I don't want to defend anyone, but at the same time, if I see accusations, I'd like to see substance and reference points, not just "they're scumbags", end of conversation.
Please, provide some details on why you think they're like that.
Especially because they could have done a basic search for what the person was talking about. Not everything needs to be cited if it’s an easy search away.
Agree with not just this comment but your comments as a new throwaway (I assume?) account in general. Though I don’t care about the CCP enough to be as anti CCP as you. Tho Ofc I am anti CCP
Not everything is a question of "does it collect data", actually.
They're collecting _some_ value from individuals, whether you understand it or not.
Your comment reminds me of Facebook's disingenuous "we we're not _selling_ the data", which just a way of distracting from the real issue. This isn't about data, it's about yet another incremental stripping down of individual's control over their intrinsics.
You seem to agree that they don't collect biometric data, so I'm not sure exactly what "intrinsics" is being "stripped down" from individuals, in this case. Could you clarify?
I don't agree to anything, because I don't know enough about it. They SAY they don't collect data, they SAY it's a zero knowledge proof. Who knows about that.
What I know is that they're extracting some value from people who (like me) likely don't understand the implications. This isn't complicated.
Because you so confidently asserted that they were "all scumbags" who have no "credibility", I'd expected much more certain evidence about the nefariousness of Worldcoin.
I, like you, believe it is very important to remind folks that they should clearly understand the value prop of an exchange before they participate in it.
But I don't think that calling "scumbags" the folks behind any proposed value exchange (like Worldcoin) that you yourself (self-admittedly) don't fully understand the implications of, is a way to achieve that.
You asked me exactly what "instrinsics" are people being stripped from. So according to that article, people are being robbed of the ability of proving their own identity. If that startup succeeds, people will no longer be able to prove their identity other than thru them. This is really dangerous stuff.
You see, the mistake you made here is assuming that without understanding the details you can't come to any conclusions. But that is wrong. There's some things that I don't need to know the details of to accurately determine they're nefarious.
> If that startup succeeds, people will no longer be able to prove their identity other than thru them.
The article doesn't at all claim that the startup seeks to monopolize world wide human authentication!
For the benefit of "bystanders", the TLDR of the article is that (1) WorldCoin is collecting biometric data (not just a cryptographic hash of it), (2) Worldcoin switched compensation of its "operators" from a stablecoin to a brand new cryptocurrency, (3) perfecting the tech to authenticate being a human while preserving anonymity itself might be very valuable, and (4) the fact that brand new cryptocurrency is distributed to everyone might make the cryptocurrency itself very valuable.
I think 1 is valid concern if true; 2, IIUC, wasn't done "behind the backs" of anyone, so it's fine, 3 and 4 are very true. Thank you for linking the article!
> The mistake you made here is assuming that without understanding the details you can't come to any conclusions.
You are free to choose whatever standard of evidence you see fit, but personally, I will always choose to reserve strong judgement (like calling someone a scumbag) until I am confident that I fully understand the details.
> You are free to choose whatever standard of evidence you see fit, but personally, I will always choose to reserve strong judgement (like calling someone a scumbag) until I am confident that I fully understand the details.
You'll get played in life a lot.
Example: you were repeating zero-knowledge nonsense, whereas I saw right thru it. You get played a lot.
A "one way function" (ie, a hash) is not non-sense. It is entirely technically feasible (eg, Apple's NeuralHash)[0]. WorldCoin still maintains that participants can elect to submit only the hash of their iris scans.
> You'll get played in life a lot [if you reserve strong judgement (like calling someone a scumbag) until you're confident that you understand the details].
I mean, I can just refuse to play games I don't understand? For example, I don't know enough about the web3 space to cast strong judgement (positive or negative) on it. So I simply don't participate in web3. (But I keep on trying to learn about it).
Personally, I think I can avoid "being played" without calling folks whose work I don't understand scumbags.
Sam Altman has done other stuff too. Like this comment asks, the same thing for you https://news.ycombinator.com/item?id=32087778. Why so focused on defending someone who failed upward (see what happened to his startup Loopt) and is now super powerful?
That's not my intention! For all I know, he might actually be a scumbag (to borrow the term from our friend in this thread). I'm genuinely trying to understand the web3 space (you can look at my comment history), which is why I started this thread by asking about whether our friend was aware of the zero-knowledge proof thing.
The quibble over the "scumbag" name-calling is because I was surprised that our friend would feel comfortable calling someone a scumbag, when they, self-admittedly, don't understand how WorldCoin even works.
You kept focusing on the scumbag thing though. Seemingly more than web3. At least three comments specified having issues with the scumbag specific issue as if it’s some insane social faux paux and even if it was, this isn’t some fancy dinner. All of this is why it’s so curious
No one said that’s the only thing that makes him a scumbag. Do we now need to dutifully lay out all our arguments for why someone might be a scumbag before saying so? That seems like a really high bar to push for. It would take far too long to do that each time especially as proponents for the person can just demand more arguments and proof.
