Slashdot Mirror

These are hardly shocking revelations. The document mentions to achieve control over two semiconductor qubits, whereas factoring 2048 bit numbers requires at least that many qubits, and probably several orders of magnitude more. The current record stands at control of 14 qubits, achieved in 2010 in Rainer Blatt's group at the University of Innsbruck, Austria, using trapped ions.

Some time ago, I wrote something on the history and possible future of quantum computing. Moreover, one also has to keep in mind that there are public key cryptosystems that most likely cannot be cracked even with quantum computers.

This stuff scales incredibly bad with time. Not even a hint of a "Moore's law" here.

There is.

I'm of the opinion adding more qubits to a superposition is going to be an exponentially hard problem.

History tells otherwise.

Oh, we already have a quantum version of Moore's law. However, the time constant for doubling is on the order of six years and not 18 months.

There is currently a huge research effort underway funded by tons of millions to make physical models of more complicated stuff: quantum simulators. With classical computers, there's basically no way to calculate properties of things like quantum magnets, quark-gluon plasmas, or high-temperature superconductors. But by using a different quantum system that we can precisely control (for example, ultracold atoms), we can tune it to simulate the behavior of the material we are interested in.

There has been an exponential increase in the number of qubits under control since the first serious experiments started almost two decades ago. If the current trend continues, we will have usable quantum computers between 2020 and 2023.

Any random notes from a clearly defined scale (e.g. C major) paired with triadic chords from the same scale will sound vaguely pleasant. Thus, this is not "what PI" sounds like as much as it's what C major sounds like.

In a similar spirit, I once wrote musical representations of quantum algorithms. I played around with various scales, and it turned out that the best sounding scale actually depended on the algorithm. But then of course the underlying data was not nearly as random as pi.

Ash is not good for jet engines. Period.

While this is of course true in some sense, there is also another problem with such massive flight cancellations as we had them in Europe (and still people are struggling to get onto their flights). People will have to seek alternative ways of transportation, which are typically unsafer than flying. Probably more than a dozen people have died in additional car accidents during the European airspace closure.

YES, please. My god I hate this aspect of Google, which is an incredibly annoying time-suck. It's even worse for me because I have a uni account that gives me access to most of the paywalled research, but only when I'm on campus, so when I'm off-campus and I want to know something I just desperately want the option to turn off all that paywalled crap.

Got an SSH account? Then you can download the papers by setting up an HTTP proxy.

// Code originally contributed by Microsoft Corporation.
// This contribution to the Stonehenge project is limited strictly
// to the source code that is submitted in this submission.
// Any technology, including underlying platform technology,
// that is referenced or required by the submitted source code
// is not a part of the contribution.
// For example and not by way of limitation,
// any systems/Windows libraries (WPF, WCF, ASP.NET etc.)
// required to run the submitted source code is not a part of the contribution

Why is Microsoft so pesky about this? It's all about patents. The Apache License requires each contributor to give a patent license for the code they have contributed. By stating that all the patent-emcumbered libraries are not part of the contribution, Microsoft does not give you a patent license, but you still have to acquire one if you actually want to use their code. So don't use this code, it's a patent trap.

Who cares if it's banned in China, he just wants to spam his service to slashdot.

And probably not even true. Internet censorship in China is usually done via fake RST packets, not via DNS manipulations.

I configured Tor to use a Chinese exit node. Here are my results:

- Chinese Wikipedia: accessible (used to be blocked)
- BBC Chinese (via bbcchinese.com): blocked
- BBC Chinese (via direct URL): accessible
- Article on Tibet in English Wikipedia: accessible (used to be blocked)
- Human Rights in China: blocked

T Mobile? I would not be surprised.

The country is definitely Germany. You can get the publication in question from the authors homepage Then take figure 1a (as suggested in hweimer's blog) and lay it over some google map, appropriately scaled.

The data is definitely centered around Germany, but tracks reach to Austria, Switzerland, Netherlands, Belgium, Poland and Cech Republic...

Contrary to what the paper suggests, the data has not been anonymized. Proper anonymization means that you cannot derive correlations between the behavior of the individuals, which was the whole point of the paper.

I don't know the exact legal situation in every European country. However, in EU countries this is regulated by the Directive on the protection of personal data, which requires for scientific use that safeguards have to be taken to prevent the identification of individuals. For some countries like Germany this means that the data has to be anonymized, although it is a grey area whether pseudonymization is sufficient.

More details on that matter can be found on my blog.

By using Tor and a few hacks, you can have a look at Chinese internet censorship by yourself.

I am German and I am potentially affected by the new law as I publish exploit code from time to time. I have written a blog entry about it, including a translation of the relevant section and some thoughts about the consequences.