This appears to be a surprisingly level-headed government initiative. The amount is enough to do something but not so much as to be a colossal boondoggle. It's the sort of fundamental research which private industry tends to under-produce because it's hard for any given company to recoup the full value of the R&D, but the target area is something with potential downstream practical implications.
> bipartisan
Also by coincidence (?) the first time I've seen that word used recently in a context that wasn't preceded by the words lack of or similar.
Advisors make these recommendations. Say the right words about the US falling behind China or national security risk and no one wants to be the person on record who said no to this.
I agree. It'd be concerning if they launched with some obscene $1 billion type financial tag with it.
If or when it makes sense, they can always increase funding for these three centers and or to build additional centers. Which is exactly what I'd expect in the coming years.
Why is it worthy of a lot more than $361 million in initial funding?
Your premise is that it'll stop at $75 million in funding. I'd bet against that premise. The US Government has a tendency to start its programs small and expand them considerably over time.
Besides, the US simultaneously has several of the world's largest companies pouring resources into quantum computing (Google, IBM, Microsoft, Amazon, Honeywell, Lockheed, Intel, et al.). It's not as though $75 million is the extent of US investment into the space.
I disagree. I'd be completely find with a $1bn financial tag with it. That's not going to be $1bn for a year but spread over something like 5-10 years. If this is a strategic technology, it is highly beneficial for the government to invest in it. They'll make the money back once the technology becomes widespread.
And I mean, look at the last paragraph
> ... the NSF that would be given $100 billion over five years to invest in American research and technology issues, including quantum computing.
Which seems low when you consider that we spent $430 bn on the F35 and lifetime expected cost is 1.5tn for more acquisition and maintenance. Governments work with larger amounts of money and when you look at science (arguably the best way to get ahead in military) it is pennies (CERN was funded internationally and only cost around $20bn over 10 years and $1bn/yr for operation). I mean the largest super computers cost only a couple hundred million, which means if Google or Facebook wanted to, they could compete.
>It's the sort of fundamental research which private industry tends to under-produce because it's hard for any given company to recoup the full value of the R&D, but the target area is something with potential downstream practical implications.
Quantum computing ? I've seen just a few months ago some article about a single start-up raising >200M ?
There is a difference between doing fundamental research and bringing a product to market.
We know enough about quantum computers to be able to build some of them in limited ways which are commercially useful in very specific domains, so then companies can get investment to do that.
But we still don't really understand all of the fundamentals, and it's always hard to raise investment for fundamental research.
Suppose you could prove that building a quantum computer that can run Shor's Algorithm on interestingly large numbers is impossible. Knowing that would be of immense value. People could stop wasting time trying to create it or defend against it. But how would anybody commercialize that information? Who has the incentive to fund that research?
Yes, I think we probably need a lot more fundamental research in the world - it also reminds me of this good TEDx talk that suggests we need more “curiosity-driven” research that doesn’t explicitly have a specific well defined purpose (if that makes sense): https://tedxsydney.com/talk/the-power-and-potential-of-curio...
> In which way are quantum computers commercially useful?
VC Bait seems like a natural fit. Just say you are making a quantum computer that predicts global warming trends using blockchain. Might want to buy a wheel barrel for all your money you’ll be leaving that meeting with.
Eh, bipartisan stuff still happens. Look at the military industrial complex. Expanding executive power, expanding the surveillance state. Pro wall street regulation repeal.
While the media paints a different picture, bills sanctioning and restricting China have also been getting through which also means bi-partisan support despite the election rhetoric.
One does not simply get into $26T in debt without a whole lot of agreement from both sides of the aisle. To view any side as pure is to be a pawn in the overton window.
I'm not sure what the mechanism is (i.e. karma threshold or moderation) but it looks like all the account's comments are dead by default and have been for a very long time.
i don't really like Trump, but i don't believe everything i read in the media. All news at this point is just push an agenda from one side of the tribe or the other.
i never hear any of his speeches live, i only catch the clips that the news shows you. So how accurate is their reporting? Are they pushing an agenda, taking comments speeches out of context? Both sides do it to their advantage.
Why don't you just watch a speech on Whitehouse.gov then instead of still asking for more people's opinion and asking everyone else to do the leg work? That's kinda the point being made
So... you don't pay any attention to Trump, but assume he's good on the basis that the media says he's bad? What's your reason for thinking they're incorrect?
