Hop on a laptop, click “download” on a book, and you’ve just directed nimble fingers of electricity to painstakingly transcribe every letter of that book. These fingers of electricity don’t write each letter like we do—on paper, in an alphabet of loops and lines. Rather, these fingers write each letter on eight switches, in a kind of braille alphabet called ASCII.
If the first letter of the book is “A”, these fingers start the transcription by flicking switches to “off, on, off, off, off, off, off, on”. A fraction of a second later, these nimble fingers of electricity will have flicked a few million switches. In doing so, they will have made a real, physical, ASCII copy of Fifty Shades of Grey materialise right underneath your hands.
If the laptop you’re on has a solid-state hard drive, there are more fingers of electricity flicking switches for you then there are real fingers on the planet. And yet somehow, you manage to effortlessly choreograph hundreds of billions of fingers of electricity to flick exactly the switches you want switched, without understanding anything of what you’re actually doing at all. You just point and click.
This is possible because some people designed you an intuitive user interface that makes choosing exactly the right dance of electricity easy. But the people who designed this user interface don’t understand this intricate dance of electricity either. Rather, they understand another “user interface” that is slightly more fundamental: a high-level programming language. If you keep digging, it continues like a Russian doll. Peer behind one neat mental picture and what you find underneath is another neat mental picture that is just a little bit more fundamental.
- SPONSORED -
We can pull apart these Russian dolls/mental pictures until we get to the baby doll/base mental picture. Here, we physicists try to draw a neat picture of how electricity works for the people who design electronic components. For instance, for the people who design switches that can be flicked by electricity.
The picture we have of how electricity works is something like this. Electricity is the aggregate flow of huge quantities of electrons. And just like the way ordinary objects move is dependent on what the gravity of a planet is, the way electrons move is dependent on what the electric fields in a material are. And every material has its own characteristic electric fields that are intrinsic to it. So from an electron’s point of view, a different material is like a different planet. The rules of how to move change.
The basic ploy of an electronic component designer is fairly simple. By making a component out of lumps of different materials melded together, she can ensure the electric fields are different in different parts of the component. Now, melding different materials also creates additional electric fields, but these are predictable and so can be part of her plan. That is, her plan to manipulate how electricity flows in a component, by manipulating the rules that dictate how electrons move.
But in addition to the predictable electric fields the component designer plans, there are also electric fields that cannot be predicted. And at room temperature, no matter how perfectly pure a material, these random electric fields are inevitable. So the electrons in the component are always subject to both predictable and random rules. And our picture of how electricity works depends on which rules dominate.
If the electronic component is less than about 20 or so atoms long, the electrons’ movement is dominated by the predictable rules. This means that, roughly, all electrons move the same. So in order to figure out the aggregate flow of all electrons, we only need to calculate how one electron moves. If the electronic component is larger than about 600 atoms long, the electrons’ movement is dominated by the random rules. This means the aggregate movement of electrons at every point is similarly random. Except, that is, for an easily calculated, small distortion from randomness due to the local predictable rules.
But if the component is between these sizes, the electrons’ movement is neither dominated by the predictable rules nor by the random rules. Here, we don’t have an elegant picture. Instead, we get a supercomputer, we teach it to follow both the predictable and the random rules, and we tell it to simulate the paths of a few million electrons. Then, we tally these up, to get a prediction of the aggregate flow of electrons.
Unfortunately, there are many people who have the awkward job of designing components of this intermediate size, including our friends who design switches that can be flicked by electricity. So in my research, I’ve tailor-made a new kind of “partial randomness” that can be strongly distorted by the non-random part of the electrons’ movement. My hope is that this will be able to greatly simplify our picture of electricity at this intermediate scale. Now, can I finish this piece by telling you what kind of fancy new component someone might design with a much neater picture? I can’t. And that’s the most exciting part.