If we cross the threshold. Until then, adding a qubit requires halving the noise floor, so the Moore's law equivalent to exponential scaling is basically adding a fixed number of qubits per year.
If we cross the threshold. Until then, adding a qubit requires halving the noise floor, so the Moore's law equivalent to exponential scaling is basically adding a fixed number of qubits per year.
Are you suggesting that gates surpassing the fault tolerant threshold will never be achieved? The vast majority of experts disagree with this.
Without fault tolerance, you have a hard wall on circuit depth because errors grow exponentially with depth. You can't make up for this by adding any number of qubits. "Halving the noise floor" means...what, improving gate fidelity?
Until then, the best we can do is if you half the noise floor (or double the fidelity) you can add roughly one ideal qubit.
The most obvious reason to suppose the fault tolerance threshold will be crossed is that simple extrapolation of progress predicts it will.
On the other hand I can make a nod in the direction of the decades of expertise any one can build in a chamber of vapor ware with the hope that it will distill into something useful.
As critical as I am, I of course congratulate anyone on taking on such difficult innovation. Nonetheless, there is no physical evidence it will amount to anything thus far. It is purely hypothetical, and the only thing theoretical in justification is the fictitious proof of its use through mathematical modeling which is not physical reality and thus closer to building a virtualized computer in a MMO that mimics Earth saying you get to play as a martian using some weird technology while it runs on classical computing.
All new tech starts as purely hypothetical. (Not to mention concrete milestones like quantum supremacy, gate infidelity trends, etc.)
(Likewise, the earliest "demonstrations" of quantum computing where they factored two-digit numbers were ridiculous for several reasons, not least of which was that you could do it in a microsecond using a classical computer. Then later, when quantum supremacy calculations were done by Google, the mathematical problem chosen was completely useless for any practical application.)
It was not until 2013, 46 years after Braginsky theorized the basic principles, that a squeezed probe was used for something other than a proof-of-concept: measuring gravitational waves at LIGO. This enabled LIGO to detect ultra weak gravitational waves that would otherwise be invisible to it: https://www.nature.com/articles/nphoton.2013.177
Needless to say, it will take even longer before these techniques are used for economically relevant applications.
In terms of computing, we have seen many new paradigms materialize within decades, even centuries, of their conception. Babbage conceived of the Analytical Engine in the 1800s, but we saw programmable computers by the mid-1900s. The transition from electromechanical to electronic computing occurred within a few decades. Every instance of these examples are real physical examples that were able to be used and demonstrated as a physical device providing utility. We can even go back to the early computation devices of the loom or even analog computing calendar/clock systems.
Quantum computing is an ambitious concept, and while I respect the academic rigor that goes into its development, the lack of concrete, physical outcomes, even in a rudimentary sense, after four decades is notable. Theoretical advancements are important, but the inability to materialize them in some substantial form, especially when juxtaposed against the timeline of prior computing paradigms, can warrant skepticism.
Don't get me wrong, I value innovation and the pursuit of new technological frontiers, but I also believe in questioning, and when the physical evidence is wanting, I'll voice my concerns. This is not to discredit the work being done but rather to keep the discussion grounded and accountable.
For a "real physical quantum computer" to exist by this definition, it should be able to carry out such operations using quantum phenomena, without any reliance on classical computing architecture for function, error correction, or result verification.
What we currently have in the field of quantum computing doesn't fit this bill. The quantum systems we have today are not independent "computers" but rather more like "classical computers conducting quantum experiments." They're akin to people playing an MMO and running an imaginary computational architecture that only exists within the confines of the game rules. In this sense, they're creating and operating within an entirely simulated environment.
This isn't to diminish the value of the research being conducted or the strides that are being made. But I would argue that the statement "real physical quantum computers exist" is, at this stage, a significant overreach. We may have precursors or tools that can manipulate quantum phenomena in interesting ways, but we're still a considerable distance from having an operational, standalone quantum computer in the full sense of the term.
Nope, wrong. Plans for quantum computers have always included assistance from classical computers. This is like saying for a nuclear weapon to exist, it must have no conventional explosives, but in fact all nuclear weapons contain conventional explosives to initiate the nuclear reactions.
You've made a number of incorrect claims here without admitting error, so I won't continue the conversation.
You could, but the premise is wildly false; tenured professors can get fired for cause (and tend to have some process protections alongside that), and tenured professors aren’t the only workers with contracts or legal protection that restrict firing to “for cause” (pretty much all high level employees with individual contracts like—but not limited to—executives have that, maybe with a high-priced buyout option, though without or with limited process protections), and most unionized workers and most (even if not unionized) public workers have both limitations to firing (or other adverse actions) for-cause and strong due process protections.