Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
A $1.2B law to boost US quantum tech (technologyreview.com)
232 points by jonbaer on Dec 23, 2018 | hide | past | favorite | 83 comments


Link to the actual legislation: https://www.congress.gov/bill/115th-congress/house-bill/6227

So, whoa - I hadn't even heard of this. Had y'all?

$1.2B on an exotic tech field - with implications that bigly threaten online privacy and security - seems like an incredible thing to get through both houses and signed into law without it being a story in these parts.

Amash unsurprisingly voted no (joining only 10 others in the house, while 348 voted aye). A quick perusal of his twitter doesn't show me any of his thoughts on it. I'd like to hear them.


>with implications that bigly threaten online privacy and security

Well, it never hurts to start researching post-quantum crypto even if large quantum computers were to prove impractical. I recall Google ran an experiment with a post-quantum key exchange algorithm [1], though I have no clue what came of said experiment.

[1] https://security.googleblog.com/2016/07/experimenting-with-p...


This is ongoing, and Microsoft and Google continued browser experiments with another algorithm (paper titled "Frodo: Take off the ring! Practical, Quantum-Secure Key Exchange from LWE") https://eprint.iacr.org/2016/659 - you can also see a recent video from DEF CON https://www.youtube.com/watch?v=SCdaYAFx7hw

In both cases, they are using a type of lattice-based cryptography known as LWE (learning with errors).


It is always an arms race.

India now allows the government to decrypt personal data:

https://www.zdnet.com/article/india-authorizes-10-agencies-t...

If anything it should tell us that Crypto works(for us whose math skills may not be up to the task) and there are not as many backdoors as technologies in the past.


Fortunately, post quantum crypto is pretty far along-- meaning cryptographic primitives that cannot be broken by quantum computing.

The cryptography challenge has seen huge progress; a surprising amount of it from one guy, DJB, an absolute legend.

What remains is a hard software engineering and standardization challenge: upgrading our apps and protocols to use quantum secure crypto.

https://pqcrypto.org


There have been a considerable number of other people involved: the complete list of people in the NIST post-quantum submission teams is pretty big.


There was a bit of discussion back when this bill was introduced: https://news.ycombinator.com/item?id=17524928


>>$1.2B on an exotic tech field - with implications that bigly threaten online privacy and security

Probably more than that...even regular computer threaten privacy and security, in 0.2 seconds they scan all faces or fingerprints in the world.

Congress probably doesn't want USA to be left behind.


I am amazed by the negativity and conspiracy theories here.

No, thanks to post-quantum encryption there is little to worry about privacy threats from quantum computing. At most, 15 years from now today's recorded messages will become decodable.

On the other hand, the advances in chemistry/material design/drug discovery/metrology thanks to quantum computing will be an amazing quantum leap [pun intended] from what we have today.


You mean ALL SSL traffic will be decodable?


Yes, if someone stores today's traffic for decoding in the future. This is not a threat model which applies to most communication. And we will easily move to quantum resilient encryption, so the future SSL traffic will not be decodable.


Metrology is actually the only field that might produce any results; everything else is just marketing.


Quantum computers will (obviously) be great at efficiently simulating quantum mechanics (which is practically impossible today). Which will lead to amazing new chemistry, which leads to drugs/cures/new materials.


FPGA/chip place and route would like a word with you.

Also, all sorts of physics solvers for optimum configuration used in aeronautics, heat flow analysis, etc.


Glad someone slipped that in... the Chinese government is partnering and investing in all sorts of public works, tech fields, moon shots and countries (infrastructure in Africa), while the US is losing ground on big ideas, solving big problems and welcoming international talent. If there were a positive use of nationalism that didn't precede war: More of this, please.


The Chinese are already conduction all sorts of quantum experiments in space, and from Earth to space and vice versa. On some occasion even in collaboration with nation actors such as Russia. The US is lagging behind so far, the discussed dollars won't undo this.


That satelite is a huge waste of money, very little scientific value, a vanity project.


Do you have any sources for these statements?



That article does not indicate that U.S. is way behind.


Here's a great report from the National Academy on progress in quantum computing https://www.nap.edu/catalog/25196/quantum-computing-progress...


Looks like an arms race to me. Whoever can get to it faster gets to decrypt the existing algorithms. Can someone more knowledgable validate this?


It sounds like the money is going through the NIST and NSF, so presumably any resulting research will become public. If the government wanted to keep the research secret, I would expect the money to go to the CIA or somewhere else.


