'I give fusion power a higher chance of succeeding than quantum computing' says the R in the RSA crypto-algorithm As they do every year, the names behind the RSA crypto-system took the stage in San Francisco for the RSA Conference's crypto panel, sharing their thoughts on the pressing issues facing the information security world. Ron Rivest (the R in RSA), Adi Shamir (the S), and Whit Diffie (of Diffie–Hellman key exchange fame) were …
>Cold fusion? Probably not
There's an interesting Outer Limits episode where a university student invents cold fusion.....
An ex-student returns to blow up a university with bombs he's made by cracking the secret of cold fusion, after being booted from the school's physics program. Todtman swears he won't follow through if authorities execute 5 people he hates. Most physicists believe cold fusion is impossible, so Todtman detonates a demo bomb by remote control. Should the government kill his 5 or risk Todtman killing 5 million?
I'd go with option 3. Ask for a demo of it (only got a bomb? Fine, we've got test ranges for that) and tell him that if it works then for a major scientific advance on a level not seen since Einstien he's awarded a pass with honours by government fiat, and has a job offer for a million a year with a thirty billion budget a year to build it outputting commercial scale energy, as he's just obviously singlehandedly solved the global warming problem if he can build something with the power of an atom bomb in his flat on a shoestring budget.
Large amounts of cheap CO2 free energy solves basically all of the worlds problems.
Water levels rising? Split sea water to hydrogen and oxygen on a large scale. Massively energy intensive of course, but if you've got practically unlimited cheap and CO2 free energy...?
CO2 Sequestration? Again, energy intensive and therefore currently expensive, but if we have practically unlimited cheap and CO2 free energy...?
Not enough drinking water in Africia? Stick in lots of desalination plants and turn the huge desert into a garden. All you need is fresh water and an oil style pipeline to move the water around. Making a new continent's worth of vegetation would also have a quite significant effect on atmospheric CO2 when it starts getting pulled absorbed by the plants. Again, possible if you have practically unlimited energy because just a matter of scale rather than a matter of new technology.
I'm not sure that qualifies as "interesting" there - just the Trolley Problem with a different backdrop.
And the best revenge on your university, after it ditched your project and you solved it on your own, is not blowing up the university, or even killing the people responsible. It's demonstrating your results, selling your results, making enough money to buy small countries, and being able to say at every press conference: "And those morons at university could have had a slice of this. Professor Smith and Professor Jones, I bet you feel stupid now." Having them remembered not for anything they achieved in their academic careers, but for being the people who were too stupid to spot a world-changing innovation - *that's* revenge. Served very cold indeed.
>Trolley Problem with a different backdrop.
The Trolly Problem is nothing new either, people have being making those decisions for centuries, especially in wartime. Someone running a philosophy course re-invented the wheel, gave it a trendy label in the 60s and thought themselves a smart arse.
PS have you even seen it to offer a critique as there is more to it ?
I'm not sure that qualifies as "interesting" there - just the Trolley Problem with a different backdrop.
It isn't. The AC's description is simply lacking.
The crux of the Outer Limit's episode is that someone discovered building a city-destroying bomb is inexpensive and technically simple. It is inevitable that others will make the same discovery, the knowledge will get out, and every psychopath out there will soon have as many nukes as they feel like making.
It's interesting that small startups are now working on fusion. Perhaps this is going to be like the spaceship field with rapid innovation and successes. The few companies I've read about are taking interesting and different approaches to the problem - but all are looking at systems much smaller than un the past.
And even with that distance we rely on having a free source of magnetic shielding from certain radiation and ozone sheilding from UV. There is a lot of things lined up for us to have it easy.
And it does raise serious questions about the realistic safety of fusion power when we figure it out, it still might be pretty dirty, most likely less than fission, but there does not appear to be any free lunches in physics.
We have contained fusion at the moment: it's been figured out.
The current models are of course lab test sizes and have previously required more energy for containment than they generate and are only run for a couple of minutes at a time, but we do have working (electro)magnetically contained fusion. Some of it even generates a net gain (as in more energy generated from it than is used containing it) hence why all of a sudden a lot of companies are interested in building fusion plants.
Remind us again how they plan to shield everybody from the neutrons produced by D-T fusion, and where we are going to get the T from?
To quote Wikipedia:
"As can be seen, the easiest to ignite of the aneutronic reactions, D-3He, has an ignition temperature over four times as high as that of the D-T reaction, and correspondingly lower cross sections, while the p-11B reaction is nearly ten times more difficult to ignite."
