Wednesday, January 20, 2016

Reality Transducer or Omniscience Engine? Five Metaphors for the Internet of Things

I think metaphors are important. They can help to organise the way we think about something, highlighting its unappreciated features, and allowing us to identify possibilities that were previously hidden from view. They can also be problematic, biasing our thought in unproductive ways, and obscuring things that should be in plain view. Good metaphors are key.

With that in mind, in this post I want introduce five different metaphors for thinking about the internet of things (IOT). These metaphors were inspired by my recent reading of Samuel Greengard’s book The Internet of Things (MIT Press 2015). I think they provide interesting insights into the potentialities of the IOT, though they may also skew our thinking in misleading ways. I want to explore the positive and negative features of these metaphors in what follows.

Before I do that, however, I want to consider more deeply the value of metaphors by using a case study from the work of the philosopher Daniel Dennett.

1. The Value of Metaphors: Dennett’s ‘Universal Acid’
Dennett’s 1995 book Darwin’s Dangerous Idea is an extended meditation on the meaning of Darwin’s theory of evolution for our understanding of ourselves. The book is replete with clever metaphors and thought experiments. Indeed, Dennett is something of a master of the art, recently collecting his greatest hits in a book entitled Intuition Pumps and other Tools for Thinking.

Given this, it’s hard to pick just one metaphor from Darwin’s Dangerous Idea to foreground my discussion of the IOT. But I’ll settle on one of the simplest and most arresting: the metaphor of universal acid. Dennett describes it like this:

Did you ever hear of universal acid? This fantasy used to amuse me and some of my schoolboy friends — I have no idea whether we invented or inherited it, along with Spanish fly and saltpeter, as a part of underground youth culture. Universal acid is a liquid so corrosive that it will eat through anything! The problem is: what do you keep it in? It dissolves glass bottles and stainless-steel canisters as readily as paper bags. What would happen if you somehow came upon or created a dollop of universal acid? Would the whole planet eventually be destroyed? What would it leave in its wake?
(Dennett 1995, 63)

Universal acid doesn’t really exist, of course. But the fiction is both provocative and evocative. Through the simplicity of the original idea (“a liquid so corrosive that it will eat through anything!”) and the well-time rhetorical questions (“what would happen if…?”), Dennett quickly has us imagining its reality.

The payoff comes later. Dennett uses this provocative fiction as a symbolic representation (i.e. a metaphor) for Darwin’s theory of evolution. As he puts it:

Little did I realize that in a few years I would encounter an idea— Darwin’s idea— bearing an unmistakable likeness to universal acid: it eats through just about every traditional concept, and leaves in its wake a revolutionized world-view, with most of the old landscape still recognizable, but transformed in fundamental ways.
(Dennett 1995, 63)

In fact, this is the central thesis of Dennett’s book — one he develops over the next 500 or so pages. Evolution really is somewhat akin to a universal acid. Once you understand that we are ourselves products of the evolutionary forces that Darwin describes — in both our bodies and our minds — nothing is ever quite the same again. Your entire worldview is dissolved and reconstituted by the process. There's some exaggeration in the metaphor, to be sure, but that's part of its charm.

I think this example reveals the power of metaphor. When you use a symbolic representation as a framework for understanding another important concept or idea, you can cast that concept or idea in a new light, appreciating both its virtues and dangers in a heightened way. Can we do this for the IOT?

2. Five Metaphors for the Internet of Things
Let’s start by asking a question: What is the internet of things? To some extent the ‘IOT’ has become a buzzword, often bandied about by people who don’t really know what it is but recognise that it is important. The gist of the IOT is relatively easy to grasp. We are all now used to the idea that computers can ‘talk’ to each other: sharing data and information through the internet. We are also all now used to the idea that other devices can connect to one another using the same infrastructure (e.g. smartphones, tablets and wearables). What we may not realise is how many ‘things’ are (and can be) imbued with similar degrees of connectivity.

This is what the concept of the IOT enables us to see. Indeed, the takeoff in terms of the number of objects and devices that can now connect via the internet is truly astonishing. Cars, buildings, warehouses, thermostats, clothes, shoes, traffic lights, cameras, watches, flasks, and eyeglasses are just some of the things that have now been tagged and imbued with internet connectivity in the past few years. This trend can be expected to continue. Greengard cites some figures from Cisco Systems that illustrate the potentialities:

[A]pproximately 12.1 billion Internet connected devices were in use in April 2014, and the figure is expected to zoom to above 50 billion by 2020. In fact the networking firm [i.e. Cisco systems] says that about 100 “things” currently connect to the Internet every second but the number will reach 250 per second by 2020. Overall, the Internet Business Solutions Group at Cisco Systems estimates that more than 1.5 trillion “things” exist in the physical world and 99 percent of physical objects will eventually become part of a network.
(Greengard 2015, 13)

This may be an over-estimate, but it gives a sense of the eventual scope of IOT. And connectivity is just a small part of it. The real impetus for this is to collect data from the things imbued with internet connectivity, mine this data for useful information, and then use it for making better decisions through both human and automated decision makers.