Typically it would be the NSA that worked on secret cryptographic technology, although this might distract them from their charter of bugging civilians... (Having a quantum computer may not help with that because they would probably want to keep it secret and not build enough to decrypt everybody's SSL)


There are two major points to clarify this characterisation:

· Existing algorithms are not proven to be secure, even classically. Quantum encryption provides an additional level of safety, but against unknown and unexpected attacks.

· Practical quantum computers are very speculative, much more so than nuclear fusion or AI. They might not be built in the next million years.


seems like you’re getting downvoted for expressing quantum computing skepticism on Hacker News.

I bet $150,000 that by January 1, 2040, a quantum computer will not exist that is able to factor a 2048-bit product of two primes. Anyone want to take me up?


NIST thinks that 2048 bit RSA will be secure until 2030. Presumably their analysis does not include a quantum computer, because if it did they wouldn't recommend 3072 bit RSA for security beyond 2030. Their analysis is indexed against DES (so cracking 2048 bit RSA using the best known methods would be equivalent to cracking a hypothetical 112-bit DES).

It's possible that you're right that a quantum computer won't exist in 2040 that will break 2048 bit RSA, but that doesn't mean it won't be broken anyway by then.


Of course, for all we know factorization is in P or even NC (as the GP notes, we don’t really have proofs). I’m more hopeful we will find a proof of P!=NC or P!=NP than there being a quantum computer that beats RSA2048 in 21 years.

It aught to be lower-hanging-fruit to prove that factorization is not in NC (polylog depth, polynomial component circuit) and therefore NC!=NP => (P!=NP or NC!=P) so at least one of the two probably true statements is true...


this is like saying the same about computers in the 60s. there are many applications besides cryptography, and there is already fairly robust research into post-quantum crypto.


does post quantum crypto require a quantum computer to implement?


No, that is the main point. Quantum computers break some forms of asymmetric encryption, but there are "post quantum" protocol that you run on classical computer and neither a quantum nor a classical computer can break them (a conjecture, as usual with complexity theory).


thanks!


Still this works the best: https://www.xkcd.com/538/


First you need to find out on who to apply the $5 wrench. That's a lot easier if you can monitor all communication.


Glad that President Trump signed it. Will foster new substantial research and development in material science as well as the mathematics for the QC field.

I personally think health-benefits from Quantum computer modeling of chemical reactions, and protein folding will bring us much closer to 'per-person' pharmaco-genetics than current methods.

Also having a specific government structure established and dedicated to this field is important.

"...

The law also establishes a National Quantum Coordination Office,

calls for the development of a five-year strategic plan and establishes

an advisory committee to advise the White House on issues relating to quantum computing.

..."

http://androidtechnews.com/trump-signs-legislation-to-boost-...


Where is the funding going to come from?! As a non-American, I get worried by all this funding when America is in serious debt. :(


I don't disagree with the important question of "Where does the money come from?" (Perhaps, it's also, "What didn't we fund because of this?") But FWIW, America is not unique in its serious debt situation and as a percentage of GDP, it's not unusual. Many EU nations are in the trillions of debt ballpark. Also China, Canada, Australia, etc. In fact, some of those nations exceed America in national debt per capita. [1]

I write this as a Canadian. We're also a nation of serious debt and I sometimes wonder where money for my government's big projects comes from...

[1] https://en.wikipedia.org/wiki/List_of_countries_by_external_...


From the link it seems that all major economic powers have a large "external debt". Then who are the creditors? If every country paid their debt, where would the money end up?

And I see from the article that "... while a country may have a relatively large external debt (either in absolute or per capita terms) it could actually be a "net international creditor" .

Wouldn't it be better to focus on this "balance of debts vs. assets?"


>Then who are the creditors? If every country paid their debt, where would the money end up?

At least in the US, most of the funds would end up in private citizens or corporations in the form of treasury bond repayments.


>Then who are the creditors? If every country paid their debt, where would the money end up?

Bond holders. The us treasury holds bond auctions all of the time to raise money.


Then who are the creditors?

The best analogy I think is that we borrow the money from our future selves. If we invest it well then our future selves will enjoy the wealth in retirement. If we fritter it, we will be destitute in our old age.


Accountants and economists will think in terms like cash flow, net present value, opportunity cost, etc. In that way of thinking rationally you want as much debt as you can carry. State actors with control of money supply are even less worried about solvency.


The governments owe money to the public.


The debt between European countries started with the second world war.

I was taught that this is one of the strategies chosen to discourage further wars, as declaring it would destroy billions of your own money.


How is destroying billions if virtual currency a bigger discouragement that destroying billions of real, physical assets?