As with all of these threads, someone needs to tell me why I shouldn't be optimistic about this. I'm not a physicist, and only partially grok all the technical challenges with making something like this work. I'm skeptical, but that's because so many others with so much money have been working on Fusion, so why is Helion different?
If you want to feel optimistic, solar pv tech that's been improving over decades should be hitting the same price point around the time they intend to actually produce net energy for the first time. Other pv tech in the pipeline that may or may not work aims to make it cheaper still.
So if this works great, if not it's not the only shot we have.
Also the most important price point is "less than fossil fuels" which renewables have reached already in most of the world. That's when you start displacing large amounts of carbon.
But more new cheap, clean energy sources are always welcome news.
I doubt renewables are cheaper once they start making up a majority of the supply (except in small countries that can get most electricity from hydropower). Sure batteries or other storage technology may get there eventually but that's very much TBD.
80 or 85% is the figure that people usually quote as being when solar and wind will start having issues (the exact division between the two varies by latitude mostly), but at that point you're saving money on 85% of your energy expenditure so things have to go pretty drastically wrong for you to end up in the negative.
As this post notes, with cheap power you can make liquid fuels for less than it costs to dig them up.
This web site here will give you example renewable only grids for most of the earth and how much battery and power2x renewable gas you'd need. It's generally not a lot, wind and solar do the heavy lifting.
You get lots of batteries in sunny places, but that's just because it's cheap and we know it works well with solar. But yes anywhere with hydro gets to take advantage of cheap renewables much more easily.
Interesting. I'll have to look into it more closely. Sounds like the "climate crisis" (don't like that term but everyone knows what it means so it's convenient) is basically solved if everything you've said is true.
Yeah, it's always been a political problem, not an engineering one.
If you're effectively getting paid to release carbon into the atmosphere, then you're going to release a lot of carbon into the atmosphere. You're also going to get really pissed if someone wants to start charging you for it rather than paying you.
Luckily there were enough people who pushed for renewable subsidies to get them over the hump where even supervillains would use them for their own selfish profit.
Still to be seen whether we left it too late, how long disinformation and lies will hold us back still and so on, and whether any petro-states will start a big war to distract from the new reality but I'm generally positive (or as positive as you can be when you realise some of the largest corporations in the world and their pet politicians were literally gambling the future of the human race just so they'd get a fat bonus while making everyone else sicker and poorer, again).
I have a friend who did their dissertation particle physics at a fusion lab, modeling Alcator c-mod tokamaks, and was reached out to by Helion for a R&D position.
The general consensus among him and his peers is that the hardest challenge to overcome is the energy capture. As interesting as it is, it's not efficient enough to be useful yet.
The company is by no means snake-oil, but there's a reason the energy-capture part of the video is a bit vague.
I have always wanted to believe that fusion will be a real energy source someday. Maybe this is finally it. Not the first time someone said that though.
This made me wonder just how much he's actually putting in vs. what he has / is investing in general. Reports say that he personally invested $375MM into Helion[0] and various sources say his total net worth is...$200-$200MM[1][2].
But yc definitely also has solar startups, and Sam Altman appears to believe in those as well[3], as one would expect.
Conclusion: It's unclear how "all-in" he is (but it's a pretty big bet), and online "net worth" estimators are even more terrible than I'd assumed.
Techno-optimism is destructive when it gives people an excuse for inaction. If you're actually putting a lot of effort or resources into a possible solution, that's the opposite of inaction.
Plenty of people are tackling the political side of things, with a notable lack of success. I'm glad there are people working hard on the technological breakthroughs that could make the task a lot easier.
ur talking about vc who got us into this mess and continue to invest in consumerism and growth at all costs, not a bystander that stepped in to save the day
It's not "vc" who got us burning vast amounts of oil and coal. That predates modern vc, and isn't even exclusive to capitalism. The Soviets and communist China were enthusiastic participants.
If the uber-riche (like SA) are doing anything less than betting their comfortable livelihoods on climate tech, they're not doing enough. The rest of us don't have the resources to do anything other than vote and hope change literally comes from the top.
sorry for my imprecise words. what I meant is that they are exclusively interested in market solutions for climate change, using capitalism to cure itself
1) I picked up low voltage electrical engineering along the way, just never did a degree. As in: PCB design, tons of robotics and very small run (<100) bespoke devices.
2) I’m oldish (mid 30s) and have a family and am semi retired. So it would be burdensome to do a college degree, especially with the additional liberal arts requirements.
My point is: if I had the degree, I would “come out of retirement” and apply for jobs like this one, but I don’t, and it would take several years to get it even at an accelerated pace.
