The Story of Micropayment

The marketplace with the largest potential in the world is the World Wide Web. Billions of web pages linking to trillions of web resources like pictures, text, videos and physical objects (IoT), make this network the ultimate place to make money; for both individuals and companies.

Just to make one clear statement: – Trading data is not supported by the web infrastructure today.

Looking at the nature of web, the hyperlinks is what connect web resources between web pages. The web has the property of collecting data from many underlying sources from various domains. One web page can have links to many other web pages. These links provide a dynamic content giving a personalized value to the person who consume the data in the end. This is the decentralized nature of the web; and also the feature that makes web so special that it has become the biggest sharing network in the world.

However, a payment system of today does not work on the decentralized principles of sharing data since the web protocol (HTTP) is a stateless protocol. This means that the protocol does not «want» to keep the state of the connection alive while payment processing is taking place (typical when ID’s and stuff are being verified). Today’s payment systems only work on one domain content at a time, for instance against a web server that keeps the connection alive while the payment is being processed.

So for a specific domain, one can offer to sell a product or service using a payment system like PayPal or Stripe for the money transfer. But for a decentralized web, where the links address external resources outside the original domain, the payment system does not work. It is not possible to shop around on the web and pay as you go. Also the web protocol doesn’t “want” to become stateful and facilitate for this type of payment process; This is due to the network design in keeping the web heterogeneous. The web needs to be heterogeneous – it is the essence of the web and what makes the web a global phenomenon.

The Holy Grail Tech challenge

There are thousands of networks in the world but only one super popular web network – the world wide web. It’s popularity has everything to do with the fact that the network is decentralized with a stateless interface across domains; so each service provider can somehow keep their IP and create value to the data they share among each others. Removing this property the web becomes an unscalable single entity.

So the questions everybody asks themselves; can we do payment across domains without ruining the nature of the web? Can we trade data across multiple stateless domains so the value of the data in the linking path is kept, and at the same time do settlement between domain owners and data consumers?

Several architectural projects are targeting these questions. And once answered the ultimate Holy Grail of data transaction and network design will be found.

Researching distributed technologies

For many years we have been researching distributed technologies. Among these Blockchain has been a natural architecture to look into.

Blockchain is a decentralized networking architecture that targets to solve the Holy Grail challenge. Our plan was simple; to target the endpoints in the network, and mirror the linking path distribution to a distributed ledger. Doing this it should be possible to trace the distribution and process payment among participant that is part of the transaction. In theory this should work. But in practice this does not.

The problem we faced was that the latency for the verification process incased for each payment made. Even if the process was mirrored into shared memory containers for concurrent processing, the latency was still to high and increasing by volume. Trying to solve this by generating token reserves (token pool) the delay was still major using programmable interfaces (APIs). Additionally, maintaining these APIs for each new service that wants to be part of the network is impractical and costly in the long run.

One of the main disadvantages of using APIs to communicate with another service provider is that you have to know the data structure in advance. This limits the possibility to communicate with a future service. For each new service a new implementation is needed. This is not scalable and maintenance is a nightmare. So even if one wants to embed Blockchain technology to handle transactions between web pages, the main problem is how the communication is done and how much it can scale with low maintenance cost.

Lesson learned; The only way to solve the API dependency is to solve the communication at a lower stack level.

Solving the Holy Grail challenge

Once you manage to solve the lower stack level communication, Blockchain and web linking networks are closer to a ultimate networking experience. We call this networking experience the Real Time Web; a new upgraded Linked Data network that follows the web principles of sharing data across domains and at the same time is able to offer data trading following links across the global web network; a 100% heterogeneous decentralized networking system.

Real Time Web is a new distributed web network

The network is enhanced with several new native properties. Here is a list of some advantages:

1. The linking network has become full stack event driven. Distribution is based on data that changes states, no lookup or request is required. This means that no database is necessary to bridge the communication.
2. P2P communication can be achieved between any web resources globally. Full stack also enables control of the endpoint without interfering with the data between endpoints; so privacy can be kept for the consumers regardless of which domain (service provider) they are using to initialize the communication.
3. The predicate or software agent becomes part of the data transformation related to the semantics of the service. This means that the Linked Data transformation can be done decentralized where the domain owner or service provider can keep their IP and business logic intact and does not need to expose it to the consumer.
4. The linking graph can dynamically be altered and in real time change the transformation of data in the network without reconnecting or revalidating clients.
5. There is no middleware processing.
6. Physical resources and IoT objects are supported globally and ownership of web resource instances can be transferred in the network at runtime.
7. Scales like the web.
8. Almost zero latency (limited by the networking connectivity and networking routers only).
9. Low networking footprint in the higher stack level reducing the energy consumption in the network by ten fold.
10. Future software development can be done autonomous and controlled by AI services. Reducing the time to develop software services to a minimum.