Is this coming from him directly? I would be really surprised, he has plenty of advisors for that. What comes out of him directly is more visible on ie twitter, different story
Better or worse, little of what the executive does is really because of the president directly. The president chooses the sort of people to put in positions of power then the whole ship largely sorts itself out. But the president gets to be responsible for everything because they are nominally in charge, and people keep pretending they are superheros with all sorts of ability to personally oversee how the US operates.
We can certainly say that if Trump didn't want quantum computing research to happen his administration wouldn't be doing it. So if it does, he has de facto supported it. Even if by inaction.
It’s a peculiarity of the US system that decisions by the federal bureaucracy tend to be attributed to one person. You can be relatively confident that Donald Trump has no idea what a quantum computer is.
The popular suggestion - with regard to your specific example - is that the president himself is directly supporting those federal actions in Portland (against the wishes of state leadership and law enforcement) because appearing strong on crime is politically beneficial for his re-election. There is no other reason for those officers to be present, refuse to wear uniforms, or use civilian vehicles. The
But then, I’m sure you’re fully aware of this and are just making hay.
Trump is also directly supporting the US beating China in technology. He doesn't care how it is done or what the technology is. This is just like he doesn't care how riots are stopped just that they are stopped. It wouldn't be the least bit surprising if he told somebody on his team to get it done and they handled the arrangements.
>because appearing strong on crime is politically beneficial for his re-election. There is no other reason for those officers to be present,
Maybe he genuinely wants to stop crime? You shouldn't assume somebody's reasons for doing stuff when you don't actually have any proof.
> refuse to wear uniforms
Do you have an example? All videos I have seen have uniforms, just not the normal police uniforms.
> or use civilian vehicles.
This is extremely common throughout the west under certain circumstances. It is especially common for federal agencies more so than state / local. The reason you use civilian vehicles is to prevent the person you are trying to apprehend from realizing you are police. If somebody knows they are going to be arrested they may flee or grab a weapon. Some of the people they are arresting are quite possibly the same people who have thrown bricks at police officers. Some of the police have been hit in the head with bricks. Its completely understandable an officer wouldn't want give the suspect additional time to get a weapon when dealing with people like that.
The explicit purpose of a uniform is to associate an individual with an organization (e.g. the Department of Homeland Security, the Federal Bureau of Prison, etc) that vests them with specific authorities - for example, the authority to detain and charge citizens with crimes.
These actors in Portland do not wear patches, badges, names, or other identifying features that associate them with any authority. Camouflage clothing can be purchased by anyone with the Internet and a credit card. A badge number can’t. This distinction is important to maintain a free society.
I’m not going to explain the history of secret police and political disappearances. Go read about the Dirty War in Argentina for a taste.
>The explicit purpose of a uniform is to associate an individual with an organization (e.g. the Department of Homeland Security, the Federal Bureau of Prison, etc) that vests them with specific authorities - for example, the authority to detain and charge citizens with crimes.
Nice job changing the goal posts.
>These actors in Portland do not wear patches, badges, names, or other identifying features that associate them with any authority.
Are you sure? I have heard there actually are badges on the uniforms. I have only seen pictures and videos on my phone so I could not tell personally if that is the case.
Will you please provide a video / picture of the uniforms?
> Camouflage clothing can be purchased by anyone with the Internet and a credit card. A badge number can’t. This distinction is important to maintain a free society.
Anybody can buy a fake police uniform and badge as well. Amazon has multiple for cheap.
I doubt I would know a real police uniform from a fake one.
Hardly. You contend that these actors are wearing uniforms. I argue that camouflage is generic and doesn’t serve the purpose of a uniform as it doesn’t indicate under what authority they serve.
> Are you sure? I have heard there actually are badges on the uniforms. I have only seen pictures and videos on my phone so I could not tell personally if that is the case.
It is being widely reported as such, yes. Additionally, new federal legislation has been put forward just in the last week to require federal officers to wear identification in the future.
> Anybody can buy a fake police uniform and badge as well. Amazon has multiple for cheap.
I doubt I would know a real police uniform from a fake one.
>It is being widely reported as such, yes. Additionally, new federal legislation has been put forward just in the last week to require federal officers to wear identification in the future.
I'll ask again. Do you have a video or picture? I have heard that these uniforms do in fact have a badge and identification on them.