I'd like a citation for this "sudden...interest in building fusion plants." Because at the moment it looks like pure fantasy
The T is produced in the reactor itself. The Li blanket surrounding the vessel is bombarded by high energy neutrons from the reaction, producing T. This is similar to how it is currently produced in the Watts Bar 1 reactor by using a burnable absorber rod containing LiAlO2 pellets. High energy neutrons from the reaction bombard the pellets, producing T by thermal neutron irradiation.
I'm in Free-UKland, it is February (just barely) and the local fusion reactor is keeping the room's temperature marginally above Absolute Zero.
The heating element that uses electrical energy made from stored fusion energies (fossilized plant and animal stuff mostly) - the desktop PC unit - contributes a share to the room's "warmth", too.
For you to be "toasty warm" you must be in daylight and on the bottom end of the planet. Yes?
"For you to be "toasty warm" you must be in daylight and on the bottom end of the planet. Yes?"
Not necessarily. I'm in Sonoma California, it's just gone Sunset (630PM) and I have had the office zone of my HVAC turned off, and a fan in a window blowing fresh air in since 10AM. Smells of cherry blossoms, and the temperature in here is ~75F (~24C). I'll probably return the HVAC to normal operation in an hour or so.
Well I'm with him [Ron Rivest] 100%, I've never believed you will get anything really useful from quantum computing. I would go as far as saying you are more likely to see cold fusion succeeding. ....... druck
Well I'm with Clarke's First Law: When a distinguished but elderly scientist states that something is possible, he is almost certainly right. When he states that something is impossible, he is very probably wrong, druck.
And be a huge fan of Laws Two and Three too. ..........
[2] The only way of discovering the limits of the possible is to venture a little way past them into the impossible.[3] Any sufficiently advanced technology is indistinguishable from magic.
Proving something is formally impossible has been done many times. It includes not only relatively recent theorems like the Halting Problem, the Uncertainty Principle, Chaitin's Omega, and a vast number of others (FPL and PCL are a couple of my favorites), but also older ones such as the impossibility of squaring the circle (which is actually relatively recent too, come to think of it, and is a corollary of the proof that pi is transcendental). And, of course, any positive theorem has an "impossible" converse: Euclid proved the infinitude of prime numbers, which means he also proved the impossibility of finding the largest prime.
There's a tremendous difference between those sorts of formal proofs and claims that something is practically impossible. In the latter domain there are some areas we may be fairly sure of, at least on the grounds that a practical demonstration otherwise would have such drastically unfortunate consequences that we needn't worry about being proven wrong - there won't be anyone left to mock us. Those include the impossibility of perpetual-motion machines (cooked by over-unity runaway energy production) and faster-than-light travel (breakdown of causality means all bets are off). In others, though, even the highly knowledgeable are often just making largely-unfounded guesses when they declare something impossible.
Personally, I'm still dubious about the economic and to some extent the physical feasibility of practical QC (real QC, not adiabatic shit), but I admit they've scaled up faster than I expected.
'faster-than-light travel'??
note that 'speed of light' is not about light... it merely means you got there faster than light would have...
all of this is STILL based on a 50 year or older theory!! when we ** properly ** get into space, we may find something totally unusual happens!!
until then, its just stupid presidents and limited budgets keeping us down here... >:(
'man will never fly'
'man cannot survive speed in excess of 20 MPH in this new-fangled automobile'
'we will never travel faster than light' then how did UFOs get here??
the problem with most scientists, is that they will not believe anything except **absolute** truth.. they only have 'theories' ... It may be ststisicall prossble that life is on other planets, it is *only* a theory until an alien walks up and shows them their spaceship...
what is the USE of fusion power ?? we can then stop polluting the planet with carbon and radioactive waste..
I remind people that the current 'quantum computer' is covered with SOOOO much chatter, when AFAIK it is not really more advanced than the original 'turing computer' ... It can just do millions/billions more computations in a very short time...
GO ON, geeks, do explain it ...
LOLOLOL... :D :D
I found a link that *may* explain the 'quantum computer' problem..
https://www.technologyreview.com/s/615180/quantum-computer-race-ibm-google
it can do something a normal computer would take 10000 years to do - BUT!! it all has to be cooled to absolute zero, meaning it is very big..
these are usually math problems, etc.. we will have to wait for them to do something 'worthwhile', like Turing's pattern recognition machine that helped the war effort..
and until we can get 'room temperature superconductors' it will not be useful in mainstream computing!!
... *only* a theory ...
I don't think that word means what you seem to think it means.
Incidentally, there are no "absolute truths" in science - science doesn't work that way. For absolute, unshakeable truth that cannot be pulled down, no matter the evidence, you need religion.