I agree with your point. There should be some skepticism that debt discourages, rather than encourages war. In more militaristic or authoritarian nations, if the debt were great enough, it could be an incentive to invade the neighbor/nearby country you owe it to in order to destroy it (assuming that scenario).

One of the several reasons Iraq invaded Kuwait, is that Kuwait refused to forgive $14 billion in loans (to Iraq) accrued during the Iran-Iraq war. Iraq's annual GDP in the 1980s was $40b to $60b depending on the year, to give you some idea of their ability to deal with the $14b in debt held by Kuwait (it was a considerable sum to them in the 1980s; it might be like the US owing Panama or Singapore $5 trillion today).

https://en.wikipedia.org/wiki/Invasion_of_Kuwait#Dispute_ove...


Another point of context, the US spent $2.4T on the Iraq+Afghanistan wars.

Invading Kuwait over $14B debt is a strange leap, since Iraq could far less expensively simply default. Anyway, it's the wrong direction -- $14B debt to Kuwait would make Kuwait less interested in invading Iraq for risk of forcing default. But either way, the value of the oil spoils (in either direction) dwarfs the cost of the debt.


Because generally, the rich are the ones who miss out in the former, while the poor miss out on the latter. Someone whose house gets destroyed will lose out a lot more from that, then from some small fraction of their pension fund going belly up because the belligerent nation has stopped making bond payments.

Since wars tend to be started by the rich (and their proxies), and fought by the poor, this puts skin in their game.

At least, that's the reasoning. I don't really agree with it.


External debt is not the same as national debt and does not in it self say very much about the health of the public sector economy, since it includes private debt including accounts held in the country by foreigners for business reasons, tax evasion etc.


Debt is irrelevant when you can literally print money out of thin air. There are many creative ways of doing so, particularly when you're a global superpower


Not sure why this is being downvoted. You are correct. I suspect the downvotes are because people don't read carefully and assume you are endorsing it, despite the fact that nowhere did you say it was good or ok, simply that they can.

There have already been indications that the US intends to print their way out of debt. We have already printed billions over the past few years. If you look at graphs and charts of inflation, it's breathtaking.


Your last paragraph is very misleading. While it's true that the US has printed billions of dollars in the last few years, that is not all that significant when you consider that we have trillions in circulation. Also, our inflation rate is far from breathtaking, and has been consistently hovering around 2% a year for decades (https://fred.stlouisfed.org/series/T10YIE)


I'm not generally a "don't trust the government numbers" kind of person, but in the case of inflation their numbers are wrong. They don't capture true cost of living changes and the reduction in purchasing power.

This article gets at some of the things I had in mind when I wrote my comment: https://www.investopedia.com/insights/how-will-fed-reduce-ba...

Also I apologize for lack of coherence. I'm very tired and should have gone to sleep 6 hours ago.

Side note: I really like your user name :-)


The Bloomberg article you linked in reply to another comment about the mis-measurement of CPI was very interesting, but even the alternative measurements discussed in that article estimated inflation to be below 3%.

The Investopedia article discusses how the Fed will reduce their balance sheet, which is actually more or less the opposite of printing money. During the Great Recession, the Fed bought a lot of financial assets (i.e other people's debt) in order to restore faith in the economy and prevent a complete meltdown. Now, they have to decide whether to sell these assets or hold them to their maturity. Regardless of what they choose, they will be receiving cash either in the form of the sale price or dividends.

The Federal Reserve has a mandate to keep inflation rates low, and they generally do a very good job at that. That doesn't mean we shouldn't worry about our national debt, but I wouldn't be too worried about hyperinflation now or in the near future.


> If you look at graphs and charts of inflation, it's breathtaking.

I don't see what you're seeing. Inflation has been close to (the Fed's target of) 2% for the last 10 years. That's much lower than its historical values. So if anything the US has chosen less inflation in recent times.


If the 2% is accurate (which I don't believe it to be. There's no shortage of people much more educated and informed than I that are talking about this), then it's certainly not "breathtaking" as I said. I had in mind a graph (which I inconveniently cannot find now) that showed inflation over the last 100 years prior to the US coming off the gold standard. There was basically none, but it has since sky rocketed.

I did a quick google and found this right at the top: https://www.bloomberg.com/news/articles/2018-05-02/the-consu...


> inflation over the last 100 years prior to the US coming off the gold standard. There was basically none, but it has since sky rocketed.

Sure, but I don't think that the high inflation between 1934 and 1976 can be used to judge the current system with an independent central bank and an explicit 2% inflation target.