Allegedly these officers did identify themselves when they made the arrests but due to the sounds (like people shouting at the officers) we cannot hear it on tape.
> This is precisely my point.
Every uniform and badge can be copied and sold to other people. If it becomes illegal to have a fake uniform people could still buy them illegally. Cops will also sometimes sell their old vehicles so even if they roll up in a police car you wouldn't know. It would also be possible to paint a car to look like a police car as well.
It sounds like it is impossible to ever arrest somebody without you complaining that there is no way to know if the person is legit. What is your solution to this problem? Never arrest anybody?
So, that's a pretty major issue, and the president was presumably presented with a meaningful choice on it (if he wasn't, then there are other, very serious, problems). Whereas I would seriously doubt that he even rubber-stamped the quantum computing thing; even if he did it's not something he'd have any context on, and would really have to rely on the experts. They're very, very different things.
In a parliamentary system the quantum computer thing would be signed off by a ministerial department, and might be rubber-stamped by the minister, but probably would not be attributed to the minister by the press and certainly would not be attributed to the PM. The "have weird paramilitaries attack our own civilians" thing would definitely go to the PM and would require a decision from them, and would be blamed on them.
The "[whoever] admin did whatever" thing, for every little thing the government does, with named whoever, is a peculiarity of the US system, but everywhere attributes major decisions which the executive leader signed off on to the executive leader.
However, all research institutions take a cut of grant money in what is usually called "Facilities and Administration" cost. My undergrad rate was 50% of the proposed sum. Obviously the exact rate varies, but the fee remains.
Ex. If a professor is writes a proposal for 1M. When the funding agency awards it they tack on another 1M for the institution.
This is why "expensive" research is favored in universities, over theoretical work. A big part of tenure review is how much money you have brought in in funding, because that's a direct measure of how much money you made the institution.
So 75M is more like 30M-40M of directly funded research, with the rest going to the institutions.
To zeroth order, it works out to roughly 30 small research groups for 5 years.
At a place like Berkeley, where I did my PhD (but not in anything quantum related), a PhD student and a postdoc both cost a professor something like $75k/year in salary and benefits (money not spent on postdoc salary basically goes to student tuition). Add in the professors salary as well (which is often substantially supported by grants), figure $150k/year for the prof. So in people costs alone, let’s say 1 professor and 2 trainees, that’s about $300k/year, or $1.5M/5 years. Multiply by 30 professors supported, and you’re at $45M, which is 60% of the $75M. And yeah, this is assuming that experiments are free.
Now is 20-30 professors a lot? If sufficiently narrow and esteemed, absolutely. There were 29 people at the 1927 Solvay Conference. Give them an extra couple months where they aren’t sweating a grant, and that’s a lot of time to think.
Agreed. This is what you get when you measure performance in terms of profits.
To quote Major General Smedley Butler:
"A racket is best described, I believe, as something that is not what it seems to the majority of the
people. Only a small "inside" group knows what it is about. It is conducted for the benefit of the
very few, at the expense of the very many. Out of war a few people make huge fortunes."
Higher education is a rachet.
Students pay ever-higher tuition, which at some level is backstopped by the government. If you don't go to college and pursue prestigious degrees the job market will be inaccessible, beyond basic skill entry level.
Graduate students are exploited laborers for the promise of a PhD and a letter of recommendation for a postdoc. Drawing from a global pool of talent for cheap research labor. If you don't do everything your advisor says you will lose your visa, or not get a publication and your academic career is dead-on-arrival.
Professors are in an environment that encourages pursuing expensive application-based research and high impact publications to pad their resumes for tenure. If you don't rake in the money and publish in "prestigious" journal (usually the walled off ones) you will not be considered for tenure, a raise, etc.
Tax dollars fund research institutions that take their generous "administration" fees and then hand over findings to walled garden journals. If you don't fund research we will not make the breakthroughs that gave us out "comfortable" way of life.
In the US, student athletes (until recently) could not profit off their likeness and all the money went to the school.
At the end of the day, the team of MBAs and administrators make fat salaries. This wouldn't be so bad if institutions used the revenue to improve education and dissemination of knowledge. Instead, they build fancy stadiums and raise tuition.
I think these overhead numbers are quoted as the percentage on top of what gets spent, rather than the percentage of the incoming cash. 50% overhead means that buying a 1M machine involves giving 500K to the institution.