So I guess regret is meant literally. I wish I had just done the degree in my 20s, because I actually don’t want to do it now, and even if I do just decide to do it online or something, it will be more annoying at this point in my life than it would have been when I was 20.
I'd urge you to reexamine now. When your kids are grown up and out of the house and you're a little older you might think "it would have been so easy to go back and get my degree then and look where I could have been now." Remember that although rare, there are people getting PhDs in their 60s and 70s.
There was a post on here about a guy who got his degree in less than a year. He used an online school that allows for self paced courses.
Also, you might not really need a degree depending on your body of work. I imagine most startups these days will happily ignore that requirement if the candidate has demonstrable experience in the field.
On the flip side, a company like this probably has low pay, long hours, and it will be a decade or more before you see your work come to fruition (if at all). So if you are financially secure enough already, the work might not be for you. Probably best to pick up some hobby projects that scratch the itch.
Whaaat? What kind of oldish is that? You need a good friend kicking your rear.
You need to find something meaningful. The nihilistic trap will do no good to you.
I don't know about "nihilistic" but thinking you're "oldish" in your mid 30s could be interpreted that way. If you live to be 70 you're only halfway through, eh?
Anyway, you're young, already semi-retired, with "plenty of money, and live a very comfortable life" "spending time with [your] wife and children", and your regret is not being able to help create fusion directly (as opposed to "merely" investing)?
Not to be that guy, but uh, "must be nice".
Seriously though, congratulations! You seem to have won life. :)
I'm curious about your motivation for your original comment, do you not want encouragement to go back to school?
No I don't want encouragement to go back to school. My point was that helion is a really cool company, and anybody with the credentials to work there should should be happy about it.
That part is the good part! But more is needed. They will expect more from you. A role model that inspires meaning to them. That comes out of whatever you are doing and studying.
And congrats for achieving that position where you are, that's brilliant. A lot of people is unable to form healthy functioning families. But you in the inside will want more, you'll eventually be hungry for things that aren't transient.
You're in a great position to study classic philosophers and ask the big questions in life and see what comes to you from that.
If you know software, most engineering projects have software development needs, for example method development. You could be contributing to an engineering project without an engineering degree
Tell that to people without enough money to be semi retired before 40 that actually pursue their degrees like everyone else. Some of those, sometimes single parents, have to work full time during during yheir studies (a had one of those in my class for a while).
You are in such a priveleged position, if you want that engineering degree, go get it.
You can do all but the most cutting edge engineering on your own. People still build cars and planes in their garages. Lots of amateur radio hobbyists build their own radios. Even if you got an engineering degree you'd most likely spend 10 years working on boring projects before you got to do anything novel.
My post literally meant: I am an engineer who never got a degree. Oh man it would be so cool to work specifically for helion, but a degree is a requirement. I wish I had gotten it back in the day, because I would apply for one of the positions.
The job postings all say: "We believe in hiring individuals. If you do not check all of the boxes, we still encourage you to apply. If this job isn't a fit, we have many other openings that may be one. "
So maybe if you have relevant experience and can show it, apply anyways?
-> As the plasma expands, it pushes back on the magnetic field. By Faraday's law, the change in field induces current, which is directly recaptured as electricity. This clean fusion electricity is used to power homes and communities, efficiently and affordably.
Thermal power goes into the expansion of the plasma I'd figure, just like the hot gases in an internal combustion engine cylinder. Only here the work is done against a magnetic field instead of a piston.
This seems to be a magnetohydrodynamic generator. Shoot a charged plasma through a coil and get electricity out. Experimental MHD plants have been built for fossil fuels.[1]
MHD have pretty low efficiency (otherwise the world wouldn't have bothered with turbines, pistons/crankshafts, etc.). They claim 95% recovery. That is amazing. That would in particular seem to imply that the radiative/thermal losses less than 5%, which is very strange (classic thermal radiative power is 4th power of temperature and add bremsstrahlung to that). You'd usually need high density (or size sometimes can do for it) for the radiation/thermal to not escape much. They seem to have demonstrated efficient recovery at conditions far from the real fusion where such radiative losses would be expected to grow disproportionally.
According to Wikipedia, there's an inherent limit on MHD generators due to plasma instability. That's also the problem with most fusion systems, where everybody struggles to maintain a stable plasma for seconds. The idea behind this Helion thing seems to be to run in pulsed mode, so long term stability isn't required.
Pulsed fusion systems have been around for a while. There's another startup, First Light Fusion, also claiming to have a commercial energy source Real Soon Now.[1]
People have been claiming that for pulsed fusion since at least 1971, when laser confinement was a big thing.
