While users navigate some process supported by an application, it memorises what has happened so far: state. On the web, state is kept by the large companies, giving them the upper hand in dealing with customers. Perspectives restores that balance. We show how this does not preclude sound business on the web.
The Web is built on HTTP. It is a distributed protocol: anyone can publish a page (just set up a webserver) and everyone can reach that page (just use a webbrowser). It is also a stateless protocol. Consider the following simplified webshop interaction:
- A client expresses interest in an article.
- The webshop owner offers payment options (fully automated).
- The client pays.
- The webshop owner ships the article.
Now, suppose both parties have amnesia. They cannot form memories at all! Now the process falters at step 3, as the client will ask: what do I have to pay for? And, overcoming that, the webshop owner would simply not ship an article, as he has forgotten what the payment is for.
This is what ‘state’ means. Somehow, somewhere, a memory must be formed, a record kept, of how far the process has proceeded. In direct interaction in a physical shop, our personal memories – our brains – would do the job. So the question is: looking at the software parts that support the process of buying online, what parts do memorize? The answer cannot be: the software layer that implements HTTP – it is stateless.
In the current Web, memories are formed at the servers. The application layer owned by the companies stores all data that represents the state of interactions (well, actually a small part is stored in cookies in the clients browser). This means that, even though the protocol is decentralised, the state of the process built on that protocol is not. It is very much centralised and this imbalances the relation between client and company to the disruptive proportions we witness today with Amazon, Facebook and all the others. Control of the data layer enables monetisation. In other words, own the data layer and you’ll get rich.
This explains, partly, all the excitement about blockchain. It is a technology to decentralize the data layer. This opens up a vista where the monetisation that happens at the data layer could be democratised.
But this is also exactly what Perspectives is about. If you think about Perspectives as a protocol (Perspectives Exchange Protocol – PEP), you’ll see that
- it is built on HTTP (decentralised)
- it is used to store all data at the nodes, perfectly distributed according to use and ownership.
In fact, Perspectives goes far beyond Blockchain in that the latter merely duplicates the entire database, thereby ensuring it cannot be monopolised by a single party. Perspectives however, splits it up into individually owned and shared parts. This has been called extreme sharding: each user keeps his own stuff (and that what is shared by others).
The Perspectives Runtime (PR) is able to use PEP to that end, because it relies on the models in terms of contexts, roles, properties and actions. So the protocol is tied up with the way IT support is defined, modelled.
Monetisation and tokens
Albert Wenger of USV explains in this post how a decentralised data layer (as made possible by Blockchain) can be monetised using tokens. Such tokens, often called utility tokens, pay for the use of the deployment service that blockchains offer. Usage is inherently metered, in these services and that allows for precise payment in direct proportion to use.
Paradoxically, this is only possible because the infrastructure is logically centralised – however decentralised it may be, physically.
Perspectives is different. Data stored over the various nodes is logically coherent, but never centralised. Whereas a Blockchain is a deployment service, Perspectives is not: the motto is bring your own device. There is no node that has an overview of everything that is represented – very much different from Blockchain’s replicated databases.
So, having no tokens, can the decentralisation of the data layer that Perspectives brings, be monetised? How can companies or individuals create and retain value? The answer is in the models. Without models, the PEP is useless. What data should be sent to whom? This is why we have created a market for models, using a subscription model. Subscriber fees can be thought of as ‘tokens’, in this respect: they will use it to compensate model builders for their creative work. This process, too, is fully automated, as it is in Blockchains. Fees will be distributed per user according to some measure of use. This creates an incentive to come up with good models.
Separation of concerns
Value creation and retainment is a process in which four roles play a part:
- the creator of models who will receive compensation for his efforts;
- the end user who will pay for the creative produces he uses;
- the repository operator, running a website, who will be compensated for his costs;
- a subscription service operator, setting up payments options – who, again will be compensated for his work.
End users pay their subscription through an operator of their choice. Hereby they up their credit count, a running balance in terms of a (supra)national currency such as dollars or euros. They also create monthly reports of use.
End users affiliate with repository operators, usually a single operator. However, as it is the operators discretion to distribute some model or not, an end user may affiliate with more than one repository. Compare repositories to shops.
Repository operators play a role in such usage reports. They aggregate these reports in value accruements for creators and augment their balances accordingly.
Subscription service operators, finally, make it possible for creators to debet their balance and get paid in hard currency. As do users, creators can choose a subscription operator to their liking.
Notice how repositories do not handle hard currencies directly: this is strictly the domain of the subscription service operators. They may be likened to exchange offices.