Just saying
... they only have 'theories' ...I'll just quote the perfectly clear and simple wikipedia entry on what a theory is as opposed to your completely ignorant interpretation (emphasis mine):
In modern science, the term "theory" refers to scientific theories, a well-confirmed type of explanation of nature, made in a way consistent with scientific method, and fulfilling the criteria required by modern science. Such theories are described in such a way that scientific tests should be able to provide empirical support for, or empirically contradict ("falsify") it. Scientific theories are the most reliable, rigorous, and comprehensive form of scientific knowledge, in contrast to more common uses of the word "theory" that imply that something is unproven or speculative (which in formal terms is better characterized by the word hypothesis). Scientific theories are distinguished from hypotheses, which are individual empirically testable conjectures, and from scientific laws, which are descriptive accounts of the way nature behaves under certain conditions.
Theory?? I think you need this one..
https://dictionary.cambridge.org/dictionary/english/theory
eg
"If something is possible in theory, it should be possible, but often it does not happen in that way:
In theory, the journey should take three hours, but in practice it usually takes four because of roadworks"
as is often said, theory is totally different from 'practice'
Actually it was Goethe in Natur und Kunst who wrote
In der Beschränkung zeigt sich erst der Meister.
In effect he says that the Master shows himself by working successfully within limitations.You could say that's far more appropriate for Turing with his Baby computer that used absolutely minimal resources (he planned to use war surplus vacuum tubes like EF50s, and delay lines filled with alcohol rather than mercury) than the vastly expensive proof of concept quantum computers that so far haven't delivered actual computing. They may get there...but one of the three people who rediscovered Cocks's work that was concealed by GCHQ surely has a right to be sceptical.
This post has been deleted by its author
Rivest has worked on a lot of things. I expect he's interested in a lot more.
And, to be fair, we don't know that factoring is a hard problem for conventional computers. The complexity class of factoring (for DTMs) is unproven. It's clearly in NP and co-NP, but P is a subset of NP (so it might also be in P), and it's suspected to not be in NP-Complete (or NP-Hard, which is a superset of NP-Complete, assuming P != NP). It's subexponential, but no one knows if there's an efficient classical algorithm for factoring large semiprimes.
Obviously it's in BQP (Shor's algorithm).
This post has been deleted by its author
Ireland bought voting machines and eventually scrapped them. Paper and pencil isn't perfect, the STV PR isn't perfect. But voting machines, the US Electoral College, Israeli list for entire country PR and UK FPTP are all worse.
Even though we can't currently form a government. That's not to do with PR, but Civil War politics and one party being far better at Social Media and traditional media PR. Belgium didn't come to a halt because it took ages to form a government.
The USA State (Washington) having online smart phone votes is mad. A signature on a low resolution touch screen? Making a paper print out of it isn't an audit trail.
The ancient Athenian idea of voting for a pool of candidates (on paper of course) and then randomly by lot selecting the final officials (to avoid corruption) has some merit. Perhaps combine that with the PR STV larger constituencies.
"That's not to do with PR, but Civil War politics"
The two are not mutually exclusive. AIUI one of the effects of FPTP is to exaggerate* the ratio of votes when determining the ratio of seats. That's likely to make a stalled outcome less likely.
* I remember reading a long time ago that the ratio of seats is proportional to the ratio of the squares of the votes.
The Electoral College was explicitly *designed* to somewhat dilute the power of big population centers, and it does that just fine. It's working as designed. Some of the states wouldn't have joined the union without it (nor without the two Senators per state design of the Senate).
Just because the "right" (i.e. Left) people didn't like the outcome this last time doesn't change that.
Ironically, it was the losers who were floating a plan to manipulate the College. When that failed, they decided that they hated it again.
The EC was for two reasons:
Dilute the power of big population centers.
Be a check against an unqualified president.
The former is of questionable value (the senate does that very well, and the reforms I'd propose for that would not eliminate that while making it much more representative: it should be multi seat electorates with a preference system of voting) and the later is apparent to not work.
Things that absolutely have to change for the US to be a democracy would be removing the many months of lame duck governments (there really is zero reason for these to exist, and the emergency things that do need immediate decisions can be done and require a vote once the new government sits) and centralization of election administration so there can be confidence in the election process.
Even if (stop sniggering at the back) the software is flawless and the hardware solid.... Cosmic rays can still bugger things up with a nice bit flip in the wrong place at the right time (in fairness any and all voting machines should be using full ECC memory but what do I know) as I was listening to this morning available here from the Radiolabs broadcast which did have me wondering why anything which has any large scale implications (death, destruction, being voted in) should have as a matter of course but doesn't. I mean it's not like ECC is that horrifically expensive that it's not worth the extra outlay to be safe surely?
