It doesn't matter that much if one is slightly faster in Javascript or rendering when Firefox will halt up for 5-10 seconds rendering a new tab. Maybe it's faster than Chrome, but if I have to wait for it, it doesn't matter. It doesn't matter how much Firefox devs work on "UI sluggishness" if it's a single thing can lock up all input to the browser.
It is being done all the time. What is this fraudulent nonsense even doing here on
/. Was this not already debunked enough?
No, it's not debunked; that is, no one has shown it not to be what's claimed. However it has been shown that even if it is what it claims it's no better than an optimized classical simulated version. It's like someone claims they have a quantum chicken, and it may be quantum chicken, but it still can't cross the road faster than a fast non-quantum chicken.
Simply having the source code doesn't mean you have the ability to actually use the source code to make bug fixes should the need arise.
And yet, it still means that you can fix it, or even rewrite it in something else, if you want. Not having the source code means this is between much-more-difficult and impossible. The lesson here should be that everything we use should be open source, including compilers and libraries, not "well in theory I might have problems, so screw that whole open source thing
Here's an example where the 5th amendment makes a positive difference.
Prosecutor: Your honor, we don't have any evidence, but we're pretty sure he killed that man because he's all shifty looking.
Judge: Tell us why you killed that man.
Defendant:
Judge: Let the record show the defendant has refused to answer the question. This court is holding him in contempt. I order him confined in prison until such time as he consents to answer the question. Bailiff, take him away.
Defendant: I didn't kill the man,
Judge: You're lying, that's perjury, go to jail.
Defendant: But there's no proof!
Judge: The confession you refuse to give would be proof; refuse to give it, you're in contempt, deny it and it's perjury. Jail!
They also didn't have OSHA and other worker protections back in the 1930's. A lot of men were injured or fell to their death constructing the Empire State Building on a practically slave wages.
Sure but the question here isn't "can we do it without safety regs" it's "can we even accomplish this given modern technology". However, the Empire State Building is about 434m (according to wikipedia). I would be very surprised if the requirements didn't go up with height, and at nearly double that, I'm sure things get considerably more difficult. Maybe if they have Gibsonian-style nanoassemblers... but that's a little beyond modern technology.
The comic (as previously posted) was amusing and also wrong; a user-level exploit might be able to get you those things, if credentials aren't encrypted. Browser exploit can probably scrape your pages or similar, which is of course bad. However, a system-level exploit can do all this and more:
Such exploits may be less bad for you, but would be considerably worse for any of the large services you rely on, potentially exposing the entire userbase.
This may be somewhat theoretical, but only because most people generally have enough sense to patch system-level exploits quickly. Most apparently not including Microsoft.
Why can't you imagine this? One of these costs $130, off-the-shelf. They have eleven total, all around the world, which is $1430, off-the-shelf. Add in some more for the sensor setup etc
Then, how long and how many people does it take to test them all regularly after they're installed? And how hard are they to install on an existing data center, vs dropping one of these on the floor, slapping some RFID stickers around, and walking away?
I imagine this is a trial run and IBM could probably come up with an even cheaper bulk solution if they need to. But it sure sounds like a lot less overall
Github did this recently too which was annoying, because it was useful. They're not entirely clear why
How will we be able in the future to distinguish between "fake" media and "real" media. As media is being used in legal battles, eventually there will have to be a requirement to determine an 'authentic' footage, which means we will need some form of protocol, file format and/or tools which can create media which can be proven as 100% original and unaltered in any way.
Signed image/video
While you could arguably hack apart a device and get at its key, doing so while leaving no trace brings us back into physical forensics, and almost certainly significantly harder to do. Add a few different measures like light-sensitive markers, exposed EPROM, etc. It's not about preventing hacking (or even difficult), it's about simply making alteration evident.
It's pretty obvious who is and who is not reading the article here:
In short, there seems to be no evidence, at present, that the D-Wave machine is going to overtake simulated annealing for any instance size.
The author concedes that it is possible that this may happen, but:
Well, I concede that almost anything is possible in the future—but “these experiments, while not supporting D-Wave’s claims about the usefulness of its devices, also don’t conclusively disprove those claims” is a very different message than what’s currently making it into the press.
Additionally the author wants this to succeed because of possible results of its failure:
Academic QC programs will be decimated, despite the slow but genuine progress that they’d been making the entire time in a “parallel universe” from D-Wave. People’s contempt for academia is such that, while a D-Wave success would be trumpeted as its alone, a D-Wave failure would be blamed on the entire QC community.
Seriously, read the whole damn article.
The problem is that it's not faster, and while there's a study that concludes it is, the blog post specifically invalidates this:
Namely, the same USC paper that reported the quantum annealing behavior of the D-Wave One, also showed no speed advantage whatsoever for quantum annealing over classical simulated annealing. In more detail, Matthias Troyer’s group spent a few months carefully studying the D-Wave problem—after which, they were able to write optimized simulated annealing code that solves the D-Wave problem on a normal, off-the-shelf classical computer, about 15 times faster than the D-Wave machine itself solves the D-Wave problem! Of course, if you wanted even more classical speedup than that, then you could simply add more processors to your classical computer, for only a tiny fraction of the ~$10 million that a D-Wave One would set you back.
About the paper claiming it's faster:
As I said above, at the time McGeoch and Wang’s paper was released to the media (though maybe not at the time it was written?), the “highly tuned implementation” of simulated annealing that they ask for had already been written and tested, and the result was that it outperformed the D-Wave machine on all instance sizes tested. In other words, their comparison to CPLEX had already been superseded by a much more informative comparison—one that gave the “opposite” result—before it ever became public. For obvious reasons, most press reports have simply ignored this fact.
Same thing that happened with hotmail. They switch to Windows servers, it crashes and burns horribly, so they switch back. There's no quality control, no development, it goes to hell, and everyone switches to the far superior service Google offers (since they decided to grow their own and not acquire youtube).
Then they switch everyone over to zune.com or something to try capitalizing on their name
Yes, and that there is a "brick wall". First, the article may be wrong; exascale might hit by (or before) 2020. They've got 7 years. That's a long time in terms of technology; the first teraflops supercomputer was 1996, merely 17 years ago. Speed increase can't happen indefinitely, but we're not talking about indefinitely, just exaflops. Even if this is not achieved by 2020, they have not hit a "brick wall", because development will continue until it is achieved. There is nothing even slightly theoretical making exaflops unachievable.
In short, even if the article is right, it's wrong.
Sure but they're one of many. Even if one of the many don't accomplish this, surely another will. If not by (or before!) 2020, sometime later. People aren't just going to give up if it doesn't happen by some arbitrary date. This is my real point.
These days, how much is really revolutionary anyway? So many new supercomputing announcements are "we threw N parts at this, so it's Yflops".
The optimum committee has no members. -- Norman Augustine