> I did a quick google and found this right at the top: https://www.bloomberg.com/news/articles/2018-05-02/the-consu...

The proposed measure merely claims to be faster to respond to changes in prices. Note how the new measure is lower than CPI during the crisis, but higher elsewhere. In the long term they should be the same. So I don't think that this suggests that inflation is being systematically underestimated.


There's no debate about the obvious fact that intentionally inflationary currency has less inflation than intentionally deflationary currency. The debate is whether the inflationary currency generates more economic growth than deflationary currency (by stimulating productivity in otherwise idle labor/capital), and most economic policymakers believe it does.


While you can certainly keep increasing the monetary supply, I wouldn't say debt is irrelevant - the costs of servicing that debt will start getting high for the US. Debasing your currency is one way to erode that debt but it will come with its own problems


No, you can't, if you care about future generations. Printing money is literally stealing from the least priviliged and their already vanishing savings.


A few billion is literally nothing compared to our overall balance sheet. If you’re worried about American debt, you need to look at the pentagon and social security, not our research efforts.


Why are you worried if $1.2B of American taxpayer dollars is going towards quantum tech if you're not an American taxpayer?


America isn't in serious debt. But let's pretend for a minute that it is -- so, how do you intend to make them pay?


When people talk about US debt, they're not talking about something like your credit card bill. When the US needs to take on debt, they do it by issuing Treasury bonds.

These bonds are a contract saying that in exchange for your money today, the US government will pay you a fixed interest rate every month until the bond expires. In order to raise the money to pay off these fixed interest rates, the US government can either tax its citizens or issue more bonds.

For the past hundreds of years, our government has fulfilled its contracts and paid off the interest on all of its bonds. If we ever neglected to do so (AKA defaulted), it would be MUCH harder for our country to raise money in the future. Also, much of our government's debt is owned by American citizens, so we would be shooting ourselves in the foot, hurting domestic investors, and possibly causing a global economic crash.

That's what's making us pay, not some loan shark knocking on the White House door


In addition to the points you make above, having the world’s most trusted debt also helps to make the US dollar the world’s most stable currency, thus affording the US enormous influence in global fiscal policy. If contracts around the world started getting settled largely in Euros or Yuan, the US would find itself in a much less powerful position at the global negotiating table. As with most things in life, those who control the money have the power.


Well, obviously the US is not going to tell its creditors to go screw themselves. They'll just re-structure the debt in a less favorable for the creditor way. And the latter will agree to that because it has no better options.


That will still tank the economy.. creditors no longer have liquid funds and trust is gone. Any future bond issues will have very interest rates, while the debt now is not actually very large compared to gdp in the U.S and is quite serviceable, if the interest rates become high it will become difficult to service it causing runaway effects and future investments/spending will be hampered leading to all sorts of problems


We almost did that by accident a few years ago. It was not a good idea.


>> If we ever neglected to do so (AKA defaulted), it would be MUCH harder for our country to raise money in the future.

Really? Do the creditors have a better option?


Money isn't real. It is a promise to produce value. If the promise is broken, the alternative is to stop playing the game.


So if your whole wealth is in US dollars, how do you stop playing the game?


From the appropriation bill. Don't worry about the debt. The debt is used to create money.


Or did he .... We won't know until someone opens the box ...


But still predetermined at the moment of big bang.


well maybe in this universe, but not one next door where the box contents were different, or in the much more sensible universe where they put him in the box and didn't open it at all


This likely has much more to do with the U.S. government wanting to break encryption than say...find cancer treatments. Quantum computer efforts were probably already funded in a much smaller way through the NSA's black budget, but they must of thought they might as well make it public and give it a much bigger budget.


Too little too late. The dialectics of lead are not likely to be surpassed by #2, since being #1 applying this is highly likely to be inherent.


di·a·lec·tic /ˌdīəˈlektik/Submit PHILOSOPHY noun noun: dialectics 1. the art of investigating or discussing the truth of opinions.

--------

I really don't understand your use of that word here.


OT: 'dialectic' and 'dialectics' are two of the most misused words that I see on regular basis. It is as if people want to use that word in particular, no matter the subject. Really interesting phenomenon.


Sometimes people just like to dialectic a word into a sentence, even if they don't know what it means.


Congressman, did you or did you not just use the word 'pronk' as a number?


There are several other definitions, though I agree the word does feel awkward here regardless of which one you choose (tho some are better than the definition you used). [1]

[1] https://www.merriam-webster.com/dictionary/dialectic


Could you elaborate on this?




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: