NASA Backs Quantum Computing Claim

narramissic writes "Canadian startup D-Wave's demonstration via Web link of a prototype quantum computer in mid-February was met with skepticism in the academic community, but NASA has confirmed that it did, in fact, build a special chip used in the disputed demonstration. According to an article on ITworld, D-Wave designed the quantum chip and then contracted with NASA to build it."

How can we trust NASA?

[simdan (207210)]

After all, aren't they the ones that filmed a moon landing in some studio?

Sorry to bring out all the conspiracy nuts, couldn't resist. :-P

contracted NASA??

[paranoid123 (633401)]

Since when was NASA in the contracting-to-manufacture-computer business? NASA is more of a bureaucracy with a collection of labs all over the nation. They usually hand out the contracts. When they need computers they usually contract IBM or Silicon Graphics (maybe not lately) to do so.

Re:contracted NASA??

[Seumas (6865)]

How else is NASA supposed to afford diapers?!

Re:contracted NASA??

[bugnuts (94678)]

Several government agencies, especially national labs, do lots of civilian work. Often the labs do the basic research, and companies turn it into products at affordable prices.

If you read TFA, it stated that only certain agencies had the equipment to make and run the chips in the first place.

Re:

[TopSpin (753)]

NASA leases facilities and performs contract work routinely. This is how they keep valuable people and justify maintaining plant and equipment for which they have no immediate need. The classic case is wind tunnel time; both the facility and the staff can be leased by private parties.

Griffen was recently lobbying Congress [nasa.gov] (see pages 7-8) about this; apparently he would like some red tape cut to permit NASA to do this with certain Shuttle facilities where it currently isn't allowed.

Existence does not imply functionality.

[Dan Slotman (974474)]

The existence of a chip does not imply that said chip actually works.

Re:Existence does not imply functionality.

[IthnkImParanoid (410494)]

Or maybe it does work, with the unfortunate side effect of the computers constantly crashing into Mars.

Re:Existence does not imply functionality.

[Bobzibub (20561)]

No, it would simultaniously crash into Mars and not.

Go Team Canada!

Scientists Develop First Irish Computer

[mrbluze (1034940)]

Well, you see me boy, they did and they didn't. Or, t'was and 'twasn't, or, to be sure, to be sure, they might've but they mightn't've.

Re:

[Dan Slotman (974474)]

I know nothing about quantum computing—I can't comment on whether it is likely that the chip performs. I do think it is likely that NASA delivered a chip that does exactly what the specifications say it is to do. The question is whether the specifications describe a functional quantum chip. If I recall the original article correctly, there were questions about whether the demo computer worked at all, much less scaled to a useful level.

Does that NASA built a chip mean anything?

[stratjakt (596332)]

Does it mean the chip works, and it actually performs quantum calculations? I see nothing in the TFA where NASA confirms or denys the actual function of the chip, just that they made it based on D-Wave's design.

I still don't see any proof that anyone computed anything quantumly. How hard is this to prove, anyways, to all the quantum physicists in the house?

Re:Does that NASA built a chip mean anything?

[sco08y (615665)]

How hard is this to prove, anyways, to all the quantum physicists in the house?

IANAQP, but I think it's pretty hard to prove given that you can simulate a quantum computer with a classical computer. (Source. [caltech.edu])

But, if you have lots of qbits then you can simply argue that it's running too fast to be a simulation:

"Take for example a system of only a few hundred qubits, this exists in a Hilbert space of dimension ~1090 that in simulation would require a classical computer to work with exponentially large matrices (to perform calculations on each individual state, which is also represented as a matrix), meaning it would take an exponentially longer time than even a primitive quantum computer." (ibid)

So I'm thinking that when they get to their 64 or 128 qbit device that we know for certain that it's genuine.

I wonder how long it'll be before Intel and Motorola are selling quantum computers and arguing about the qbit myth?

Re:Does that NASA built a chip mean anything?

[tftp (111690)]

If they can routinely discover RSA secret keys then you probably don't need a quantum computer to verify that. An Intel 386 box would be enough.

Re:Does that NASA built a chip mean anything?

[tbo (35008)]

Disclaimer: IAAPRQC (I Am A Physicist Researching Quantum Computing).

