Date

Everything is a wave, Louis de Broglie postulated in 1924. This idea is now central to Quantum Mechanics and won him the 1929 Nobel Prize for Physics. Smart man. In 1927 he also introduced (but abandoned due to criticism) a “pilot wave theory” that physicist David Bohm rediscovered decades later and developed into an alternative interpretation of Quantum Mechanics. Some people think it’s useful.

The burning question in early 2016 is how can we possibly apply Louis de Broglie’s insights to create a better social network? People stop me in the street and interrupt my dinners to ask this question. It’s on everyone’s mind. I feel that now is the right time to share with you, my dear audience, how we shall do this. One more social network will surely solve all of our problems, right? Of course it will!

First, some perspective

Since everything is a wave, so are you. A big messy bundle of waves, in fact, because you’re made up of a massive quantity of cells and atoms that are each also bundles of waves. That’s cool, but not news.

When we look into our physical selves we see waves, but what about when we look outward at our behavior? Well, that would seem to be waves too, because everything is. A big messy bundle of waves, likely. And not without wave-particle duality.

Backing up a little

Your behavior — the collection of actions or choices you make throughout your life — mostly seems like a foggy walk, doesn’t it? You do what you do, one thing after another, and don’t usually look too far ahead or behind. It’s probably rare that you analyze your behavior in detail, and that’s completely normal. Still, it’s obvious that life is a series of events and actions. We each basically do one thing at a time, switching frequently between different types of activity.

Each type of activity we engage in, no matter what level of abstraction, has an actual timeline of distinct historical actions. There is an actual history — a homogeneous timeline — of each time we’ve gone out for a beer, for example. Same for brushing our teeth, speaking to Jennifer, or pressing the H key on our keyboard. Anything.

We could theoretically maintain a timeline for every type of action, going into as much detail as we care for. Mostly we don’t digitize our own behavior like this, but it's quite possible and surprisingly easy.

From timelines to waves

Any timeline of events we choose to digitize, provided it is homogeneous, can be approximated by a periodic function. Treated as a wave, basically.

Waves, whether mathematical or physical, have the useful property of superposition. That is, we see the effects of constructive and destructive interference when two waves of similar type occupy the same space. We know how waves work so I’ll spare you the overview.

Social activity

We begin by approximating a specific timeline — a series of like points — as a sine wave. For this example, the timeline is all the times I’ve gone out for a drink (but it could be anything). I generated the timeline by pressing a button each time I went out; before, during, or after, it makes no difference.

Generating the wave

Let’s say I get out for a drink weekly, on average. This leads to a sine wave with a half-period (trough to peak) of 7 days, with a phase (or offset) such that the trough begins the most recent time I went out. The peak of the sine wave occurs 7 days later.

The amplitude of the wave at any given time becomes a rough probability that I’ll be out for a drink at that time. Or you can look at the amplitude as the rough probability that I’ll say yes to a proposition of grabbing a drink around that time. If I’m asked when the amplitude is low, such as in a trough, I’ll probably say no. If asked near the peak, I’ll probably say yes. With a probability gradient between. To be clear, this sine wave is not prescribing my future behavior but is only describing probabilities based on my history.

Let’s say I get out for a drink weekly, on average. This leads to a sine wave with a half-period (trough to peak) of 7 days, with a phase (or offset) such that the trough begins the most recent time I went out. The peak of the sine wave occurs 7 days later.

The amplitude of the wave at any given time becomes a rough probability that I’ll be out for a drink at that time. Or you can look at the amplitude as the rough probability that I’ll say yes to a proposition of grabbing a drink around that time. If I’m asked when the amplitude is low, such as in a trough, I’ll probably say no. If asked near the peak, I’ll probably say yes. With a probability gradient between. To be clear, this sine wave is not prescribing my future behavior but is only describing probabilities based on my history.

