Skip to toolbar

Community & Business Groups

A Web of Ledgers

Twenty five years ago the Web changed how we share information. Today anyone with Internet access can access trillions of documents instantly and for free. But the same has not happened for payments. It is still cheaper and faster in many cases to send a bundle of cash via Fedex than to send an international wire. In order to change that, it’s important to learn from the experience of the Web.

What made the Web – and the Internet for that matter – so powerful, was the fact that it allowed systems to offer interoperability while placing minimal requirements on them. Interoperability enabled competition because even smaller providers could reach a global audience. And the technical minimalism of the Web platform resulted in tremendous diversity.

Can we do the same for payments?

Payments are different

Some would argue that payments are different than information. Information can be copied, money must not be. But we believe that the end result – a truly frictionless and competitive world – can still be achieved by focusing on the same core principles: interoperability and technical minimalism.

This week we published the first draft white paper of a new protocol for payments across payment systems. In the same way that HTTP has enabled documentation systems to interoperate, we hope that the Interledger Protocol (ILP) will do the same for payment systems.

What could ILP be used for?

ILP will make payments global, instant and free. Aside from the direct benefits, lowering transaction costs by orders of magnitude often enables entirely new use cases. When we think of the internet we don’t think about how much postage we’re saving, but all the new services that it has made possible.

The Web today, it’s full of crutches – business models which have evolved as workarounds for the lack of an efficient payment system. Whether it’s annoying ads, frustrating paywalls or gatekeepers with outrageous revenue share, these are artifacts of the underlying limitations.

If my browser requests a file from a server and there is a tiny cost associated with the licensing and transmission of this file – why can’t my browser send a tiny amount of money to cover that? Web browsers already govern other precious resources – CPU usage, memory, screen space and prevent websites from using too much. Extending that model to money is daunting but by no means impossible.

Do you have ideas about the Internet of Value and how it would change the Web? Post them in the comments below! And watch this blog where you’ll find our thoughts, demos and hacks around the IoV idea.

How does it work?

At its core, the Web is a giant global graph of hypertext documents connected by hyperlinks. ILP models the world of finance as a giant global graph of ledgers connected by liquidity. The systems providing this liquidity are called connectors and a key feature of ILP is that these connectors do not need to be trusted, meaning anyone can create one.

Although the idea of a financial graph of liquidity is not new, this groups aims to build it according to the principles of the web, without requiring a global stateful system, such as a blockchain or central operator. In other words it is a protocol in the true sense. ILP is designed for international payments, but it can just as easily be used to make a payment on a local network between two ledgers with no internet connection.

For more details, check out our technical white paper.

What is a ledger?

A ledger is a system which tracks the ownership of one particular type of asset. You can think of it simply as a list of accounts and balances. Banks operate ledgers as does PayPal. There are even distributed ledgers such as Bitcoin.

It’s important to note that ledgers aren’t just about money. Ledgers can track the ownership of any asset or liability. This might include stocks, bonds, commodities or even discrete assets, such as artwork or real estate.

A matter of trust

How can it be possible that connectors don’t need to be trusted? After all, they are moving money!

ILP uses a simple form of cryptographic escrow, which ensures that connectors only get paid if they’ve fulfilled their obligation first. This eliminates the risk that a liquidity provider might default, sometimes called Herstatt risk, after a German bank that defaulted on its payment obligations when it went bankrupt in 1974.

Because of this risk, large banks today often use a method called payment vs payment through services like CLS, which is exactly what ILP does as well.

A key feature in ILP is the ability to chain these escrow transfers together. This is what turns it from a neat way to connect ledgers into a giant global graph of liquidity.

Moving to an ILP enabled world

In order to participate in ILP, a ledger only needs one very simple feature: The ability to do escrowed transfers. Bitcoin and most modern distributed ledgers already support this. For others, it is very easy to add and can even be provided by a third party. Importantly, ledgers do not need to speak ILP. They can continue to use their current protocols, be it ISO 20022 or Bitcoin. They can even continue to use any transport, be it SWIFT, TCP/IP or carrier pigeon. The connectors smooth over these differences.

The community group

There is a lot of work to be done in order to make the Interledger idea a reality. We believe that we can succeed only if we bring together a wide variety of stakeholders.

Aside from the obvious parallels between existing Web protocols and ILP, we believe that the W3C through its lightweight open processes has uniquely proven to be a standards body that is very accessible for individuals while still being able to interface with the industries its work affects.

We hope that we were able to give you a sense of the magnitude of the change that the financial industry is currently undergoing. If you would like to be a part of that change, please join the Interledger Payments Community Group and help us to develop the standards that ledgers around the world can adopt to enable ILP.

4 Responses to A Web of Ledgers

  • what exactly is the change that a bank will need to do if it were to adapt to inter ledger? What if the counterparty banks do not change?

    I work as a financial services software consultant; am excited about the possibility


  • In general all a ledger needs to do to participate in Interledger transactions is enable cryptographic escrow.

    What that means is that it needs to offer escrowed payments, where the release (or return) of the escrowed funds can be triggered programmatically using a cryptographic condition. To learn more, please check out the whitepaper

    I’m also working on a blog post to be published soon that goes into more detail about the cryptographic conditions.


  • Hi Stefan,

    Locally executed but globally replicated deterministic transactions proved themselves as better alternative to globally distributed transactions coordinated with a two-phase-commit protocol.

    The protocol is not resilient to all possible failures and cannot guaranteed reconciliation. When one of its assumptions (that nodes do not experience stop or Byzantine faults, data in the write-ahead log is never lost or corrupted, and nodes can freely communicate with each other) turns out incorrect, a human intervention may be needed to remedy the consequences.

    The Protocol for Interledger Payments apparently aims for the best of both alternatives. Perhaps its only weakness is the inability to uphold the main idea behind the concept of transactions – to make the consistency of performance causally independent from anything that might go wrong within the tolerated limit per ledger.

    I visualize the future Internet-of-Value as a loose federation of lots of smaller permissioned ledgers, each operating with individual stateless consensus protocol, rather than as an interoperability between few large ledgers, each operating with a stateful consensus.

    In that federation, liveness of a ledger is not causally dependent on a stop-faulty computer or partitioned network link and the interledger transactions provide guarantees that safety of a ledger cannot be spoiled by one or two stop-faulty nodes of another ledger.

    Despite the differences, I appreciate the Protocol for Interledger Payments as an important conceptual milestone. Please, get in touch with me if I can contribute to its further tuning or implementation.


  • Imagine a time in the not too distant future where you pay for your lunch with Apple stock. Or the football ticket with XRP. Or your friend desperately needs ETH but you have only XRP. Just pay him in XRP and he’ll receive ETH. ILP will do the rest


Leave a Reply

Your email address will not be published. Required fields are marked *

Before you comment here, note that this forum is moderated and your IP address is sent to Akismet, the plugin we use to mitigate spam comments.