The Web Renaissance
The Advent of the Dynamic Application Protocol Framework
New Introâ€”July 27, 2010
I wrote this essay in the summer of 2000. I have kept it hidden here and I rarely bring it up. In fact, I considered removing it from my web site because I was beginning to doubt my vision.
Recently I was introduced to the technology and company that has invented the Dynamic Application Protocol Framework I describe here. I was wrong about it being based on XML. It is based on the Kynetx Network Services (KNS). I will explain more shortly.
We are in a deep state of Web Noirâ€”technological Dark Age obscured by the apparent brilliance of the Internet, as we know it. The darkâ€”that noirâ€”is what we don’t see, what we don’t know because it doesn’t yet exist.
Whatâ€™s missing is technology infrastructure. Iâ€™m not talking about physical infrastructure here. Iâ€™m talking about the logical infrastructure where both humans and devices live and do their work. Itâ€™s the way weâ€™re all connected, and what we can do withâ€”and throughâ€”those connections. The real world of people and devices changes constantly. Natures, functions, identities, relationships all change. Yet we have few if any truly useful ways to support that dynamism beyond the store & forward facilities of Web and email servers running over a worldwide TCP/IP network. While what we have is a miracle-grade advance over what we knew a decade ago, itâ€™s still profoundly limited. In fact, itâ€™s so limited that in some cases the best we can do is leverage the worst from bygone ages.
Take the firewall. Nothing exposes the feudal nature of our enterprises more than the firewall: the modern expression of a city-state in constant hyper-vigilant siege modeâ€”a castle with stone walls, turrets, parapets, a moat and a drawbridge through a single entrance under a threatening portcullis. What could be more static and retro?
Or take our object models. None has ever answered the question of how anybody or anything can find any other object in a world full of multiple object models, each intended to get along only with its own kind. Microsoft’s latest â€œ.Netâ€ announcement is yet another in its continuing series of attempts to tempt and then crush the whole world into its own model. It never worked before, and it won’t work this time, either.
The dream of distributed computingâ€”a world where computing power is distributed everywhere it needs to be and works interoperably between anybody and anything–remains exactly that: a dream.
The Web Renaissance
But Web Noirâ€”our dark ageâ€”will end. We are headed into a Web Renaissance as radically different and significant as the European Renaissance that profoundly changed western science, medicine, the arts, government and everything else. What is needed is a significant shift in the availability of new and the innovation of existing network infrastructure. They key to the shift in infrastructureâ€”and the advent of the Web Renaissanceâ€”will occur because a significant new invention has appeared.
In the anthology The Greatest Inventions of the Past 2000 Years, (Simon and Schuster Edited by John Brockman) Bart Kosco nominates calculus as the most important invention:
â€œThe world today would be very different if the Greeks and not Newton/ Leibniz had invented-or discoveredâ€ calculus. Our world might have occurred a millennium or two earlier. Calculus was the real fruit of the Renaissance. It began when thinkers took a fresh look at infinity at the infinitely small rather than the infinitely large. And it led in one stroke to two great advances: it showed how to model change (the differential equation) and it showed how to find the best solution to a well defined problem (optimization). The first advance took math from static descriptions of the world to dynamic descriptions that allowed things to change or evolve in time. This is where â€œrocket scienceâ€ became a science. The second advance had an even more practical payoff. It showed how to minimize cost or maximize profit. Thomas Jefferson claimed to have used the calculus this way to design a more efficient plow…
Calculus lies at the heart of our modern world. Its equations led to the prediction of black holes and gravity waves. We built the first computers to run simpler calculus equations to predict where bombs would land. The recent evolution of calculus itself to the random version called stochastic calculus has allowed us to price the mysterious financial derivatives contracts that underlie the global economy: Calculus has led us from seeing the world as what Democritus called mere â€œatoms and voidâ€ to seeing the world as atoms that move in a void that moves.
Calculus was a breakthrough because it made the dynamic as real as the static. Newton and Leibniz formalized calculus to overcome static notions about relationships between objects that were anything but. Pre-calculus math models were simply incapable of sufficiently representing relationships between objects that move in space and time. As Kosko points out, the discovery of a model that was capable of modeling change (as well as other aspects of calculus) had a profound effect on all aspects of science.
What leaves us in Web Noir is the need for a dynamic application relationship model. The difference is that, in the Internetâ€”case, we donâ€™t need a dynamic mathematical model but rather a dynamic application protocol model. Today all Internet servicesâ€”file, print, directory, security, and so onâ€”are accessed using a static application protocol model. Communication to any Internet service must be agreed upon in advance. There is little or no leeway to these specific application rules.
As a result, even a simple protocolâ€”such as the Lightweight Directory Access Protocol (LDAP)â€”can take ten years or more to stabilize. In the meantime little or no progress outside of the bounds of the protocol can occur. Therefore, the process of developing a static protocol not only takes a long time, but also has the effect of freezing innovation while stabilization proceeds at a glacial pace.
Dynamic Application Protocol Framework
So we need a dynamic protocol model. And now we have the basis for one. In fact, we’ve had it for a while. But now it’s starting to catch on in a serious way. The key to the calculus for the new Renaissance is Extensible Markup Language (XML). XML has the promise of filling the need for a dynamic application protocol framework.
With a dynamic application protocol framework, independentâ€”or discreteâ€”services can exist and innovate at their own pace without the limitations and boundaries of a fixed protocol. XML provides the framework for provider of a service and a consumer of a service to exchange meaningful information dynamically and independently. The freedom that a dynamic application protocol framework will give to innovation and implementation of Internet infrastructure will have an immeasurable impact on the future of network computing, as we know it.
It is impossible to overstate the significance of the use of XML as an application protocol framework. The notion of a protocol framework alone is immensely profound and full of impact. This is why I compared it to the Calculus. The Calculus provided the foundation for modern science to shift so profoundly that we have not yet begun to decelerate from its impact. An application protocol framework will be the foundation for a similar significant shift. The effect of a shift in services will be so immense that it will be difficult to isolate the cause until well after the fact.
Without a major shift in infrastructure we are shooting rocks into the air with slingshots hoping they will somehow go into orbit. A dynamic application protocol framework is the missing solid rocket fuel needed to put network computing beyond the stratosphere and into orbit for the first time. I mean, the first time. Until then, we look like the Wright Brothers claiming the discovery of space exploration. And the time it seemed so profound, in retrospect, it was only 12 seconds and 120 feet.