Navigating the Frontiers of Investment Today
Sore tail alert! Bruised butt warning!
Caution to my readers. Don’t try to plump down in a virtual chair.
I fear my friends at Google declaring “quantum supremacy” are cruising for a bruising.
Many of you live much of your life in a digital or virtual world, inscribed on smartphone screens. You defer to analog chiefly for meals and mattresses.
Now and then during the day, perhaps on the street, you might step in something analog. But you don’t take it seriously, like a lottery ticket or a 4K image or a WEP code for Wi-Fi.
The truth is most people talking about analog and digital don’t know what they’re talking about. That includes me, much of the time. But today we are holding a short class to prepare my readers (including lawyers and computer scientists at Google and high-level administration authorities in the White House and the Federal Communications Commission) who may not be fully comfortable with the differences between analog and digital.
Let me try to break it down for you…
Understanding your Analog Mind
These people live in a digital world these days and imagine it is real. But it is not. It consists of on/off switches that translate into bits, bytes, numbers and mathematical symbols — which are not real in themselves. They only point at real things.
When you point, your dog or your lawyer may well look at your finger rather than at the enigmatic object that you are pointing at. Pointing your finger is a symbol rather than an object. Confusion is widespread.
Think of maps and territories. Maps can be rendered digitally, as a series of rostered numbers defining rows of pixels each conveying colors in digital code. Territories consist of analog potholes and trees and clanking machinery and inaudible beetle chirps and fractal topologies and thermal noise and the infinite expanse of the electromagnetic spectrum.
Ultimately, territories consist of atoms and molecules in Brownian motion at around two billion oscillations a second, fraught with relativity and quantum entanglement and possible strings in 11 dimensions.
Hey, your map has almost nothing to do with the territory. The correspondence between the two is all in your analog mind.
That’s the essential gap between digital and analog. Digital is symbolic and mathematical. Thus it can be ferociously fast. But to connect to specific objects, digital symbols always need to be interpreted. Analog is mimetic and attempts to simulate what it sees.
Because the actual world is ineffably complex, even analog simulations of it are only grossly accurate. In general, analog employs modulated sine waves, captured from the electromagnetic spectrum, governed by the speed of light.
Electromagnetic waves have charges and voltages and magnetic moments and degrees Kelvin and vector potentials and quantum spins and phases and other real-world properties and textures. But for electronics, they mostly reduce to wavelengths (distances between peak amplitudes or voltages) and frequencies (the number of cycles per second).
Fiber optic signals, for example, run around 1550 nanometer (billionths of a meter) wavelengths and 153 terahertz frequencies (10 to the 12th or million millions of cycles per second). These are called infrared or below red in the visible spectrum.
Your eyes can perceive smaller wavelengths between 400 and 700 nanometers and higher frequencies around 700 to 400 terahertz. These comprise the dominant modes of sunlight.
Wavelengths and frequencies are related by the speed of light: divide the wavelength into the light velocity in the relevant medium and you get the frequency.
Because even digital machines such as computers partake of the real world, they also use waves. But rather than plunging into quantum enigmas, they only measure two states, high and low, to signify the ones and zeroes.
Because analog uses the entire wave to communicate information, engineers for centuries have dreamed of analog computers and telecommunications systems. In theory, using the entire wave is hugely more efficient than merely measuring two states of the wave.
It’s the analog temptation and it has lured many companies and investors to their ruin.
Analog is indispensable for sensing and interacting with reality. But if you want to compute or communicate numbers, you had better go digital.
Until the mid-twentieth century, most telephony was analog. Long-distance was a terrible problem because analog errors add up over long expanses of wire or air.
With digital, you only have to be sure about whether the current or voltage is off or on. With analog every point on the wave matters, so noise on the line mingles with the signal and distorts it. The longer the distance the greater the distortion.
What do these lessons of vanity and virtuality mean for investors?
The frontiers of investment today open in such categories as the internet of things, 5G wireless, Wi-Fi 6, and blockchain. All these systems have to sense and report on the real world.
5G, for example, promises to link millions of devices or things in a single square kilometer (5/8th of a mile) compared to thousands of links per kilometer in prevalent 4G networks. Every link entails a real-world connection that has to capture an analog signal or wave.
As the number of connections multiplies, analog technologies offer huge opportunities. Prominent will be analog chip foundries, analog chip designers, and analog sensors of all kinds. Count them. There are more than a dozen real-world sensors in your smartphone.
But the opportunities don’t stop there…
Analog Computing Gets a Makeover
Blockchain also opens huge opportunities, because every link is a target for hackers. As links multiply, the attack surface for hacking expands exponentially. The security patches increase additively. That’s why the more money you spend on internet security the more the bad guys gain. To fix the internet will require a new architecture, not more patches. A mastery of analog will be essential.
But the reason this is the computer age is the magic of simple digital symbols. Engineers have learned to avoid analog whenever they can.
If everything is a symbol, you can employ the huge frequencies of digital on-off switches to compute any set of numbers accurately and at fabulous speeds. These computational speeds, moreover, increase year after year as chip technology advances.
With every advance of “Moore’s Law,” doubling computer power every two years, engineers learn how to inscribe the transistor on-off switches more densely on silicon microchips.
All this is lost in analog computing. Analog is only theoretically more efficient. In practice, it bogs down in noise and distortion. In analog, the problem of computation moves from the actual processing back to the preparation of input and the interpretation of output.
Quantum computing is a case in point. Quantum computing is, in fact, retro-analog computing.
As MIT physicist Seth Lloyd has written, the entire universe is an analog computer, constantly calculating its state. All the answers you want are instantly available all the time.
The problem is framing the questions and interpreting the answers. In other contexts, this is called the problem of prayer. In quantum computing, it is termed the problem of distilling the qubits at input and reading the answers at output.
All computers are ultimately quantum computers, as I implied in my book Microcosm: The Quantum Era in Economics and Technology (1990). As analog machines, they are inexorably special-purpose devices, governed by the specific question they can answer.
As analog devices, quantum machines, though, should not be confused with programmable computers. People who declare “quantum supremacy” should be careful where they sit down.
Editor, Gilder’s Daily Prophecy