How can we make sense of the IOT? This is where the metaphors come in. One thing that struck me while reading Greengard’s book was the different ways in which he described the infrastructure of the IOT and its various possible uses. Although he never explicitly labelled these descriptions, or referred to them as metaphors, I think it is worth doing so for the reasons stated above. So, without further ado (further? hasn’t there been plenty already?) here are five metaphors for thinking about the IOT:

The IOT as the Apotheosis of Connectivity: Dali’s famous painting “The Apotheosis of Homer” depicts the epic poet’s ascension to the divine. It was a common scene in classical art. This reflects the original theological meaning of the word ‘apotheosis’, which is to elevate something to a divine or holy status. I use the term deliberately (if advisedly) here. The suggestion behind this first metaphor is that the IOT represents an elevation of connectivity to a sacred status. This is perhaps the most basic metaphor for the IOT. As Greengard puts it, the IOT “extends connectedness far beyond computers and into all the nooks and crannies of the world” (Greengard 2015, 16). Optimistically, this augurs an end to loneliness and isolation; more pessimistically, it augurs a struggle (which many already experience) to maintain privacy, solitude and disconnection.

The IOT as a Reality Transducer: ‘Transduction’ is a term used in many fields. It refers, roughly to the process of converting one thing into another. In biophysics, for example, it refers to the conveyance of energy from one electron to another by changing the class or type of energy. And a ‘transducer’ is a device for converting one kind of quantity (e.g. pressure) into another type of signal (electrical). I use the term loosely here to refer to the potential for the IOT to convert physical reality (“real” reality) into digital or virtual reality. If everything is tagged and imbued with digital technology then it can become part of the digital world. If the tagging incorporates control technologies (e.g. robotic systems) then the physical reality can be manipulated using the same tools as the digital world. Greengard says this represents an ‘advanced state where physical and digital worlds are blended into a single space’ (Greengard 2015, 18). This opens up many interesting, and possibly disturbing scenarios. It may deconstruct our concept of reality (remove ontological distinctions between the virtual and the real), and lead to the programmability of the world (through advances in nanotechnology).

The IOT as an Omniscience Engine: This is another quasi-theological metaphor. In orthodox monotheism, God is supposed to be an omniscient (“all-knowing”) being. I have my doubts whether such a being is possible, and I’m not even sure that the concept of ‘omniscience’ is coherent, but nevertheless I think there is something about the IOT that warrants the metaphor of the omniscience engine. Kevin Ashton, coiner of the term ‘internet of things’, wrote an article in 2009 lamenting the fact that ‘people have limited time, attention and accuracy — all of which means they are not very good at capturing data about things in the real world’ (quote from Greengard 2015, 20). In other words, he was lamenting the epistemological limits of humans beings. The hope was that the technology behind the IOT would be better at capturing data about reality. This suggests to me that the IOT can be viewed as a way to make the world more knowable. In the limit, this implies that the IOT could know everything about everything. But ‘knowledge’ is a tricky concept. Mass data collection is not knowledge: making sense of the information collected by the machinery of the IOT will be key.

The IOT as a Global Neural Network: This is an obvious metaphor. It moves us beyond the IOT as a tool for connectivity and data collection, and focuses on its other capacities for data mining/processing, and physical action in the world (through human or robotic agents). When you include those capacities, you see the potential for the IOT to form a global neural network. Neural systems in humans and other animals perform three basic functions: they collect sensory data; they process this data; and they initiate actions in the world. The infrastructure for the IOT can perform the same basic functions. This is interesting in its own right: a single or multipolar agency could emerge from the architecture of the IOT, which raises many fears and hopes (cf Bostrom Superintelligence). But it also enables us to appreciate the next metaphor.

The IOT as a Hivemind Platform: This metaphor appeals to the concept of a superorganism or hivemind, i.e. something akin to a bee colony or, to use a fictional analogy, the Borg of Star Trek. The idea is that the IOT is potentially all-encompassing. We might think that there is some hard and fast distinction between us (human beings) and other ‘things’ in the world. But there isn’t. There is no reason why we cannot be among the things that are incorporated into the infrastructure of the IOT. In many ways we already are. Wearable tech and prosthetic implants can be imbued with internet connectivity: they can be controlled and updated from the ‘cloud’. In the near future, we will be able to use these implants and devices to connect with one another on a brain-to-brain basis. Indeed, there are already some examples of experiments doing exactly this. As the potential for this kind of connectivity grows, we could transition to a something like a hivemind society. I’ve explored some of the issues arising from this in my previous posts about the Borg-like society.

So there you have it. Five metaphors for thinking about the IOT. I think these metaphors enable us to see potentialities within the technology that may previously have been hidden from view. But I also accept that they may be hyperbolic and overblown. Are there better metaphors to be adopted? I’m curious to know what you think.

No comments:

Post a Comment