Its bonkers to me that the CEO of a company with no quantum hardware effort has such a large voice, especially given Nvidia sells the most important competitor hardware and therefore has an interest in downplaying progress. It’s like asking Elon whether trains are promising, or fuel cells.

They could have, but then they would have had to cite different works. Quantum computing kicked off only 20 years later for a reason: it required additional nobel-prize worthy discoveries (including by this year’s laureates) to reach the many orders of magnitude improvement in device quality.

The cited discoveries were not enough for quantum computing because of the huge amount of charge noise present in these devices. The transmon with its charge noise insensitivity was the breakthrough discovery that enabled quantum computing in superconducting circuits.

I am so dismayed a the statement today from the President of the @royalsociety.org that I want to publish here the letter I wrote to him yesterday urging him to address the clear concerns of the UK scientific community.

The tragic is that there is some real, accelerating progress in this field but these announcements overshadow those actual breakthroughs (such as googles surface code demo), which are harder to make sound flashy

I guess I was thinking academically, he’s old enough so that his PhD students now have PhD students themselves ;)

They did not. Read the facts from one of the grandfathers of quantum computing: scottaaronson.blog?p=9170

They did not. Read the facts from one of the grandfathers of quantum computing: scottaaronson.blog?p=9170

They did not. Read the facts from one of the grandfathers of quantum computing: scottaaronson.blog?p=9170

I don’t think it’s “shut up and calculate” - as long as all the interpretations are not distinguished by different predictions for experiments, they are one and the same physical theory. Interpretations in physics are only useful if they lead to differing quantitative predictions.

I‘d also say that a lot of the services provided by journals are obsolete. Who needs copy editing into a format that you don’t want anyway and that introduces sometimes serious errors? Arxiv wrapper journals (like quantum) are the future.

Also these red eagle eyes just look evil.

Very interesting article that raises a problem beyond just AI: that researchers can overhype any topic with almost impunity. Even after grants have finished the question of „have you really achieved what you promised?“ isn’t seriously asked as long as some papers came out that get cited

A first step would be to provide access to SN journals free of charge to members of the affected institutes.