https://www.huffingtonpost.com/entry/princeton-professor-sexual-harassment-not-punished_us_5a01d203e4b0368a4e872655 ]]>

Critics of this policy should remember that STOC is a conference, and the provision of an accessible 10p version is a good service provided by conferences. Conferences are not journals (it is a separate issue that our journals have problems too…)

]]>These researchers (and you) know much more about this than me. My intuition is quite simple: suppose that you came to a computer chip manufacturer and told them that instead of having a $100 price tag they are allowed to use a budget of $1,000,000 and that instead of having a chip that fits inside a pocket they can make it as big as a house. I would imagine that given these additional resources they would be able to make the chip more reliable, and I am not sure why things would be different for constructing quantum qubits.

]]>The rationale for this view is the following: if you regard the noisy quantum circuit as the correct physical model for computations allowed by quantum devices, and if for every fixed amount of noise the computational complexity allowed by the model is BPP, then this reflects on what you can expect from engineering efforts, and gives a strong reason to think that BPP is indeed the correct complexity class for noisy quantum computers.

]]>All that I meant is that with enough resources one can manufacture a single component (e.g. gate) whose failure probability is arbitrarily close to zero.

“Engineering” here is not a derogatory term but it is in contrast to “no go” theorems such as the uncertainty principle that tells you that some quantities can never be measured no matter how much resources you expend.

]]>