I personally believe that the pulsed/inertial would be the fusion, be it energy generation or space engine (after all the only net positive artificial fusion device so far - thermonuclear bomb - is "pulsed", and so we have only a matter of engineering to scale it down :). My bet though would be on something like Sandia Z-machine or on laser (as we've had remarkable laser efficiency improvements) as they clearly create fusion conditions (and much closer to the bomb confinement wise) and burn significant amounts of fuel (thus allowing for typical established ways of energy generation) than for example Helion where the plasma clouds don't carry much fuel and thus can't produce significant amounts of energy.
At least we know it's not written by and for engineers or scientists, but let's be polite and just call it BS. There is so much things need to happen between plasma expanding and consumer electricity in the copper wire.
> Helion now needs to figure out how to engineer machines that don’t break
This is actually honest. It encompasses all the past, current, and future researches. The rest doesn't matter as much. Wish them the best.
Helion addresses that in their FAQ (https://www.helionenergy.com/faq/). TLDR, they have produced six working prototypes, the most recent which has been continuously running for 16 months. These haven’t produced net electricity yet, apparently due more to scale than to any unsolved scientific or engineering problems. Prototype 7, “Polaris”, is aimed to scale up the design to produce net electricity by 2024. This new hiring spree is to figure out how to mass produce it.
Wow, so start-up claims to have a fusion reactor demonstrator running for 16 months, while all the large projects are struggling with their firat prototypes. And we just believe said start-ups claims?
Achieving fusion is not that hard, lots of projects have working fusion. You can even do it yourself with a homebrew project costing about $1000. What's hard is getting more power out than you put in, which GP mentioned Helion has not done yet.
You don't have to believe it. Literally nothing in your life will change whether you choose to believe this bit of information or not. Either Helion will solve nuclear fusion, or turn out to be another failed startup, but in this instance, your personal beliefs will have no impact on which one happens. So you may as well suspend both belief and disbelief and just wait and see what the reality turns out to be.
Not true if you're in a position to work on the project. In that case, it matters how confident you are in their claims: if you believe them entirely then you might want to help, while if you disbelieve them then you won't bother.
The most basic physics (e.g. the actual reactions involved) is solved but how plasmas actually behave in practice is not well modeled. A lot of improvements in fusion efficiency come from tinkering with settings and seeing what works. A caveat that I am not an expert in fusion, and am getting this from the book "The Future of Fusion Energy".
Viable in the sense that fission power was viable before the Manhattan project - theoretically possible, but still an awful lot of experimental physics work left to do, not just engineering.
Fusion has been theoretically viable for decades, but we've never produced a sustainable fusion reaction that produces more energy than we put into it. We know it's possible because we can see the sun. But we haven't proved that any of our theoretical reactor designs can actually work.
Of the current fusion startups I'm generally most optimistic about Helion because they have a plausible solution for generating energy that produces electricity rather than just heat. Having to couple large amount of turbine with potentially neutron-activated liquids is a big driver of costs in nuclear plants and this neatly seems to sidestep the problem by turning the momentum of charged particles into electricity directly with no heat engine involved.
Aneutronic fusion (https://en.wikipedia.org/wiki/Aneutronic_fusion) - some fusion reactions simply don't result in many neutrons being created, rather all protons and electrons being created.
When a neutron is created, it has a lot of energy (heat) and it has 0 charge. This means it cannot be contained by electric/magnetic fields, and it cannot create an electric potential in metal directly. It must run into something else, transfer it's heat energy, and that something else must then transfer energy into electricity (usually this means heat up some water and run a turbine).
That's hard, lossy, and messy. And it's how all major fusion projects have always worked (including ITER). If instead, someone got an aneutronic fusion reaction to work (no theoretical issues, just engineering problems), they could produce pure protons and electrons, trap them with an electric field, and use the high speed motion of the charged particles to induce a current/voltage in a nearby metal.
This would solve the following problems:
1. Easy to contain high temperature outputs
2. Easy to protect the inner wall of a reactor (minimal interaction due to electric repulsion)
3. Easy to generate useful (electric) energy from heat energy
This would definitely be a gamechanger. It's like going from "use the sun to boil this water and turn the turbine" to "use the sun to induce a photoelectric effect." Categorical difference.
A significant cost driver of most energy sources is the work involved in dealing with the high temperatures and high pressures- cooling towers, steam chambers and turbines that can withstand superheated, high pressure steam, etc. It's a lot of concrete, thick steel walls and safety regulations as each one is basically a small bomb should the containment fail.
If they can actually produce a 50 megawatt generator the size of a shipping container and requiring inexpensive, easy to source fuel, with sane emissions, every power producer in the world will be chomping at the bit to buy them.
My state is currently in the process of putting up solar panel fields on the sites of coal power plants that are being decommissioned, because it is getting harder to find suitable locations with cooperative local governments. Many see them as eyesores, along with the added trouble of putting in new HVDC lines out to the fields.
