become competitive with big banks. It is a system by which lots of small fish can collaborate to challenge the big fish eventually. The genius lies in the Settlement mechanisms which reduce liquidity needs of small banks.

Hope you understand my perspective.

@Billbags: Thanks for the questions.

P2P: System is a mixture of P2P and non-P2P. P2P when exchanging assets and paying another person with an account at the same bank. Non-P2P when payments cross the boundary of an individual bank.

Mining: There is no mining.

Banker and Government ran: Banks can issue as many assets as dictated by daily operations and reserves. The ledgers are maintained in a decentralized fashion. This kind of decentralization is very useful in mediating interactions between banks, and the article explains why.

Basic ACH 2.0: Yes, it is supposed to be ACH 2.0 + Stock Market 2.0 + FedWire 2.0 + Contracts 2.0.

The paper considers the following question: What system can we design if financial institutions (banks plus others) tracked asset and liability balances using public decentralized cryptographic ledgers with inbuilt scripting languages?

Mike Hearn's initial wiki article for Micropayment channels is unidirectional, i.e. the payment can flow only from one party to another and not in the reverse direction.

Example 7, https://en.bitcoin.it/wiki/Contracts

Is there a protocol that allows micropayments to flow in both directions? The solution of creating 2 channels with Bitcoins locked up by both parties is obviously valid, and I am looking for something different.

Is there a way by which, only one party locks up bitcoins, opens up a channel with another party, and the micropayments can flow in both directions?

Here is proposed an alternative architecture called "The Internet of Money (IoM)" to build a Ripple like system with the following differences:

1. No in-built math based currency.

2. No requirement for a singular consensus process. Gateways use a decentralized consensus pool, outside their control, to run their ledger. We replace a singular consensus process (Ripple) with multiple ones, and make up for the disadvantage at higher protocol layers. This simplifies the Ripple consensus engineering challenge.

Reading the Bitcoin technical forum with its numerous resident geniuses has been the best educational experience of my life. I welcome everyone to post comments, questions and point out mistakes / errata. Please bear in mind that the document is an early draft.

Current document version is 0.2.9, please discard the old versions.

Version published through google docs: https://docs.google.com/document/d/1Bc-kZXROTeMzG6AvH7rrTrUy24UwHoEcgiL7ALHMO0A/pub

An aesthetically better pdf version: http://issuu.com/meherroy/docs/an_architecture_for_the_internet_of/01
Pdf works best with Chrome.

A third backup, suitable to download the file: https://www.scribd.com/doc/241975638/An-Architecture-for-the-Internet-of-Money

Comprehensively crediting work that came previously is always a challenge; omissions (if any) should be treated as arising out of ignorance.

Hyperledger thread for the same: http://forum.hyperledger.com/t/an-architecture-for-the-internet-of-money/157