I have no doubt their chip actually exists. That's not what people are skeptical of. There are more fundamental questions, a few of which I'll list below, along with my guesses as to the answers:

1) Does their chip demonstrate global coherence?
Maybe.

2) If yes to (1), can they maintain that when scaling up to larger numbers of qubits?
Almost certainly not with anything like their present design, unless they move to implement quantum error correction and the massive amounts of overhead that entails.

3) If no to either (1) or (2), can they implement a practical algorithm that gives at least a sqrt(N) speed-up over classical computers without global coherence?
Possible, but would be surprising if true. This is probably the main thing the academic community is skeptical about--we want to see some peer-reviewed research from D-Wave on this.

4) Why is all the press coverage so horribly wrong and misinformative?
Because it's more fun to make jokes and stupid statements about quantum mechanics than it is to actually write a clear and well-researched article. Also, talking to an actual physicist is far too scary for your typical J-school grad.

See this post [scottaaronson.com] on Scott Aaronson's blog for a much more informative and detailed analysis of D-Wave's claims.

Re:Does that NASA built a chip mean anything?

[ColdWetDog (752185)]

Thank you. Maybe.

Re:Does that NASA built a chip mean anything?

[tbo (35008)]

It's good to hear that there are at least some journalists with an interest and an aptitude for science. I think the entire quantum computing academic community has been a bit bummed out about the quality of media coverage lately. Scott Aaronson's blog [scottaaronson.com] has a number of posts discussing this issue, including a letter that he wrote to The Economist about its particularly bad D-Wave coverage. There is also some good news--Scott got asked by Scientific American to write a summary of Shor's algorithm--but mostly reading press coverage of our field is just maddening.

Science is hard work -- is it really surprising that interpreting scientific research, and translating results into layman's terms, is in some ways almost as hard?

No, it's certainly not surprising. I get a reminder of how hard it is to explain this stuff every time I try to tell someone what I do and their eyes glaze over. I don't claim to be good at explaining it, whereas science journalists seem to be quite good at making stuff entertaining and bringing it down to a layman's level. The problem is the completely uncritical coverage of miraculous claims, and the glaring technical errors that horribly distort the science. Is it common for journalists/editors to run a draft of their article past an actual scientist in the field? If not, why doesn't this happen? Pride? Deadlines? Journalism guidelines?

After being burned on a previous interview, I'd now be very reluctant to give an interview about my work without the reporter agreeing to run a draft past me for me to check for technical accuracy. Do science journalists honor that kind of request? If not, can you give me a journalist's perspective on what I can do to ensure the resulting article is accurate? I ask because I've got a paper coming out soon that might attract a bit of media interest.

Re:Does that NASA built a chip mean anything?

[MayonakaHa (562348)]

I thought the first rule of quantum computing is you may or may not be talking about quantum computing.

to all the doubters... the chip works!

[N3wsByt3 (758224)]

I'm a tenured professor in quantumcomputing and I can assure you the chip works! This is based on a paper I often require for my students, and I would hang my own Ph.D. on it's credibility.

O, wait...

This was meant to be posted here: http://slashdot.org/article.pl?sid=07/03/09/145221 9 [slashdot.org]

Sorry, my mistake!

First...

[suv4x4 (956391)]

First they ignore you, then laugh at you, then attack you, then you win.

The bad part is that fakes share the same fate, except the last bit.

Re:

[zCyl (14362)]

The bad part is that fakes share the same fate, except the last bit.

Isn't that more like:

1. They ignore you.
2. They laugh at you.
3. They attack you.
4. ...
5. Profit.
6. Move to a small island.

It's real!

[Pedrito (94783)]

I know it's real. I've actually seen it in action. An unfortunate side-effect is that my cat suddenly died... and didn't.

Not this again...

[posterlogo (943853)]

NASA doesn't necessarily "back quantum computing claim" of D-Link. They just confirmed they made a chip for them. Didn't we already find out a month or so ago that, according to their own admission, it's not a true quantum computer, but it MAY use some quantum principles in its design? As far as I care, even that claim hasn't been verified.

D-Link?

[SethHoyt (1024709)]

I didn't realize it took quantum computing to power my wireless router...

"backed the claim"

[MaggieL (10193)]

Insofar as I can tell, JPL has backed the claim that they made the chip; nothing further.

its about time!

[stim216 (881386)]

Finally! A computer that can run vista!