If instead you could get significantly more power, and equally clean, from the same site, power companies will ditch solar in a heartbeat. No need to buy more land or run new lines, just decommission your coal and nat gas power plants and put up a few of these.
Their big problem is not manufacturing, not personnel, not extraction of useful energy from kinetic motion of charged particles, and not even physics of light-nucleus fusion. Those are, in principle, probably all solvable.
Their big problem is that they depend for fuel on 3He, and there isn't any. Not on Earth, and not even, usefully, in the lunar regolith. (By the way, there should be a better name for 3He as there is for 3H, tritium. "Helionium"?)
There is one source for 3He, and that is decay of 3H, tritium. Tritium can be made in high-neutron-flux processes such as in a CANDU reactor, slowing the neutrons with heavy water until they fuse with the 2H deuterium atoms in it. Then, keep it for 12+ years waiting for it to decay. We actually do manufacture a noticeable amount of tritium, and then store it in missile silos for the requisite 12 years, purifying out the 3He half periodically and topping up the tanks. Tritium is stored there ready to be funneled into the middle of a hydrogen bomb shortly before lofting to its unlucky recipients.
Operating these things is supposed to produce a little bit of 3H, which in principle could be collected, banked, and harvested. But each 3He they burn is very unlikely to make the 3H they will need to get another 3He some years on. They have to extract, anyway, the tritium they make, to avoid fusing those by accident, producing useless and harmful hot neutrons.
There is a competing technology, that has been worked on at Princeton and generically labeled "field reversed configuration", FRC, also burning D-3He, optimized for spacecraft propulsion. They hope to loft a test article by 2035.
So there are a few practical uses for this technology, if it can be made to work at all: to power spacecraft exploring the outer solar system, and maybe even for a moon base; but not for practical energy here on Earth.
For energy here, solar + wind + storage are the key to forestalling civilizational collapse. It might still be possible to prevent worldwide collapse if we radically increase the investment rate in these known-practical technologies. But not if we don't, and we haven't, enough. If civilization can be saved from collapsing, this tech could become interesting someday. Otherwise it will stall when civilization collapses out from under it. It is safe to say that would be a shame.
Android autocorrect thinks there is a word "trillium" already. So, maybe trelium or terlium. Or live with it, as we do lama and llama. Dreilium? Internet would probably go all in for threelium. Or heliumy-mc-heliumface.
Running deuterium-rich enough to net significant 3He out would seem to emit enough neutrons to render the reactor vessel strongly radioactive, and require the sort of cooling and shielding that makes the pure D-D reactors being pursued by others grossly impractical.
6% as neutron radiation is the figure I've seen, from Helion and other sources.
According to the head of MIT's fusion department, even D-T neutrons only mean you'll have to bury the activated materials for a few decades. A much lower amount of lower-energy neutrons isn't going to cause any major problems.
Exactly, the neutron induced radioactivity is orders of magnitude lower than in fission by-products and is a minor issue.
The problem is more with neutron embrittlement that weakens the materials exposed to fast neutron radiations. For DT fusion in tokamaks the idea is to replace the inner core every year or so.
Helion claim embrittlement is not much an issue for them. First, because their neutrons are relatively slower and second, because they can choose neutron resistant materials since they do not fear plasma contamination thanks to the pulsed nature of their operation: in between pulses they vacuum the vessel before injecting the exact mix needed.
In tokamaks, the materials facing the plasma are designed primarily to avoid emissions that might contaminate the plasma, embrittlement comes as an afterthought. Also in DT fusion, it's not 6-10% of the energy released as slow neutrons, but 100% as fast neutrons.
Conclusion: there is a neutron problem but it seem minor and fairly solvable
I am corrected. Operated just D-rich enough to breed its own 3He (plus tritium, presumably for sale), the material damage, waste heat, and radioactivation from neutron flux produced by the D-D side reactions are much more manageable than for a plant relying on capturing those neutrons for their heat.
The waste heat from a 2 MW plant, ~200 kW, would not be negligible, but capturing and handling 200 kW in warm (guessing?) lithium-deuteride saturated heavy water is different in kind from extracting 20 GW (probably the minimum size of a the D-T plant) from thousands of tons of molten metal.
Each dollar diverted from renewables to fission brings climate disaster closer. Besides displacing much less carbon emission, in operation, it doesn't start displacing any at all for a decade. During that decade, what you pay out for coal or NG could have bought the solar farm instead.
Coal producers love fission nowadays, because it means another decade of dependence on their product. They see the writing on the wall, so they only hope to put off the end. Promoting nukes is their lifeline.
Is that really how the financing of these projects works, though, to say that dollars are simply diverted from renewables to fission? I think not. As a civilization, we surely have the capacity and the imperative to do both. (That said, we should absolutely stop giving trillions in subsidies and financing to fossil fuel projects.)