To summarize:

  • The wave is based on a timeline of actions all of the same type (going out for a drink).
  • The wave itself is now of that specific type (going out for a drink).
  • Different waves of similar types produce interference patterns.

Superposition of similar waves

Now let’s consider what happens when your friends have also digitized their “going out for a drink” timelines and have transformed them into waves.

This collection of waves is all of the same type but with each wave having a unique frequency and phase because everyone lives life differently. A bit like this diagram, although here the phases are the same (the troughs begin at the same time):

Three waves

Superimposing these waves makes it easy to identify the constructive peaks at any point in the near future. The peaks, when high enough, indicate a high-probability of a successful meetup.

Like this:

Superimposed waves

This peak-detection process can be driven by each person’s computer uploading their waves to a shared service. An algorithm on that shared service then makes connections by identifying peaks and sends back meet-up proposals that will probably work out.

Going big

You might be thinking, what’s the big deal? Our brains already do this, and besides, we usually schedule a regular meetup. That’s true, but the simple case of a few friends doesn’t tell the whole story. What do you do when they graduate, move away, get sick, or go on vacation? What about when you want some variety but the others have different ideas? The ability to make ad hoc connections is what digitized waves offer.

When there’s 20 to 30 people who you’re fine with grabbing a drink with, but you’re in the mood for a particular mix of people, size of group, day, time, or location it can get overwhelmingly complicated to set that up, even if you’re lucky enough to get everyone into a group chat. With digitized waves, however, we can offload this problem onto an algorithm, and computers are wonderfully capable of this sort of drudgery.

Going massively social

So far, I’ve used examples of activities with people you already know. What about organizing activities with others, possibly choosing from amongst thousands, possibly with precise preferences and highly specialized interests? That’s tough, but a perfect job for computers. Now they can actually do that for us, with minimal effort on our parts: simply pressing the occasional button.

Putting it all together, it looks to me like the basis for an actually social network. As in, a social network that helps people be more social with less effort. A social network for introverts, perhaps. Or simply for social people who’d rather do something than talk about it.

Economic activity

Let’s not stop at social. Instead of going out for a drink, let’s switch to buying a 6-pack or bottle of wine for yourself. No friends are involved this time. Say you make this purchase roughly 6 times per month, leading to a sine wave with a half-period of 5 days, with the same type of phase or offset as before.

What we have now is information that your liquor store (or their competitor) would love to have, especially if a lot of other people are doing the same thing you are. It means that you could now share, in real-time, your near-term probability of interest in purchasing a bottle of wine or a 6-pack of beer.

You could even go into detail of the type of wine or beer because you don’t drink the same thing all the time. Then each sub-type would be its own wave, with all of them superimposed into a high level “want to buy a drink” wave.

With this kind of information, the store could have your usual order put aside for you, they could initiate a delivery to you, or hook you up with a loyalty discount. A competitor might try to steal away your business with even better offers.

When many people share digital waves of this type, group buying power and improved access to special orders becomes possible. A group of people could also share their waves in a type of digital cooperative to make bulk purchases or apply leverage to sellers for better prices.

Connection-making from these waves works for any type of economic activity, literally anything we purchase. Food, toys, gadgets, clothes, gas, housing, anything. All of these types of activity are on different time-scales and levels of detail, but because it’s digital there are no limits.

Added up, it looks to me like the basis for a vastly superior replacement to nearly all types of advertising, on-line or off. It even looks like a unification of on-line and off-line commerce. With this technique we know exactly what we want and when we want it. When we have this knowledge in digital form we can share it however we choose. This also lets them know exactly what we want, and so be able to better provide it to us. What else, I wonder, might we discover we can do?

There you have it

By transforming our behavioral waves to streams of digital particles (timelines) and then into digital waves, we create the foundations a social network that actually is what it says on the tin. And, as a bonus, we also see a way to unify online and offline commerce in a way that puts a big dent in the world of online advertising we all struggle to love.