On Wednesday night, in front of a packed auditorium at the Computer History Museum and a Facebook Live audience, three researchers—Pat Gumann from IBM, Matthias Troyer from Microsoft, and John Martinis from Google—talked with the Computer History Museum’s David Brock about the hopeful but unpredictable future of the technology, as well as the unique quirks of managing quantum computing projects. A few takeaways:
On when we’ll have a useful quantum computer:
“We are at the point now where we have the science developed so far that we see a path to scaling it and building a quantum computer that solves problems in the next five to ten years,” predicted Microsoft’s Troyer.
Google’s Martinis, however, was less willing to go out on a limb. “It is going to be an interesting five years as people are investing and trying to push technology forward to see what hardware works,” he said. But because it is currently not just an engineering challenge but involves scientific research as well, “we are just going to have to see what happens in the next five years,” he said.
Troyer concurred, saying that with basic science involved, “it is hard to predict [that] we will make this great invention next year.”
Continued Martinis: “It’s going to take us a while to figure out how to do everything, both in hardware and software. I have been working on this since the ’80s. We are making rapid progress [now], but it may take some time to figure it out.”
“I don’t want people to get the idea that if we don’t get this working in the next couple of years, it’s not going to work,” Martinis said.
On hiring the right researchers:
At Google, said Martinis, “We hire a wide variety of engineers and physicists. We tend to hire physicists who have taken quantum mechanics classes, but also physicists with more of an engineering background who can work on a team.” For engineers, Google looks for those who have “taken some physics classes and quantum mechanics, so they can understand the concepts.”
“We are looking for scientist/engineers, [people] who can think as an engineer when they need to, and as a scientist” as well, he said.
Microsoft needs “people who are mathematicians, and physicists, and chemists, and engineers, but mostly people who have an open mind and who can solve problems,” Troyer said.
On managing a quantum research group:
Martinis pointed out the unique challenges of running a group working on developing a quantum computer. Because “you can’t copy quantum information,” he said, “you have to design computer in a different way. If you are building a normal computer, you assign [teams] of people to the CPU, and memory, and I/O. They design their systems, and [the systems then] communicate information over a bus.
“But when you can’t copy information, when you are sharing it all the time, and everything is interacting with everything, you have to make sure that everyone on your team understands what is going on globally... so when they are designing something they aren’t messing up something someone down the hall is doing. As the leader of the group, it is a challenge.”
On the end of Bitcoin:
In response to an audience question about the future of Bitcoin in the quantum computing era, Troyer pointed out that his group calculated that once a quantum computer can be built with just over 2000 qubits, “you can crack Bitcoin.”
“We joke that then we can fund all of our programming” with Bitcoin, he said.
“If you have a secret today,” Troyer warned the audience, “don’t encrypt it with RSA if you believe quantum computing is coming.”