Published 17 September 2026 by LINO News
A Wonderful Problem
„By golly, it’s a wonderful problem since it doesn’t look easy.” Forty years later, a Nobel Laureate is still working on it. Richard Feynman – American physicist, Nobel Laureate, and one of the most gifted scientific communicators of the twentieth century – held a talk in 1981 that founded an entire field of research. He asked whether computers could operate according to the laws of quantum mechanics, and he summarized his conclusion in a single sentence: by golly, it’s a wonderful problem since it doesn’t look easy.
John Martinis – then a PhD student at the University of California, Berkeley – was in the audience. By his own admission, he did not understand everything Feynman was saying. What he understood was enough. He spent the next forty years working on the answer.
The Decision That Nobody Tells You About
Martinis opened his Lindau Lecture in the Inselhalle not with equations, but with a warning to the Young Scientists in the room: “A decision you make in the first couple years of graduate school has tremendous importance in your career.”
He meant two things: what kind of science you choose to do, and who you choose to learn it from. For Martinis, both decisions led to the same place: the research group of John Clarke at Berkeley – an experimental physicist who had built one of the most rigorous and creative laboratories in condensed matter physics – and to a postdoctoral researcher in Clarke’s group named Michel Devoret, a French physicist whose combination of theoretical depth and experimental precision would shape everything that followed.
The experiment they worked on together – probing whether a macroscopic electrical circuit could exhibit quantum mechanical behaviour – would eventually turn into a Nobel Prize. At the time, however, it began with a disaster. Before showing why the first measurements failed, Martinis briefly explained the underlying intuition: just as a skater needs a little more energy to overcome friction, quantum tunnelling in a dissipative system requires borrowing a little more energy. Turning that idea into an experiment proved far more difficult than the theory suggested.
“But you can understand that formula saying that if you have friction, you have to borrow a little bit more energy in order to be able to skate the energy, because part of that energy is going away as you transverse it – actually a good way to think about it. So here’s the experimental design. When we first started doing this – you graduate students will know – your experiment doesn’t work.”
What followed was a redesign that combined microwave engineering with quantum mechanics in a way that had never been done before.
The Janitor, the Banker, and a Nobel Prize
One night, Michel Devoret was working late in the lab at Berkeley when the janitor came in and asked what he was doing. Devoret explained using the image of throwing a ball against a wall. “It was great because the janitor understood what was going on. Maybe he got our lab cleaned up a little bit.”
Forty years later, Martinis noted, that exact conversation was part of the Nobel discussion. When journalists called after the prize was announced, Martinis reached for a different image of his own:
“And the way to talk about it is quantum mechanics, like a risk averse banker, a nervous banker. Nature allows you to borrow a Delta U of energy, which gets you over the barrier, but only for a very short amount of time. […] And that way you can escape from the system.”
Old School Physics – and Why the Data Must Fall on a Line
Midway through his Lecture, Martinis stopped to teach the Young Scientists something he described as thoroughly old-fashioned and which made the crowd laughing: „The first thing – very important – is when you make a plot, you have to label your axes. Okay? You know that when I go to physics conferences, especially with theorists, they’ll label it alpha. And that’s because theorists are so smart, they know how to read other people’s minds, and they think you can too. Label your axes with words to give you some idea what’s going on.”
For Martinis, the lesson was not really about plotting graphs. It was about building automatic reality checks into an experiment. Modern software can produce elegant figures with a click, but a graph that should be a straight line and isn’t immediately tells you that something is wrong. That was exactly what happened in their own work.
“Why is that important? Now, this is almost never done these days with automatic plotting routines and computers. And I’m really teaching you some old school physics. And the reason why this is so great: when we initially did the experiment, the data did not fall in a line. And we knew it was a disaster and we had to fix something.”
Quantum Mechanics Is Bigger Than the Periodic Table
The experiment did not confirm the hypothesis it set out to test. That, Martinis argued, was beside the point. “Quantum mechanics is not just the physics of the small. This is what people say all the time. It’s correct. But quantum mechanics is actually bigger than that. It can include US manufactured systems with a chip. You know, something that you can look at about a centimeter in size.”
The quantum playground is no longer only the periodic table of elements. It now includes devices we can design, manufacture, and engineer – and from that realization, an entire field emerged.
Feynman, 1985 – And a Problem That Didn’t Look Easy
Near the end of his Lecture, Martinis returned to the conference in Santa Barbara where a PhD student had once heard Feynman speak. “That’s where I learned about it, and it was a great talk. He’s a great speaker. Really excited.” And he quoted Feynman’s line, which Martinis had been carrying with him ever since: “By golly, it’s a wonderful problem since it doesn’t look easy.” Given that Martinis and his colleagues are still working on it forty years later, he observed, Feynman had understood the depth of what was ahead.
