Ground Control to Majorana

I hope you all understand that I realized late on Monday that I had completely forgotten to write this. I dare you to forgive me.

This week, I got a roommate! His name is Francesco. He's one of the new friends I've been hanging out with. I invited him to switch to my room because he had a kind of crazy roommate situation that was starting to take its toll on his well being, plus I figured it would be pretty fun. It does mean I have to get half dressed in the shower stalls and not entirely in my room, but that's about the only downside.

For the fourth year in a row, all of my friends will be leaving on missions. Conveniently, being year 4, I have cycled back to be with some of the originals. I have generational friends. It's very weird.

This week I've been thinking about two things: being nice to people and Microsoft's Majorana quantum chip. I will explain my thoughts on both of them. Which one will come first? You can't know until you read (that was a joke for the physics nerds).

Being nice to people is such a simple thing to do. I don't think that being nice means that you never say anything negative about anyone or anything. I think that you can still tease people and make fun of your friends (and yourself) while being nice, and you can also dislike people without being rude or mean. Being nice includes apologizing for things if you suspect you may have crossed a line. It means not making people feel like you don't value them, even if you hardly know them. With friends, they know that you care (if you did it right), so you have a little bit more leeway in what you can say that will still make them feel cared about. 

I saw 3 things that week that got me thinking about this: 2 examples of what not to do and 1 good example of what to do. The two bad examples occurred today, actually. The first one happened when I was trying to get in contact with someone about ordering lunch for a student trip that my work is hosting. They're leaving early in the morning, so I was asking if they could have the food available for pickup at the time of departure. The guy I talked to was incredibly short with me, making passing comments about my lack of knowledge of his own business. Mind you, he would be getting paid a considerable amount of money if this order went through him. There was just nothing that warranted his tone of voice, especially when I was requesting information that would lead to money in his pocket. Lesson 1: be kind in professional situations. You will be more successful. You don't need to take advantage of people at every chance you get. The second example was in my math class, when I asked a question to the teacher about a brand new concept that none of us had ever heard of: relations. His first definitions were incredibly abstract and it was hard to grasp the meaning behind them, so I asked a clarifying question about the difference between his early and late definitions. He looked at me and said, "Well, they are what they are, so I don't see the point in your question." I asked it again in a different way, and he said, "Yeah, I don't understand what that means." I tried a third and final time to ask the questions, at which point other classmates nodded in agreement. He responded, "I don't think your question means anything. The definitions just are." I then said, "Okay, then you can continue," at which point he said, "no, I just don't see the point in your question." I said "that's fine, just go on and maybe you'll answer it." "I don't see how you can feel confused when there's nothing I've said that could be confusing." At this point, I couldn't see any reason to try to explain again, since he clearly wasn't interested in answering me, so I said, "that is okay, just keep going through the material. It isn't important at this point." My classmates tried to rephrase my question throughout the lesson, since it was about the foundation of the rest of the lecture. I left the lecture with minimal understanding and I suspect many of my classmates did as well. Lesson 2: when teaching someone, patience is incredibly important because you have no idea which question will spark an epiphany. Be nice to them, since they don't have 50+ years of experience in your field like you do, and they're just trying to understand you better. The situation was really uncomfortable and I was more embarrassed for him than for myself.

A good example came at work after a meeting about people stepping over boundaries. Basically, another department was trying to tell us how to do our jobs differently than how we were trained, so we met to discuss how to help them without complying with every request and acting in the appropriate scope. One of my bosses called someone a "puppet", which was totally harmless in the conversation, since it was a fairly accurate way to describe the particular situation we were discussing. Nobody thought anything of the comment, in fact, most of us agreed. However, in an admirable demonstration of humility, my boss called us back a few minutes after the meeting ended and apologized for saying that about the person, and that it was probably inappropriate to say. He said he apologized if he offended anyone, and recognized it as something he probably shouldn't have said. I found this a wonderful example of several things, like humility and empathy. My boss' comment didn't sit right with him, so he put other's needs in front of his own pride and apologized. I think this is a huge part of the difference between being kind and not; being kind includes trying to remedy things and doing what you can when you aren't perfect.

I try to be a kind person. Most people who know me also know that I love to tease and view a lot of things satirically. When it comes to a battle between my ego/pride and the relationships I make, regardless of depth, I would like to think that I choose to prioritize the feelings of another. Lesson 3: be nice to everybody. People like you more, they trust you more, they care about you more, and I dare say that they might even love you for it, just a little bit. 

Also, side story, just before I talked to the mean catering guy, I spoke with someone at BYU's Food-To-Go scheduling. She warned me that the guy might act a bit annoyed and was very kind in helping me work out what I needed to do. I talked to her pretty casually since I had talked to her before, and my boss came over and jokingly told me to stop flirting with the phone lady and get back to work. This lady was like, 45 years old. I could hear it in her voice. I don't flirt with anybody, much less old ladies. After I got off the phone, I told him how old she was and it was a hilarious situation. I turned red so fast, even though I wasn't embarrassed. I can never predict when I blush, but I can feel it on my face. It is a curse that I have.

Now, total subject switch, but it's time to talk about Microsoft Majorana chip. I would like to forward this with the words of Stephen Hawking: "Science is beautiful when it makes simple explanations." This depends on the explainer's ability to convey meaning in a way that doesn't require technical knowledge. If I can do that properly, you can understand the most complex of scientific topics. I would like to think that I can do that. Stay with me and you will learn something really interesting about one of the hardest problems in physics without having to solve a single equation. I promise that I will keep things simple and define any crazy words you probably don't know. This should be teachable to a 6th grader.

