Google claims 'quantum supremacy' for computer

On Fri, 25 Oct 2019 05:10:02 -0400, Paul
wrote:

Martin Edwards wrote:
On 10/24/2019 11:35 PM, Peter Jason wrote:
On Thu, 24 Oct 2019 08:03:40 +0100, Martin Edwards
wrote:

On 10/23/2019 9:38 PM, Peter Jason wrote:
On Wed, 23 Oct 2019 21:05:37 +0100, ? Good Guy ?
wrote:



Your encryption is no longer safe with the authorities. Something
that
took 10,000 years can now be processed in 200 seconds.

https://www.bbc.co.uk/news/science-environment-50154993

Well, we'll just have to bury our drives in the back yard.

As an arts man, I have only the vaguest idea even about quantum physics.
Can someone explain this?

Some fanciful physics theory about a cat in and out of a box.
https://en.wikipedia.org/wiki/Schr%C3%B6dinger%27s_cat

That's one I have never got. For me the cat is in an actual position
with or without there being an observer. As for the quantum thing,
there is what I think is a pretty good explanation, but I still do not
get it. Thanks for the post anyway.

Quantum computing is limited to problems which "happen
to align with the interpretation that comes out of the
hardware". Quantum computing is not an "abacus". You
can't do payroll on it. In some ways, it almost resembles
an analog computer, in the sense that it has "measurement devices"
on the outputs.

https://arstechnica.com/science/2019...uting-efforts/

A physicist at Ars, it took him months to think up a "sample problem"
to run on a DWave. He chose to model a Bragg diffraction grating, and
got an output that kinda looks like what should come out. And he noticed
that the state the machine was in, was "similar" to how lighting effects
come through the grating.

https://arstechnica.com/science/2019...tum-computing/

They have more articles, but I don't expect this will lead to
an answer for you any faster.

https://arstechnica.com/tag/quantum-computing/

Wikipedia, about all you'll get out of this, is the notion there's
more than one kind of quantum hardware, and what they're working
on now for a particular type, is "error correction". The machines
need to stay coherent long enough to run a problem to completion,
or allow the error correction to take place. They're refrigerated,
to help stabilize them and encourage coherence.

https://en.wikipedia.org/wiki/Quantum_computing

And this is the topic which is the "threat" to cryptography.

https://en.wikipedia.org/wiki/Shor's_algorithm

The best part of the Shor's algorithm article is this dialog box...

"This section may be too technical for most readers to understand.
Please help improve it to make it understandable to non-experts,
without removing the technical details. (February 2014)"

The intro section of the article is OK. But once you hit the
"Quantum subroutine" figure, where the couplings between Qubits
are shown, it rapidly goes downhill from there.

From this we learn "writing a program for a quantum computer is
a metric bitch" :-) Notice how you need a PhD in math ? Nobody
in payroll will ever figure this out.

Today.

Next week may be different. Indeed, QC-bods are betting their careers
and using up their lives in the hope that tomorrow QC will actually be
useful for something.

The example I often use in situations like this is coherent light
beams. Lasers. At first a theoretical notion without much of a
practical application, now they are death rays used by advanced,
nuclear-powered robotic alien invaders on Mars.

QC might never be good for anything human-scale, lots of notions are
not but there are precedents for supposing it might be.

Maybe.

Eventually.

And even if not, blue-sky scientific research *ALWAYS* pays off. CERN
gave us YouTube.


*******

There are other topics in computing, that are out in lala land
like this too. This isn't the only thing that's hard to understand.

There are *hundreds* of technologies being worked on that are beyond
the grasp of the average punter or politician, many of which will
vanish without generating a single useful product or service. On the
other hand, there is a huge ring in which bitty things chase each
other at near light-speed which gave us dancing hamster sites.

On the gripping hand, fundamental scientific research *ALWAYS* pays
off.

And on the imaginary hand, the only way to discover if Mars has ever
had any freely running water would be to send up a team of areologists
to poke the planet with probes. [Or to send me up with many tools and
an Internet connection so the brainy guys could use me as an
intelligent finger.]

Designing elegant theoretical models is a lovely and human thing but
often the best way to do Science is to poke your fingers into the
wall-socket or to taste the urine.

QC is currently a bloody expensive fad, like VR helmets. Whether it
goes the same way is up to the cosmos, corporate sponsors and the guys
in the labs.

Me, I have my doubts but I've been wrong before.

J.


Paul