The Google quantum supremacy experiment of 2019 – a 53-qubit processor performing a computation too complex for any classical computer to verify – was one answer. The next, Martinis argued, requires the same leap that turned transistors into microprocessors: full integration, mass production, miniaturization.
The Panel – And the Man Who Connected It to Lindau
Earlier that day in the Inselhalle Martinis had joined a panel moderated by Rainer Blatt – an experimental physicist from Innsbruck, one of the world’s leading researchers in quantum optics and quantum information, and for many years the Physics Chair responsible for shaping the scientific programme of the Lindau Meetings. Blatt and his research group achieved the first controlled quantum teleportation between individual atoms in 2003. When he opened the panel by noting the distance the field had travelled, he was speaking from direct experience.
“When we started working on ‘Trapped Ion Quantum Information’ in 1994, we were excited if you could control a single quantum bit. Today we are discussing possible societal consequences of quantum computing. I personally find that this is a very remarkable change in perspective and perhaps a good place to begin.”
There was also a quietly remarkable detail on the stage itself: Michel Devoret, who shared the 2025 Nobel Prize in Physics with Martinis and John Clarke, had first come to Lindau not as a Laureate but as a Young Scientist – in 1976, fifty years earlier, when he met his scientific hero Paul Dirac. So far, he is only the third Lindau Alumnus in the history of the Meetings to return as a Nobel Laureate.
Blatt then asked the audience for a show of hands: how many expected their own work to be impacted by quantum computing technology in the future? Fewer hands went up than in an earlier, more optimistic poll about the broader societal impact of quantum computing.
“This is fewer than the optimistic part that we had before. Okay, so we have not been convincing so far,” Blatt remarked, before opening the discussion to the next question: what applications of fault-tolerant quantum computers might genuinely matter beyond the research community. The panel had not lacked ambition. It had simply been honest about where the field stood.
Four Perspectives on the Future of Quantum Computing
William D. Phillips, Nobel Laureate in Physics 1997 and one of the pioneers of laser cooling, answered the moderator’s question about the biggest misconception on quantum computing with a story.
“One of the possible misconceptions is 20, 30 years from now, everyone’s going to have a quantum computer in their home, you know, maybe a plug in on their laptop that has a quantum processor. To me, it seems really unlikely that that would happen. But then I remember that in my lifetime, there were people who said no one will ever need a computer in their home. There were big companies of which the chief executive officers said things like that. Those companies are not in business anymore.”
Michel H. Devoret, Nobel Laureate in Physics 2025 and, as mentioned fifty years after his first Lindau visit, now on the other side of the stage, reframed the entire question of what quantum computing is actually for: “I even like to say that we are on algorithms at the level where the Romans were on numbers. Try to do a multiplication with Roman numerals and you will understand what I mean.”
Arguments Are Okay
In his Lecture, Martinis had mentioned that he and Devoret once disagreed so fundamentally about what to call a particular phenomenon that they failed to publish the result in the best possible journal. In the panel, both sat beside each other again. – “Arguments are okay. We always use that to do the right things afterward.”
Victoria Sánchez Muñoz – a theoretical physicist from the Université Libre de Bruxelles, Heidelberg Laureate Forum Alumna, and one of two Young Scientists invited to join this panel – gave the afternoon’s most unexpected answer when asked what had drawn her to the field: “I know there’s a lot of applications for quantum, in general for quantum technologies. But to me they are like the extras in your vehicle. I can still drive without them.”
Christophe Valahu – an experimental physicist from the University of Sydney and the panel’s second Young Scientist – described a conviction grounded not in theory, but in experimental measurements. He was referring to error rates in quantum operations:
“In 2019, we were sitting a bit better than 1 in 100. And then I had a healthy level of skepticism, thinking, there’s no way we can get better than ten times, 100 times. And you fast forward a few years and it was better than 1000. And then now fast forward to today and we’re at 1 in 10,000. And that forced me to rethink the engineering challenges and to realize that there is no fundamental engineering challenge.”
Finally, it was Rainer Blatt to close the panel with a summary that is also the most accurate description of where the field stands: “The effort to understand and to control quantum systems has produced remarkable science, inspired a new generation of researchers, and expanded our imagination of what might be technologically possible. The future consequences of quantum computing will ultimately depend not only on advances in hardware and algorithms, but also on the people who develop them and the choices society makes about how to use them.”
This blog post is part of a series showcasing highlights from #LINO75. Most of the Scientific Programme from the recent Lindau Meeting is available in the Lindau Mediatheque.