Not to mention, we have a land use problem. Far too much land is given over to grossly inefficient livestock rearing, but solar installations are hardly better in land use terms, much as I recognize their necessity.
In any case, sites for solar installations are hard to find. Same deal for fission plants, for obviously different reasons, and in spite of their impeccable safety record by the numbers.
Difference being, though, that for every fission plant that overcomes NIMBYism, you get a lot more generation, albeit with considerable delay before that generation comes online, as you note.
All of which is to say that, the realities are not straightforward, it is not a binary choice. We should build as much new, non-carbon-emitting generation as possible.
Energy is the principal civilizational force multiplier. Given the damage we’ve already done that we need to undo, we need as much energy as we can get.
While we’re on the subject, let’s drop the facile pretense that any of these generation technologies are actually clean. They all have environmental externalities. I think it’s incumbent upon us to fairly represent that, and to make a whole-system analysis of this problem.
Hand-wringing about land use amounts to concern trolling.
There is absolutely no hint of shortage of land to site overabundant solar on. Solar coexists synergetically with numerous existing uses, including reservoirs and canals, pastures and farmland, and on industrial, warehouse, and even residential roofs and parking lots.
On water, solar farms operate more efficiently and cut evaporation and biofouling. In pasture and farmland, they operate more efficiently, cut water demand and heat stress, and for many crops even increase yield. They extend the life of roofs under and cars parked under them.
The area now devoted to domestic oil extraction is well over the most that would ever be needed for overbuilt solar, even without dual use.
Wind extraction, of course, has long coexisted nicely with agriculture and navigation.
So, the most effective way to get the overabundant energy we will need is to overbuild renewables.
The "environmental externalities" of renewable generation are orders of magnitude less than for others, so for purposes of planning really are negligible. Most of that is carbon emission occasioned by fabrication before the power used can be generated renewably.
> Their big problem is that they depend for fuel on 3He, and there isn't any. Not on Earth, and not even, usefully, in the lunar regolith.
It's interesting that talking with rocket people, they point out that getting enough - enough - He3 from the Moon regolith isn't that big of a problem, and then the phrase is practically repeated - "their big problem is that 3He isn't needed by anybody on Earth, since there are no fusion plants to utilize it". Classical catch-22.
Pushing against a magnetic field. There are a lot of ways to do it, including the generators that turn mechanical energy into electricity. My favorites are flux compressors though, for when you to turn the energy in a stick of dynamite into a very sharp current spike.
Their website mentions that fusion makes the plasma expand and:
> Electricity Recapture
> As the plasma expands, it pushes back on the magnetic field. By Faraday's law, the change in field induces current, which is directly recaptured as electricity. This clean fusion electricity is used to power homes and communities, efficiently and affordably.
Unlimited access to a natural resource can juice development short term, but look at where that has gotten us with both fossil fuels and Moore's Law - bloated, absurdly resource hungry software, and environmental destruction. Of course, we also got new technology and increased average standard of living, but so much waste.
Probably "free energy" is necessary to achieve a utopian, post scarcity society, but with all our prior experience with gluts of resource mostly being hoarded and used for profit-seeking, my enthusiasm for the nature of any fusion revolution is more measured than wide eyed.
10-15 days’ vacation is pretty bare minimum if you are trying to get top talent. Maybe think about adding 5 days onto those buckets to get higher quality candidates.
Is this a common number of vacation days in the US? In Germany the minimum _by law_ is 20 days on a 5 day week and I wouldn't even consider a position offering only that.
Assuming sick time is a separate pool (which is less common than it used to be and isn't mentioned), then yes in the US 15 days would be pretty standard for someone with experience but might be as low as 10 for someone without. It's almost certainly on the low side if that's a common vacation/sick time pool.
In Germany when you are sick you are sick, there is no limit. You will be paid and you can not be fired for it.
An allotted amount of "sick days" doesn't really make sense anyway, not like you can choose to be sick next year.
(Only exception is for long term sicknesses, if you are sick for more than 6 consecutive weeks, then the next week your health insurance will pay you 60% of your normal wage instead. Or if there is no hope of you ever being able to work, again insurance.)
Oh and if you are sick on vacation you get your vacation days back as you were not really able to take advantage of them.
Typically, in the US there's a progression from sick days (or just a shared pool of days) to short-term disability to long-term disability at companies that have those things.
6 weeks per diagnosis. If say you had six weeks of Covid and break your leg on the way out of the hospital and need another six weeks, your salary will be paid. And you keep your vacation days.
This seems like an insane law. I'm shocked it works at all. If a country wants to provide people insurance against sickness, totally fine. But to push it off on employers seems crazy. Let's say you are a small business, you have 1 employee. If they get sick you are now stuck paying them for 6 weeks and you can't replace them? So your small business can basically go under if someone gets sick.