This initiative is launched quicker than I expected, considering the pressure on federal budget due to the pandemic. It's part of National Quantum Initiative Act, signed into law in December, 2018. https://www.congress.gov/bill/115th-congress/house-bill/6227
$75M is rounding error compared to anything being discussed in the vicinity of the pandemic; I wouldn't really expect it to enter into the conversation.
Someone needs to invest in developing improved mental models for quantum computers. Much of the development of classical computing (particular from a user perspective) was driven by metaphors: memory, stacks, pipes, files, folders, etc.
What are useful metaphors in the application or use of quantum computers?
Quantum computers are modeled with extremely basic linear algebra. Quantum states are vectors and gates are matrices which multiply those vectors to get new vectors. How much simpler do you want? You can explain most of it as rotations around the unit circle which we all learn in high school or earlier.
The problem is pop science crap that tries to explain superposition and entanglement with complete word salad instead of "linear combination" and "product vector cannot be factored into tensor product of single-qbit states", respectively.
Anyway it is extremely doubtful that more programmers will need to learn how to program a quantum computer than, say, will need to learn how to program a GPU. They're both useful co-processors for very specific workloads. Programming will remain a fundamentally classical endeavor.
I'm not necessarily asking for better metaphors of quantum phenomena. But the opportunities or affordances of quantum computers is extremely unclear, even to experts. It's like explaining the physics of a transistor to someone and waiting for them to come up with an app design.
Taking the analogy of a GPU, how would you describe what a quantum coprocessor could do for a specialist (in, say, neuroscience or material science)? What kind of datasets would be appropriate? How should they think about what can be done with quantum processing? That's where I think there is room for metaphor development.
The short answer is that all of this info is out there, but it doesn't live in pop science articles.
For materials scientists the value proposition is clear: quantum computers will enable efficient quantum chemistry simulations. Here is an overview paper: https://arxiv.org/abs/1808.10402
Popular science has been trying to explain quantum mechanics to people for well over half a century now, and it hasn't worked one time. All we got was a bunch of woo like What the Bleep do We Know?! You need math. There's no other way around it. Any understanding of quantum mechanics or computing that is not fundamentally rooted in mathematical reasoning is worse than mush.
Most advantages that quantum computing has aren't practical for most people outside "the HN tech bubble". It's meant for specialized types of high throughput computations.
Title of the post is misleading: The newly-funded centers will address three different topics in the field of Quantum Information Science (QIS).
Quantum computing (the integer-factoring kind) is the focus of only the UC Berkeley-led consortium. This effort accounts for 1/3 of the announced funding.
The other two centers will work on Quantum networking (UIUC) and Quantum sensing (U Colorado).
Surely this amount pales in comparison to the likely level of investment in this field by the intelligence community, considering the implications for cryptography. This is like announcing $75m for nuclear technology research during the Manhattan Project.
NIST is already well into a post-quantum crypto competition. Any serious user of cryptography will have switched to the new protocols by the time quantum computers become powerful enough to crack production RSA. Cloudflare has already prototyped protocols in TLS [0]. Scott Aaronson has gone on record saying he would be "astonished" if this happened within a decade.
This isn't like the cold war days. The quantum computing research community is close-knit and people would notice researchers being hoovered up by the NSA; this hasn't happened.
Shor's algorithm is undeniably a groundbreaking result but is not the killer app of quantum computers. It's more of an unfortunate side-effect.
> Any serious user of cryptography will have switched to the new protocols by the time quantum computers become powerful enough to crack production RSA
Yes, but the NSA could still decrypt messages from the past if they recorded and stored them.
Well, anyone could yes. That's why if you want messages you're sending today to remain secret for longer than a decade, you use a cipher that isn't fully broken by quantum computers like AES. It's really just the key distribution methods that will be broken by quantum computers, they only get a sqrt(n) speedup against symmetric-key cryptography: double the key size and you're golden.
So anyone who questions quantum computing is a hater? Your argument reminds me of one interview where Elizabeth Holmes said everyone was a hater for doubting her. John Martinis didn't silence anyone, after decades, still there's no verifiable proof of quantum computing.
>...along with helping to develop curriculum for students in the quantum computing field to help expand the workforce in this area.
How do you develop a curriculum for something that has not been invented yet and might not even be possible to expand to a useful level? How can you have a workforce?
Sounds like someone was just cutting and pasting from a general policy...
...and by not letting anyone not descending in a straight line from the Mayflower work at it, they make sure it's second-rate, at best.
