Perspectives, like Blockchain, opens up a market offering a wealth of economic opportunities. By freeing up user data from the privately owned tech giant silos, innovation is boosted. Perspectives is like an open protocol: no vendor lock-in. By using a witnessing scheme, monetary transactions can be very safe, even on a distributed network. We show how some services can be metered, even though Perspectives adheres to the Bring Your Own Device philosophy.

Fat and thin protocols

Perspectives is an open system for co-operation. It is neutral as to what that co-operation is about. A large part of human interaction consists of economical activity. As Kate Raworth argues, classical economists are a little parochial in that respect. She argues convincingly that the households and commons should be included in economic science, alongside government and business. But in a more narrow sense, economy is about creating wealth and making money. In this article we explain how Perspectives fits in with economy; what opportunities it offers and how it compares to the current web in that respect.

Joel Monegro of Union Square Ventures (USV) coined the name ‘Fat Protocols‘. In a post on their blog he explains:

The previous generation of shared protocols (TCP/IP, HTTP, SMTP, etc.) produced immeasurable amounts of value, but most of it got captured and re-aggregated on top at the applications layer, largely in the form of data (think Google, Facebook and so on).

He therefore calls HTTP etc. ‘thin protocols’. In contrast, he argues, BlockChain could be described as ‘fat protocol’. Most of the value that is owned is generated ‘at the protocol level’. Bitcoin and Ethereum have a very large market capitalisation, while applications built on top of their respective chains have far less value. Is Blockchain a protocol? This may be a stretch, but it is arguably more a protocol than an application with end user functionality.

This difference between thin and fat has been explained in terms of where state is kept. The underlying web protocols are stateless, so state must be kept in the programs that use the protocol. In contrast, Blockchain is all about state. It is in essence a distributed database. So we see that, for both fat and thin protocols, value accumulates where state is kept. The current web is dominated by the client-server architecture. State is primarily kept by the servers. By now we all know that the companies owning the servers do not just keep state to make the processes run. They hoard any bit of data that can be gleaned from their users’ behaviour and make huge profits out of it.

But even though some Blockchain companies represent enormous value, they have not acquired that value because they own data. Instead, their value is closely associated with the ‘tokens’ that are issued by these companies. Much of this token value is based on speculation. However, speculation starts with expected future use, which in turn rests on the fact that these tokens are used to pay for services provided by the Blockchain companies. Take Ethereum as an example. It is in essence a deployment service: one can run a program (a ‘smart contract’) on it. But this service is metered and one pays in terms of the ‘ethers’ issued by Ethereum.

A limited number of tokens is issued. This makes them a little like securities and their relative scarcity promotes their speculative value. It is a clever construction that enables innovative companies to acquire capital and at the same time creates a market for innovative products on top of the Blockchains. Except that, as Monegro notices, companies do not make a lot of money out of such products at the top level. This may, in the end, prove to be an Achilles heel for the entire Blockchain ecosystem.

When we look at the development of the current tech giants we see in the adoption of their services over time the familiar S-curve. Early on that curve, these companies seek co-operation with others, but at the high end they elbow out competitors or cannibalise them and thereby stifle innovation. The same may happen with the large Blockchains. This problem arises whenever a party acquires too much power over others, when monopolies are formed – and here, data is power, or deployment.

Perspectives as a Protocol

With Perspectives, we have an alternative model. To understand Perspectives for the purposes of this article, you have to make a leap of faith:

• a user will, as by magic, always have access to the information he needs to execute the actions that are available to him in his role in any relevant context – and just that information;
• that data will be stored locally on his computer and not on any server.

Perspectives is fully distributed: it is a Bring Your Own Device model of supported co-operation. These two characteristics guarantee a system with maximal protection of personal data. There is no central storage of information whatsoever, unless it is explicitly designed as such (as in the taxi example given below).

We can understand Perspectives as a protocol. It is a protocol for exchanging information between users in a peer to peer fashion, governed by models of co-operation. Each situation (‘context’) requires its own model and provides roles with actions. These models ‘execute’ on the Perspectives Distributed Runtime created by Perspect IT. It uses this protocol. However, other parties may develop their own competing runtimes (indeed, we give a suggestion for one, below for the witnessing scheme). Because the Perspectives Protocol is open, there is no vendor lock-in.

Blockchain has been hailed as a promotor of innovation, because “by replicating and storing user data across an open and decentralized network rather than individual applications controlling access to disparate silos of information, we reduce the barriers to entry for new players and create a more vibrant and competitive ecosystem of products and services on top” (Joel Monegro). This is true of Perspectives, too.

Monegro sees a second necessary component: “But an open network and a shared data layer alone are not not enough of an incentive to promote adoption. The second component, the protocol token (..) which is used to access the service provided by the network (..) fills that gap.” Tokens promote economical development by providing a monetary incentive.

Perspectives on money

Perspectives has no ‘token’. But it is opiniated as to what can be free and what should be paid for. That is, Perspect IT is opiniated and has constructed tools and software to make its particular vision come true. The protocol and the Perspectives Distributed Runtime are free, in the sense of ‘gratis’. But models are not. A model represents creative intellectual work and the creator should be compensated for that work.

This is achieved by including in the protocol the exchange of proof of payment of models. The PDR checks this proof on receiving an update from another PDR. No proof means the update is rejected. So a cheating user is excluded from interaction with bona fide peers.

A proof of payment is acquired at the repository where the PDR downloads a model. The repository issues such proofs, for example as the user buys a subscription. To pay for this subscription, the user’s account is debited. These accounts are nothing more than a context secured by independent witnesses. A user charges his account using an exchange service that provides regular payment options (credit card, Paypal, Ideal, etc.). The repository operator makes models available and credits the creators account.