"Earlier I said some nasty things about the people in charge of the visa process. I would like to ask for this to be forgotten."
It's annoying when somebody touting himself as an expert gives his supposedly informed opinion, then shows he doesn't really understand the issue, or doesn't care, as long as he wants to spout a funny-but-wrong witticism.
The right to be forgotten is not, and has never been an absolute right applying to every and all things one did or said. Of the 50000 requests Google receives, as quoted, they turn down plenty, as the law allows. About half of them, if memory serves.
If people cannot get jobs because they have a criminal record, I am not sure how enforceable or fair a right to be forgotten would be in the long term. Is it really the best way to move society forwards? If I were voting for my MP or someone, I would expect their criminal record to be publicly available. Security services also need to vet candidates for such things. It is more applicable to some jobs than others. We don't fight other forms of discrimination by forgetting race, gender or age. Perhaps employers should be obliged in law to demonstrate that any refusal for a normal job was on other grounds - and then monitored/sanctioned for not fulfilling their quota of old lags.
The whole point is that the right is not a blanket "I don't like this. Take it off the internet". The problem is that, search engines specifically, are not clever enough. They aren't nuanced enough to construct and proper and accurate time line to realise that the very newsworthy arrest you experienced ten years ago was followed by the case being thrown out or never prosecuted.
Many jurisdictions also have convictions expire after a time, once the fin has been paid or time served. The "right to be forgotten" doesn't mean all this magically disappears from history, it just means that search engines have to be much more careful about what they present in search results and, additionally, have to conform to local custom, law and (cultural) standards/expectations, e.g. do you consider the legal/justice system to be punitive or reformative? The answer to that will often determine which side of the fence you're sitting on the right to be forgotten.
I believe that an applicant's criminal record should be available.
The Wall Street Journal had an article recently on a company that had a factory in Chigago with a work force that was 80% ex-cons. The conversations with the company's officers and the employees on what it takes to make that work were fascinating. It does require a lot more investment of time and effort to help these employees succeed in what may be a first for them: a steady job with a paycheck, having to be at work on time, managing finances and paying bills - a complete change in the way you live. But it works - these people are turning their lives around.
There are people out there doing God's work.
"It does require a lot more investment of time and effort to help these employees succeed in what may be a first for them..."
And then you get the ones who take all your investment and stab you in the back. Worst part is, it's hard to tell the two apart, and for some business, one backstab can break you as much as one missed day of production. And with so many potential employees available, you wonder why many employers wax cautious...
Well, given that 2019 research paper, where MIT (?) discovered they could teleport heat, you can see why Quantum Computing will only ever be a modern version of an analogue computer. Consider the 'heat teleportation" discovery from last year and consider what you're looking at in that system and why it could never work.
Take two neutral particles e.g. N1 and N2. N1 and N2 are *not* passing heat to one another, no teleportation of heat.
1) There are no real neutral particles by observation. Neutrons are not neutral, splay them out and they're formed from balanced charges. The smallest possible thing must be light, to be an electro-magnetic wave across space, light must have a splay of charges across space. So it too must be a balance of charges, not a single neutral entity. Since every neutral particle can ultimately be converted to photons, they must also be a balance of charges.
2) N1 and N2 are in motion, detect their position at t1, detect it again at t2 and its in a different place. IT'S MOVED ergo its in motion!
3) Since they are charged, there must be a force between them and that force must be electric in nature.
4) In order for that force not to move N2 with respect to N1, both N1 and N2 must be moving the same. So that motion 2) must be resonant electric. A sort of synchronized dance.
5) OK, so we have N1 and N2 particles doing a synchronized dance. But N2 does not move with respect to N1, so the N2 dance must end up at the same place with respect to N1. In other words we have a dance over an electric field that overall goes nowhere.
6) If they were not in resonance then a force would apply from one to the other as it moves and so you would be passing energy aka heat.
So quantum entanglement, the fact that motion effects correlate (whether that motion is up down oscillation, or angle of oscillation or spin or some more complex oscillatory motion), is simply because everything is doing that! The reason atomic clocks oscillate the same is not magic, its because they're doing this resonant dance. They're not going through every possible state till detected, they really are moving.
So you're quantum computer isn't in all states till it locks to the best state and it will find sub-optimal solutions for which better solutions exist. The proof is obviously to find better solution using current techniques and then prove that the quantum computer isn't going through all possible states, because it didn't go through the optimal one. I believe this has already been proven several times.