I really disagree strongly with trying to push things onto companies.
The government want to pay basic income? Great! The government want to provide health care? Great!
This to me is the main advantage I like about basic income, is it really seems like we can get rid of all these suboptimal laws with horrible abuses and bad tradeoffs, including:
- min wage
- overtime
- vacation requirements
- healthcare requirements
So businesses can just focus on what they do best and the government provides a safety net.
Like another sibling said, it's not really that different in the US, since in the US you also use sick days for a while (not six weeks though) and progress to disability.
In Germany it's just more structured. There's multiple special rules, like the employee must be working at the company for a while, it has different rules for companies with fewer employees, companies can't be in an emergency situation, there might be the possibility tax rebates.
This is mostly handled by careful accounting. You need to provision future money for it rather than living in a reverse "paycheck to paycheck" situation. I never had a German company but I had a company in a country with similar arrangements and my accountant made me learn it all.
It also has the side effect of having businesses to care about employee quality of life. Someone working 12 hours a day will probably burn out quickly.
From a legal perspective I'm not sure, but I'd say 10 days is the minimum you'd see for a full time salaried position. But it's considered low for skilled professionals and would be a dealbreaker for many people. For skilled professionals (lawyers, engineers, etc.), 20 days is a more common number to see. This doesn't include sick days, which are usually counted separately.
It varies. Some companies have 'unlimited' paid time off, others just federal holidays. I don't think there is a federal law governing paid time off. There is a law (Family and Medical Leave Act) guaranteeing up to 12 weeks of unpaid time off though.
As someone else pointed out, some companies still offer vacation time and sick leave separately. None of the companies I've worked for still do, all have combined those into a single Paid Time Off pool.
Living where the legal minimum is 24 days in addition to bank holidays and where 28 days the practical minimum with the consus being 30 days, I'll call that exploitation.
How is it exploitation when both parties voluntarily engage in an agreement? Is anyone forced or coerced here? Is there any ambiguity or deceiving language?
It is better than calling it "Unlimited Vacation" and then leaving it up at discretion. They're being up-front and honest about it. It should be totally OK to advertise a job with 0 vacation days. We're talking about fully functional and grown up adults who have great college education who can make that decision if they would like to take on a high-responsibility and high-risk job.
In fact, I would argue the complete opposite. Doing anything else is interference and forceful coercion.
If you apply for 30 jobs, advance to 5 interviews, and receive 1 offer, all while your savings (do you have savings?) are being depleted on living expenses and college debt, are you really on even footing for fair negotiation? Or are you in a position to "take what you can get" and thereby find yourself vulnerable to an offer well below market norms?
We use the word exploitative to capture that sort of scenario. It suggests that the employer is exploiting candidates' extenuating circumstances. It's generally not sustainable for either party and the side effect for the employer is high churn and low productivity/quality.
This is a made up scenario that fabricates "exploitation", but even in this case I disagree. If there are no jobs out there, then tough luck. You are still getting paid vastly more than probably 99.99% of the world's jobs.
The entitlement is simply astonishing and incongruent, completely untenable even in the most charitable interpretation.
Furthermore, I believe that in jobs like this, companies cannot afford to hire someone that is going to take 8 months parental leave, for example. That would be consider exploitation of the employer by the employee in my view.
If all companies with interesting jobs only offer 10 days vacation, then there's no real other alternative. You could try to negotiate.
I prefer to give people plenty of holidays. Enjoy life, recharge. I often come up with some of my best work ideas while or after a holiday.
(I'm often concerned about the productivity of people who don't take time off. I work with independently contracting software engineers and some of the most productive ones take plenty of time off)
Top talent rarely takes vacations, and in my experience usually works a lot more than 40 hours a week (not because they have to, but because they want to).
Go and ask any PhD student at a top 10 CS university what they think of 20 days vacation and 30 hour work weeks. I think you'll find the answer expands your worldview.
That still sounds super low to me.
I've got 23 vacation days + 5-15 holiday days (varies by year due to the holidays shifting around the schedule) per year, and we get a ton of medical leave if you need it. I rarely make use of my full vacation options, but I still take a few weeks every summer and a few weeks during our holiday seasons. The US labor market makes no sense to me...
They may want just the passionate workaholic candidates.
I've had interviews where they were pretty open about the company culture not being one where you take vacation.
There is something to be said for working people a ton of hours to compensate for talent/skill. A few notable Canadian startups are built heavily on evening and weekend labour of immigrants and young people.
They want young, naive people that acceot that kind of thing in exchange for saving the world and some petty RSUs, if any. All the while working to make early investors and founders rich or even richer.
In a way, being open about it actually makes sure you get exactky those candidates. As I said, exploitative.
Usually those claims are bullshit, but not in this case. If I had the qualifications to meaningfully contribute to this, I'd jump at the chance and minimize my vacations for the next several years.