Of course if he also succeeds in abolishing universities' non-profit status as he's recently started adding to his culture war talking point reservoir, not even Americans will be there to run this thing.
It's tempting to dismiss that write-up due to its incendiary style, but I find that many of its "factory floor" level insights ring true to me.
As an experimentalist, I would agree with the sentiment that the potential future applications of Quantum computing probably receive too much media attention, given the maturity level of existing technology.
The technical arguments as to why building a useful QC will be impossible are a bit more shaky.
First, the post seems to imply that the calibration effort scales with the number of gates in the algorithm. This is false. In reality, the number of interactions that need individual calibration is basically the number of qubit-qubit pairs exposed by the gate library. The most mainstream approach to error-corrected QC only uses nearest-neighbor interactions, and hence the scaling is linear.
Second, it is not clear to me why the number of computational basis states (2^N) is relevant to the engineering at all. Following this line of thinking, the recent Quantum supremacy result by Google already amounted to a mastery of 2^53 ~ 10^16 degrees of freedom.
Third, an argument is made that large-scale error correction will not work because of correlated errors, of unspecified nature. I think a claim like this should come with a mention of at least one concrete source of such errors, so that we could estimate its magnitude and potential severity. Note that such calculations are almost the essence of day-to-day work of a physicist.
In the absence of more detail, I can make a generic counterargument: Any phenomenon causing such correlated errors by definition affects multiple physical qubits at once, and will tend to be more macroscopic in nature. This is in contrast with the processes that limit the fidelity of one- and two-qubit operations, which is what the error-correcting code will take care of. Macroscopic disturbances are exactly the ones that we can attempt to shield against with clever engineering.
John Martinis emphasized in his talks that they didn't observe any correlations between local errors in the Sycamore and the total error can be computed through the "high school probability theory" from the local errors. This pretty much silences deniers who used this totally made-up argument against quantum computing.
Google (or anyone else) hasn't shown an implementation of an error correcting code, so we do not have data points for a model-free "ruler extrapolation" of logical error rate vs. lattice size.
In fact, I think the Sycamore qubits were "pre-threshold", i.e. no error correction gain possible even in theory. I wonder if someone will correct/confirm me. I remember the readout fidelity was particularly poor.
Furthermore, I would argue that the large readout errors make the observed scaling of total error slightly less impactful.
But don't get me wrong, it's still a monumental achievement.
Yes, to an extent, but the case is not closed by any means. From memory, the logical error rate target is 1e-12. Clearly, quite a
sizeable leap of faith is needed to extrapolate from the Martinis results to this value.
Since we're throwing trillions around these days (needed, absolutely,) I'd quietly sneak a few billion is vital things that are harder to fund in normal times.
OK, here are $x Billion, make us #1 in this field.
It's common to refer to an administration by the name of the current President than "the US government," particularly within the US where the Hill is based. You can find numerous references for example to the "Obama Administration, Clinton Adminstration, Bush Administration, etc" in the media.
There's no "agenda" at work here, you're just being needlessly paranoid.
> "The White House Office of Science and Technology Policy and the National Science Foundation (NSF) announced Tuesday the establishment of three quantum computing centers across the nation, involving an investment of $75 million."
It was a White House initiative, so the headline seems to be accurate.
The Hill is focused primarily on Washington politics so 75% of the stories are about the Administration/House/Senate/Supreme Court/Agency putting out a press release or taking some procedural step.
Turn on any NPR News broadcast over at least the past two decades and the first story will invariably start "The ______ administration…", regardless of who's in office.
Meta: whilst the title itself is a standard, neutral way for a political news site like TheHill to report a government's decisions, this is one case where editing the title for HN submission might help focus the discussion on the quantum computing initiative rather than the Trump administration. I say that as someone who's considerably more enthusiastic than the average HNer about content on here which is inherently political in nature
Ignoring all the political stuff, you do have an interesting point.
How much quantum computing will $75 million buy the US government? It doesn't seem like that much money, given that the Summit super computer at Oak Ridge cost $325 million to build.
$75 millions seems like the amount you'd need to start renting office space and hire researchers for a few years. I don't think it will be enough to fund building any actual quantum computers. Especially given that we're talking three centers, not one.
> bipartisan
Also by coincidence (?) the first time I've seen that word used recently in a context that wasn't preceded by the words lack of or similar.