This new piece of technology promises to change the world of computing forever through quantum architectures that haven't been seen before. If you haven't yet heard about it, it uses a "new state of matter" that will make quantum computing available in a matter of years, not decades.

There is just one teeny-tiny caveat: it probably won't.

Let me explain briefly how quantum computing works, then the chip, then we can talk about Majorana specifically once you are up to speed on things. You have to understand a little bit about quantum particles before you can understand how they are used in computing. 

After another guy named Albert Einstein showed that photons, the particles we know as light, act as both a wave and a particle (a wave and a dot), this even crazier guy named deBroglie (pronounced 'de broy') theorized that it wasn't just photons that were wavy, but all of the elementary particles, like electrons and quarks and other weird things. In his PhD thesis, he showed that quantum (another word for super super small) particles could be represented by a little packet of a wave. Then some guy named Schrödinger came up with this thing called a wave function, and the rest of history.

That probably didn't mean a whole lot to you. The important thing to know is that a particle isn't a little dot moving in a wave, it is the wave itself. The entire wave is the particle, it's just so small we see it as a dot. We have verified this by experiment. If you multiply 2 wave functions together, you get a probability distribution, which describes where you are most likely to find the particle when you take away its ability to move. Until you put the brakes on, the "particle" is the whole wave, so it doesn't have a definite position. We know it occupies the space between the start and end of the wave, but we don't care about that. We want to know where it really is. So, we put up a screen and shoot a bunch of particles (like electrons) at it, and we see a pattern emerge. Electrons hit some places more than others, and this pattern is the probability distribution. Different particles have different distributions, which gives us clues to their properties.

The reason we multiply two functions together is because each one describes the behavior of the particle if it were stuck on a line. When we multiply two together, we pretty much let the wave function do its thing on a flat sheet, aka a 2D surface. Guess what happens when you multiply 3 of these wave functions together? You get some wild 3D shapes. If you've taken chemistry, you've seen them before. These are the electron orbitals, like 1s, 2p, etc, but that isn't super important.

Okay, that was kind of a lot pre-intro. Now, you're probably thinking, why does that matter (bahaha pun)? Why do we care? Well, my friend, let me tell you.

Computers currently work by using electricity to put an electron in a box. You know binary? The 1's and 0's state if there is an electron in the box or not, respectively. This is memory. When you send very specific electricity through these boxes, you can change whether or not there are electrons in the boxes, which is how computers perform calculations.

Remember how the wave function basically describes probabilities for where the particle can be? Well, that turns out to be incredibly useful in computers. What if, instead of putting an electron in a box, we use something that has a 60% chance of being in the box and a 40% chance of being outside? Then we can use electricity to change those percentages as we need to. By stopping the "thing" we call a qubit, we either see a 1 or a 0. This type of computing is called quantum computing. It is incredibly powerful because instead of only doing one calculation at a time, qubits can calculate what happens if we did a whole lot of different calculations at the same time. They aren't defined until we tell them to be, so we don't have to spend energy taking electrons in and out.

Here's the other crazy thing about qubits: they're kind of stuck together. Like the couple making out every time you walk by, their "moods" are highly influenced by each other through a process called quantum entanglement. This means that if I do something to one qubit, the next one will change without me doing anything directly. This enables insanely complex computations using very little energy. Even crazier, since the qubits don't have a defined state, these chains of calculations can be done at the same time, which makes them way, way, way faster than regular computers. They can solve problems in minutes that take real computers septillions of years (a really long time, like, older than the universe's 16.5 billion).

However, there is one downside: these qubits are pretty unstable. They tend to destroy themselves at the slightest perturbation, just like a cranky toddler. If you get them really cold, they do quite a bit better, since they pretty much have no energy to complain with. They also are more likely to be wrong, since there is the slight non-zero chance that they don't exist inside or outside the box you want, they exist somewhere else entirely. Thus, they must be corrected for, which is getting so much better in recent years.

All particles have alter-egos, which really mess up their ability to act normally and tend to cause those things that upset quantum toddlers. The reason the new chip is called the Majorana chip is because it utilizes particles called Majorana (named for the scientist who theorized their existence) particles whose alter-egos are the same as their normal ones. Thus, you don't get the schizophrenic self destructive behaviors that you do with the other particles, so they are much more stable. Yes, I just used children with an incurable mental disorder to explain quantum particles. I bet you understood it, so I think it works. These Majorana particles are the "new state of matter" Microsoft has claimed to invent, although we already knew about these. They just got them really cold and got them to stay where they want, which I guess kind of counts as new matter.

Google just put out the Willow chip (named after yours truly) and gave benchmarks for its performance. Microsoft's statistics were notably absent. They have claimed that their design is scalable to the long-standing final figure for supercomputing: 1 million qubits. This would be more powerful than every computer ever invented combined, by far. However, this is entirely theoretical. The more qubits you have, the more they interfere with each other. 1 million qubits in the palm of your hand (the current size of the chip that Microsoft claims can be scaled down) might not be physically possible, we have absolutely no idea. It would be incredibly difficult, that's for sure. Their 8 qubit computer is still pretty error prone, albeit more stable.

The Google chip, Willow, has 105 qubits, making it significantly better than Microsoft's eight qubits at a similar size. The reason that Majorana has received any attention is because this is the first time that a qubit has been created that uses particles and antiparticles as described, called a topographical chip. It isn't currently better, but maybe it will turn out to be better in the future for its stability. Willow beats the current fastest standard computer, Frontier, such that it can solve a problem that would take Frontier 10 septillion years in under 5. Oh sorry, not years. 5 minutes. Crazy stuff.

Thus, Majorana has hype because it's new. It isn't groundbreaking by performance, but by concept, and time will tell if that concept is worth improvement.

I guess that's all from me. Have a good week.

Will Ott