Likewise the spin-off quantum encryption is worthless. It's just filtering by that 'successful entanglement' signal, the same flaw in all entanglement experiments.
----
Look, there's lot of obvious things in that 2019 experiment staring you in the face:
Mass for example, you have a motion that returns to the same place after some time in some repeating pattern. i.e. a form of energy that is not in motion.... that's mass. And given the only thing there is electric in nature, it must be a motion over an oscillating electric field. That "stationary electric energy" must be mass because otherwise what is it!
If N1 and N2 were not in resonance then a force would be applied between them, and they wouldn't return to the same place, N2 would return to a different place with respect to N1, and N1 a different place with respect to N2. i.e. That component of the motion you call momentum is the same motion as mass. It's just whether it returns to the same place or not with respect to the observer!
The faster it moves, the less mass it has and the more momentum it has, up to the limit case where it is all momentum and no mass... that's light/photons, whatever you want to call that limit case.
And if you push N2 at resonance with N1 at resonance, then you have something to push against. The further N2 is out of resonance with respect to N1, the less of a component you have to push against and the harder it is for N1 to push N2. You wonder why it gets harder and harder to accelerate things, the faster they go, in a magnetic field? It's because that F/2 component becomes skewed more and more with respect to your accelerator coils!
And since electrons also do the resonance dance of 2), because they too keep moving when you measure their position, then any electric field you've made with electrons has been an oscillating field at resonance. You've never experimented with electric force, only an oscillating version of it at local resonance!
Electric force itself (the non-oscillating kind) must propagate infinitely fast, since all paths through space must be in resonance, the force must take zero time to propagate. And that makes sense of course, because the same force must bind those charges in photons that must occur from 1), and since photons travel at the speed of light, the force binding them couldn't propagate at the speed of light, it would be impossible.
And mass is all relative. A photon N2 has ~0 mass with respect to detector N1, but from the point of view of the photon, N1 has ~0 mass. From the photons point of view N1 is all skewed out across motion and has no mass. All of N1's energy is in its momentum! And ~0 of it in mass. At least from the point of view of the N2 light.
In the days of the Wild Wild West, one could 'do a crime and serve your time' and that was it. You pretty much were, whoever you said you were. Nowadays, everything is recorded and any sentence is a life sentence.
The only thing people can do nowadays to finish with the 'doin yer time' part, is to change their name and move to another country immediately after leaving jail.
"Blockchain is the wrong security technology for voting, I like to think of its as bringing a combination lock to a kitchen fire,"
What a marvelous quote.
The problem with blockchain is that it's a nifty technology that doesn't seem to be a really good answer to any known problem.
Allow me to point out that using quantum technology to crack encryption is an entirely different class of problem than fusion power. Cracking encryption -- especially for encrypted material you already have -- only requires your cracking technology to work when everything is just right. It can fail dismally most of the time and still be extremely useful on the rare occasions when it works. Fusion on the other hand would appear to need to work VERY reliably in order for it to be useful for anything other than rapid urban renewal.
Recently, there have been some very positive announcements about fusion power. And a lot of major corporations take quantum computing seriously enough to invest in it.
This, however, is not to say that Ron Rivest is wrong. Fusion has been ten or twenty years away before. And the challenges of keeping a quantum system isolated from the classical world rise exponentially with its size.
My opinion is that it's not safe to assume quantum computing will work, and it's also not safe to assume quantum computing will not work. Prepare for the worst in both directions.
1: " ... the cat was neither dead nor alive; after being alternately poisoned and not poisoned for over a century, it had become greatly pissed off, miffed, irked and annoyed and it was simply missing."
2: "... when they opened the box, they found neither an alive cat nor a dead cat but a highly confused dog."
3: " ... this experiment, when first performed in the distant Galactic Supercluster, resulted in the superposition of two entire cats. Cats, of course, having much mass this then became a matter of blast radius."
By the time ITER pops its Francophone cork-- full tritium-dueterium fusion experiments in 2035 after all, after several rounds of gratuitous PhDs are awarded, and even then only if the construction stays on schedule (in France?)-- we'll have broken the R of RSA with quantum computers. And even if ITER works, it will be 10-20 years before one sees Real Energy(r) as ITER will only demonstrate something more than break even is possible... maybe this time? unlike the previous umpteen fusion tries costing many billions of euros?.
It would be lovely to have affordable fusion power. Maybe it would help to have working quantum computers to calculate what is needed for successful cost effective fusion on a terrestrial basis. Probably means scrapping ITER before it lights up and building ITER2, ah, cushy tax payer financed jobs all around.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
