hintjens/Draft intro take 4

## Draft intro take 4
The Internet Protocol dominates 100% of the networking market but this was not always inevitable.  In 1995 Microsoft launched Windows 95, a desktop product that had no Internet but instead used a proprietary network, the Microsoft Network (MSN).  The MSN and AOL walled gardens failed, as did every other attempt to capture the growing Internet in private networks.  Following Internet's move onto every computer, society, economics, and even politics have gone digital to an extent that was impossible to foresee even ten years ago.

Why did the Internet dominate?  It is owned by no-one, not patented, not subsidised, and represents no dominate market interest.  Unlike the GSM stack, Windows, Internet looks like an orphan technology that should have been easily quashed by private interests.

The answer is exactly in this lack of centralized ownership.  By 1995 Internet was already the standard for connecting the leading edge around the world.  With no patents or taxes on the technology, and a clear set of interoperable standards, competition on materials and services was extremely efficient, and prices fell perfectly towards zero in a friction-free curve.  Sending data by wired Internet is today ten or twenty orders of magnitude cheaper than sending it by mobile phone networks, which are heavily owned, patented, and taxed.

Starting in the late 1990s, the World Wide Web started to define the digital economy.  Built on a protocol for sharing "pages" of data, the Web rapidly spread as the cheapest and simplest way of publishing and thus sharing knowledge.  Again, HTTP forms a natural monopoly connecting browsers to servers.  Again, with no patents or taxes on the technology, and a clear set of interoperable standards, competition was efficient and prices of HTTP-related products and services fell towards zero.

Internet and the World Wide Web dominated as a natural monopolies owned by all their users, who received the full benefits of the competition between suppliers.  Suppliers also benefitted, reaching a wider market than ever before, and building new profitable businesses on top of the now commoditised and unprofitable stacks.  Again, the mobile phone networks provide a counter-example: no new businesses layered on top, and all benefits of the natural monopoly going to suppliers, not users.

As more and more of our economy goes digital, however, we are seeing the emergence of a new class of natural monopoly that mimics the mobile phone business more than it does the Internet or the World Wide Web.  That is the emergence of dominate centralized web service providers: Google, Facebook, Youtube, even Twitter.

These centralized service providers present dangers for privacy and for freedom of knowledge.  The power of these natural monopolies is clear: in 2009, Google's Android operating system for mobile phones, which is primarily a vehicle for delivering Google services (and ads) had 1.9% of the US smartphone market by Q2.  In 2010, 12 months later, it had 17% of the US market.  [!-- need to get reference for these numbers --]

But in terms of economics and growth, this centralization is even more dangerous.  There is no effective competition to dominant centralized services, except other less dominant alternatives.  There is standardization, no interoperability, and no layering of new businesses on top of the natural monopolies.  Where the dominant suppliers do offer new business layering, it is within captive walled gardens: the iPhone marketplace, the Android marketplace, and so on.

In this article we will argue that decentralized services are possible, as well as highly desirable.  We will argue that the technical difficulty of connecting service providers in a distributed fashion can be solved, and we will explain how we propose to solve it, using free and open standards, and scalability layers that connect service providers using those standards, and clear patterns for distributing data and work.  We will present examples of successful decentralized services, and then we will present 0MQ, which is a simple messaging library focussed on this goal.

0MQ demonstrates how large-scale distributed services can be built with basically zero friction.  It provides reusable patterns for service interoperability that are intended to become new free and open standards that will create new natural monopolies owned by their users, the benefits going to their users.

Finally, we discuss how the principles 0MQ is based on can be rephrased in terms of the Internet stack and what steps need to be taken to make the "scalability layer" a natural part of Internet.
	The Internet Protocol dominates 100% of the networking market but this was not always inevitable. In 1995 Microsoft launched Windows 95, a desktop product that had no Internet but instead used a proprietary network, the Microsoft Network (MSN). The MSN and AOL walled gardens failed, as did every other attempt to capture the growing Internet in private networks. Following Internet's move onto every computer, society, economics, and even politics have gone digital to an extent that was impossible to foresee even ten years ago.

	Why did the Internet dominate? It is owned by no-one, not patented, not subsidised, and represents no dominate market interest. Unlike the GSM stack, Windows, Internet looks like an orphan technology that should have been easily quashed by private interests.

	The answer is exactly in this lack of centralized ownership. By 1995 Internet was already the standard for connecting the leading edge around the world. With no patents or taxes on the technology, and a clear set of interoperable standards, competition on materials and services was extremely efficient, and prices fell perfectly towards zero in a friction-free curve. Sending data by wired Internet is today ten or twenty orders of magnitude cheaper than sending it by mobile phone networks, which are heavily owned, patented, and taxed.

	Starting in the late 1990s, the World Wide Web started to define the digital economy. Built on a protocol for sharing "pages" of data, the Web rapidly spread as the cheapest and simplest way of publishing and thus sharing knowledge. Again, HTTP forms a natural monopoly connecting browsers to servers. Again, with no patents or taxes on the technology, and a clear set of interoperable standards, competition was efficient and prices of HTTP-related products and services fell towards zero.

	Internet and the World Wide Web dominated as a natural monopolies owned by all their users, who received the full benefits of the competition between suppliers. Suppliers also benefitted, reaching a wider market than ever before, and building new profitable businesses on top of the now commoditised and unprofitable stacks. Again, the mobile phone networks provide a counter-example: no new businesses layered on top, and all benefits of the natural monopoly going to suppliers, not users.

	As more and more of our economy goes digital, however, we are seeing the emergence of a new class of natural monopoly that mimics the mobile phone business more than it does the Internet or the World Wide Web. That is the emergence of dominate centralized web service providers: Google, Facebook, Youtube, even Twitter.

	These centralized service providers present dangers for privacy and for freedom of knowledge. The power of these natural monopolies is clear: in 2009, Google's Android operating system for mobile phones, which is primarily a vehicle for delivering Google services (and ads) had 1.9% of the US smartphone market by Q2. In 2010, 12 months later, it had 17% of the US market. [!-- need to get reference for these numbers --]

	But in terms of economics and growth, this centralization is even more dangerous. There is no effective competition to dominant centralized services, except other less dominant alternatives. There is standardization, no interoperability, and no layering of new businesses on top of the natural monopolies. Where the dominant suppliers do offer new business layering, it is within captive walled gardens: the iPhone marketplace, the Android marketplace, and so on.

	In this article we will argue that decentralized services are possible, as well as highly desirable. We will argue that the technical difficulty of connecting service providers in a distributed fashion can be solved, and we will explain how we propose to solve it, using free and open standards, and scalability layers that connect service providers using those standards, and clear patterns for distributing data and work. We will present examples of successful decentralized services, and then we will present 0MQ, which is a simple messaging library focussed on this goal.

	0MQ demonstrates how large-scale distributed services can be built with basically zero friction. It provides reusable patterns for service interoperability that are intended to become new free and open standards that will create new natural monopolies owned by their users, the benefits going to their users.

	Finally, we discuss how the principles 0MQ is based on can be rephrased in terms of the Internet stack and what steps need to be taken to make the "scalability layer" a natural part of Internet.