On Fri, May 6, 2016 at 2:40 PM, juan <juan.g71@gmail.com> wrote:

On Fri, 6 May 2016 13:48:00 -0700
Sean Lynch <seanl@literati.org> wrote:

> On Fri, May 6, 2016 at 12:27 PM, juan <juan.g71@gmail.com> wrote:
>
> >
> >
> >
> > related
> >
> > http://www.scottaaronson.com/blog/?p=1400
> >
> > http://www.scottaaronson.com/blog/?p=2448
>
>
> These are both about D-Wave's machines, which are not universal
> quantum computers. IBM's is universal,

Yes, that's why I said "related" =P

What I found interesting about the d-wave 'affair' is that they
can't even prove that the machines are using some kind of
quantum effect. I'd expect the masters of the universe to be a
bit less sloppy in scientific matters.

They can't "prove" it, but AIUI given the current evidence there would have to be something pretty strange going on for it not to be quantum effects. But even if they ARE using quantum effects, they still need to prove they're faster than any classical algorithm for some problem.

> though it's only 5 qubits, not
> even as many qubits as they used to factor the number 15. Still, I
> think there's plenty of reason to be skeptical of QC, and even if
> you're not that skeptical of it, we've still got decades before it'll
> be cracking even 1024 bit RSA.

I just learned about "topological qbits" - they seem even less
practical than ibm's stuff, at least for the time being...

Depends on what you mean by "practical". Perhaps fewer applications due to the fact that you can only approximate the answer (with accuracy increasing as you increase the number of braids), but potentially more scalable because the braids are more stable. If IBM's computer can't be scaled up to the point that it's more than a toy, but TQCs can, then the TQC would be the practical one.