tl;dr:
"Funkhouser sat in the rear seat, and Fisher sat in the driver seat on top of a buckled seat belt. <... > Fisher engaged Autopilot while the car was in motion on the track, then set the speed dial (on the right spoke of the steering wheel) to 0, which brought the car to a complete stop. Fisher next placed a small, weighted chain on the steering wheel, to simulate the weight of a driver’s hand, and slid over into the front passenger seat. Using the same steering wheel dial, Fisher reached over and was able to accelerate the vehicle from a full stop. He stopped the vehicle by dialing the speed back down to zero."
It's not control of individuals photons (that would be boson sampling), but gaussian pulses. Still a tour-de-force of a demonstration, but actual practical implications are very limited.
Not necessarily true. Google's result used a programmable universal quantum computer for a specific task; the point of the experiment was not just to show that it can do this specific convoluted task faster than classical supercomputers, but also to show how powerful the quantum computer they've built is and how well they can control it. What they then did is go on a publish a dozen papers of running many other algorithms on it (of course, for those algorithms the quantum computer is not yet powerful enough to show advantage).
On the other hand, the device used in this experiment is a single-purpose non-universal sampler with very limited applications.
Unfortunately, the same complexity considerations do not hold for Ising problems. While many NP-hard problems can be formulated in Ising form, it is often not hard to get a “pretty good” solutions to these problems. DWave and collaborators have spent a decade trying to come up with exactly the same thing as demonstrated here — namely, a problem specifically designed to demonstrate quantum advantage of any sort — and as of now did not succeed.
To answer your question specifically: DWave does not allow the same level of control over qubits.
> Are there types of problems where this is known to be faster now?
Not yet. This is commonly called "quantum advantage" or "quantum supremacy", i.e. proof that on a certain (even artificial and with no practical applications) problem a quantum computer can perform better than classical state-of-the-art. So far even for the task of simulating quantum circuits (which quantum computer should definitely be better at!) we don't have enough cubits to have a go at quantum advantage.
Depending on connectivity (full connectivity like on ionQ vs planar connectivity like IBM / Rigetti), we need between 100-200 and thousands of qubits to show quantum advantage (denser connectivity -> lower qubit number requirement). And that's just for a synthetic problem with no applications.
Good point. This _should_ be illegal. And there is definitely a need for better data protection legislation, to protect consumers against exactly this power imbalance.
Well the question really is economic incentives. If it is too hard to do (say, doesn't scale) or can be simulated efficiently-ish on a classical computer -- no one will do it.
Do you think there is a significant chance that quantum will never take off (i.e. there are non-obvious limitations that will prevent quantum architectures like superconducting qubits / trapped ions / quantum dots /... from ever outperforming classical supercomputers)?
Related, what in your opinion is the best indicator (or would be the best indicator if demonstrated) of the potential of quantum devices?
A lot scientific libraries have virtually non-existent documentation (often just a bunch of html generated by Doxygen). NetworKit is one example, but it applies to way too many of them. Caffe is another example mentioned below
"The overarching message is that there is no single cause of the tuition boom. The reason for rising costs differs based on the type of institution and the state it’s in, and even varies over time."