So we have four parties participating in an economic fashion: users, creators, repository operators and exchange operators.

How do the two service operators make money? Interestingly, even though Perspectives is distributed (BYOD), a service operator can apply a form of metering for deployment. Consider the exchange operator. He will participate in contexts where users can charge their accounts and he will also participate in contexts with repository operators. This may amount to a lot of data, depending on the size of his business. But as the process is well understood and modelled out, we actually can construct a pretty good idea of the load a single customer puts on his infrastructure in terms of memory usage, cpu usage and bandwith usage. This forms a base for metering and fees. It also creates a competitive market.

Will people pay for models? By now, the public has become familiar with the idea that ‘free’ software and services on the web come with a price. It’s a hidden price. It is loss of privacy in the sense that when one deals with organisations, they will know infinitely more about you than you know about them. This may lose you opportunities for loans, jobs, memberships and housing; it may drive up the price of all sorts of goods and services. But there is a larger cost to society as the super capitalists in Silicon Valley evade taxes and thus do not contribute to the infrastructure in terms of roads, airfields, energy systems and educated employees that they make use of. In the end, the survival of a nation depends on its ability to tax the population – including businesses.

So yes, we think people are, more than ever, willing to pay for services, especially when they see they acquire real value for their money. Recently we have seen a rise in subscription models, as argued by Ev Williams, the CEO of Medium. Just think of Spotify, Medium, Netflix and others. People will pay for content and service.

So, even if Perspectives does not issue tokens, it supports financial accounts and creates a financial incentive for creators and the two service providers we have described above.

We’ve lightly touched the subject of ‘witnessing’. A financial account, stored as data on one’s own device, is an open invitation for deceit. How easy it would be to hack one’s PDR and increase his own credit! This is where witnesses come in. Any organisation or individual with a name to lose can participate in a transaction context as an independent witness. These contexts are created in such a way that witnesses form an independent communication channel about transactions. This makes it impossible to carry out such a hack an get away with it. But again, witnessing has a cost in terms of computing resources, so witnessing parties may meter this and require a fee for their service. Interestingly, such a party does not need a full distributed runtime. As long as a program uses the Perspectives Exchange Protocol and knows how to act according to the governing model, it can participate in the network. Thus, a company may create a program that is far more efficient than the all-round general runtime and still provide the witness service.

But there is more. For example, users may want to pay for an archiving service that backs up data they no longer need direct access to. Or they may want to pay for a deployment service that provides them with a PDR that is available 24/7 (their own laptops will be switched off every now and then. Bots defined in models – yes, it is possible – act only when a PDR is active).

Central services on a distributed network

Far more interesting is the economic case for centralised services. This may sound like a contradiction in terms, for a distributed system, but let me explain. Think of a taxi service (like Uber). In essence, this is a simple market where drivers offer their services to passengers. Passengers will put up a request for transportation and drivers may bid for them. Now obviously, this will not work unless there is a central market. Somewhere we need a node in the network where supply and demand meet. Notice, however, that just a little data is needed for the actual matching: just the essentials such as when, from where and to what destination.

A company could provide that service. Again, the service could be metered, this time including monetisation of the matching process. Notice how such a service would differ from, say, Uber: the company would have access to far less information about its passengers and drivers. Moreover, users would have all that information themselves, too. This would allow other creators to draw up a model that includes one’s taxi rides in an expense scheme, for example. But most of all, services like this might spring up in cities and areas independently. There is no benefit for a world wide operating company like Uber. What use is it for a passenger looking for a ride in Amsterdam to deal with a company that knows all drivers in New York? Uber’s service requires massive investments. But using Perspectives is a different matter entirely.

Pirates and freeloaders kept at bay

Now all this is good and proper as long as users behave. But what happens when a pirate decides to put the models he’s acquired on a repository of his own? This repository would give out proofs of payment while no payment was actually made. Such pirates can be banned by an organisation that requires repositories to be certified. The organisation would be ‘central’ (but not necessarily a world wide operator).

However, suppose a more clever pirate steps in, who forks the PDR (it will be Open Source) and throws out the payment proof checker? Now freeloaders can use this forked PDR and the pirate repository and they can interact without problem. True, but only amongst each other! The bona fide PDR’s will still require payment proofs. So now the community splits in two.

One might think that the freeloaders and pirates will win, will drive out the bona fide paying users. But we think different. This is because we think businesses will put up their models just with bona fide repositories. They also have good reason to crack down on pirates, just like the film and music industry did with the p2p filesharing pirates in the nineties. Even though p2p filesharing still exists, there is a vibrant music and film industry online. The same will happen with Perspectives models. The combination of fair prices, high quality and good service and user experience will make the freeloaders a fringe phenomenon.

Finally, think about discovery. A search service like Google could never be operated on a distributed netwerk. Or could it? To start with, there has not been an economic incentive to develop such a service. Secondly, some resources have a natural hierarchical index. Addresses are such resources (designed for the purpose). Actually, the taxi service we’ve described is another example. Resources that are distributed geographically and used or consumed locally, have such an index. More examples will be discovered as the case for distributed discovery becomes economically attractive.

Resources that can be indexed in such a fashion, lend themselves very well for distributed discovery. Think of a marketplace for second hand goods. Someone looking for, say, a lawn mower would like to search in terms of geographical proximity (rather drive 20 km to fetch a lawn mower than 100 km). If the search does not yield satisfying results, widen the area. And this is key. ‘Widening the area’ means using not just a locally deployed index, but geographical neighbouring services, too.

No doubt we’ll see a lot of ingenuity in coming years with respect to phenomena like this. A global search engine is convenient, certainly. It also promotes intellectual laziness. This wil change!