Schlagwort-Archive: emergent system

Making Sense of Blockchain Mania

A spectre is haunting the Internet – the spectre of blockchain. Some claim it to be a disruptive technology, others think it is bullshit. In favor of the bullshit point of view the blockchain narrative lacks convincing use cases and looks like a solution in search of problems. On the other hand, a number of reputable companies seem to believe that this blockchain thing does somehow matter. Are we witnessing a mass delusion or is there more to it?

In a recent post, Peak Blockchain, I argued that that blockchain craze has a lot in common with the way Second Life, an online 3D toy world, was being hyped and then forgotten a little more than a decade ago. Perhaps there is more to this analogy than just the way attention rises and then drops in both cases, blockchain and Second Life. A real but boring trend may lure behind the exciting (at least to some) surface and its catchy name.

Second Life reached its peak of attention during the first half of 2007. Conceptually it never made sense: Except for certain types of computer games and some special-purpose applications, 3D worlds rarely make for usable user interfaces. Just imagine having to browse the shelves of a virtual 3D library instead of just googling or having to use an on-screen replica of a good old typewriter to write a letter instead of using a word processor – a user interface should support relevant task rather than needlessly replicate constraints of the physical world.

Despite the obvious flaws of Second Life as a tool and platform for anything, many well-known companies fell for the hype and built experimental presences in this virtual world. At least one of them, IBM, went even further and attempted to turn Second Life into a collaboration support business. Was everyone crazy?

Not entirely. The 3D toy environment of Second Life was merely the bullshit version of a real trend. Around 2007 the web had evolved from a distributed hypertext system into an interactive application platform (“Web 2.0”). Blogs had appeared on the scene (e.g., WordPress – 2003, wordpress.com – 2005). Cloud computing was on the rise, although nobody called it so yet (e.g., Google Docs/Sheets/Slides – 2006, Dropbox – 2007). Social networks and media were evolving (Myspace – 2003, Facebook – 2004, Twitter – 2006). While Second Life itself did not make much sense, it symbolized in its over-the-top manner what was going on as a proxy instance and gave it a name.

Looking at the recent blockchain mania from this angle, today’s trend-behind-the-craze might be the automated interaction of networked artifacts in emergent systems of systems. The evolution of the Internet has not stopped after turning all our computers, tablets, phones, and watches into cloudtop devices, user interfaces to (collections of) services that reside on the network.

Today we are moving toward an Internet of Things (IoT) where thing means anything that is a computer without looking like one. Networked computers come in a variety of shapes today. At home we find entertainment devices streaming audio and video into our living rooms, voice assistants that let us talk to the Internet, and home automation letting us control anything from lights to heating through our smartphones. Then there are connected and increasingly, autonomous cars as well as charging infrastructure for electric vehicles. Manufacturing equipment from machines and robots to entire factories continues to be automated and networked, as does agriculture.

Trend graphs showing search interest over time
Search trends for “blockchain” (blue) and “IoT” (red) over the past five years

Technically these cyber-physical systems are networked computers as we know them. However, they consist of increasingly autonomous entities forming emergent systems of (semi-)autonomous systems. On the one hand, autonomy is the key feature in cases like autonomous driving or autonomous robots. On the other hand, many IoT devices lack traditional user interfaces and their ability to interact with users in other ways than over the Internet remains rather limited.

As a consequence, IoT devices need the capability to interact with cloud services and with each other without requiring human intervention and perhaps also to make transactions autonomously. For example, one blockchain application that has been mentioned is charging of autonomous electric cars – a car buying electricity from a charger. This interaction must be secure, whatever this may mean in the particular case, and there is no such thing as a PGP key signing party for IoT devices.

Blockchains and cryptocurrencies are unlikely to solve any of the pertinent problems, but the problems are real: transactions between devices, managing access to continuously produced machine data, keeping manufacturing instructions under control while getting them where they are being needed, and so on. I suspect this is the grain of truth in blockchain mania. At some point everyone will drop the burnt term but continue to work on the actual problems and actual solutions.