Feasibility of this fusion work aside, I think it must be quite amazing to really believe in all these things to come true, it's like living in the future.
I wonder whether Sam really believes in all these ventures, or if he's just building hype? I mean according to his timeline the whole world will run on Helion fusion energy in 15 years, A.I. will write all of our software code, we'll all pay in cryptocurrency using our biometric data and quantum computers will be a proven, practical technology (among other things). Would probably be amazing if these things came to pass (at least for some of us), but if I would go around and tell people stuff like that everyone would think I'm crazy. Then again I don't live in SV, so maybe that's on me.
(Disclaimer: speculation ahead, never having met Sam).
> I wonder whether Sam really believes in all these ventures, or if he's just building hype?
Part of being a VC is having many moonshots in your portfolio; if only some of them work out, that's still a huge win in total, even if the others go bust.
So he likely "believes" all these things in the same way that he believes in his portfolio: not that literally all of them turn out to be true, but some of them will, if you just believe in enough of them.
This is incorrect — Worldcoin orbs[1] use iris scans to create privacy-preserving proofs (these are called zero-knowledge proofs[2]) of uniqueness client side, which are just metadata for creating social graphs. The data is never used beyond the construction of the proof.[3]
If you're going to be critical you should at least understand why the idea is exciting to some. :)
Belief is a probability. Utility is an outcome propagated backward subject to your value function. Belief times utility minus opportunity cost can lead to low probability events having high utility such that their expected value even after accounting for a high opportunity cost is still worth pursuing.
If you take this approach with only one bet than you don't minimize your risk. Probabilistic reasoning doesn't have any guarantee for a particular sample. You can get unlucky. In order to safely pursue low probability high utility gains you have to make multiple uncorrelated bets such that your winners pay for your losers.
So the perception of this approach being crazy is exactly backwards. Rather than being crazy in his bold vision of the future, he is actually a cautious person who is trying to minimize the riskiness of his beliefs.
Apologies; I didn't mean to imply you thought he was crazy. I was trying to refer to the easy to reach perception of someone who makes a lot of low probability bets as being crazy, but was sloppy in my word choice.
A big part of moonshots like this is not being fatalistic. It is possible but not guaranteed. And if leadership does not have absolute conviction, belief in the best outcome future being possible, it will definitely not happen.
Speaking of "in 15 years" -- it's not too early to start doing y2038 blogging.
Write yourself a thoughtful post once per month about y2038 issues and you'll have a lucrative set of consulting jobs starting in 15 years and ending about one year later.
> Helion has demonstrated the concept[4] in a D–D reactor with plasmoids that fire once every three minutes, and it is now seeking $15 million in private financing over the next five years to develop a full- scale machine that could use D–T fuel to reach the break-even point, when it generates as much energy as it takes to run. The company hopes that its reactor could eventually reach the hotter conditions needed to fuse deuterium with helium-3, another combination that produces only α-particles and protons, with no neutron by-products.
>produces only α-particles and protons, with no neutron by-products.
Magnificent use of weasel wording there. D-He3 is aneutronic, but of course a mixed plasma hot enough for that reaction is going to have tons of fusions among deuterons, producing plenty of neutrons. I'm not sure if the sky high price of helium-3 is worth just reducing the neutron output of a reactor, rather than eliminating it.
It's intentionally a hybrid of D-D and D-He3. The output of D-D is half He3, and half tritium which decays to He3 with a 12-year half-life. So the only external input is actually deuterium.
I've seen Helion and other say that the combined reaction will release 6% of its energy as neutron radiation. That's low enough to skip a heat cycle, which is the real benefit since it substantially lowers your capital cost. That's why Helion estimates a cost starting at $0.01/kWh, and less as they get into mass production.
I know that I should be skeptical of all fusion claims, so I'm not asking anyone to predict if this actually WILL happen from Helion, but this machine claims to be smaller than other designs. Does anyone know how small this approach to fusion could scale down? Is it conceivable that it could deployed as the power source for an ocean-going ship, or could it be small enough that it could be part of a campus-scale microgrid?
That is conceivable, if it works as intended. Small enough to fit in a shipping container is small enough to put on a ship or on a campus. Of course, that's a big "if", and I don't have much to go on besides material that Helion itself puts out.
As a very much non-nuclear engineer, I found this video [1] explaining different fuels, size, magnetic field strength, and the orders of magnitude physics and engineering that goes into current fusion efforts.
Curious where Helion would put themselves on the nτE vs ion temperature plot he uses to evaluate how feasible a technology is.
Hartwig gave a similar talk for SULI a few weeks ago. Really awesome stuff and accessible too. Note that he is assuming steady state confinement. Fuel viability is a more complex beast for high-density low-confinement-time plasmas.
I used to work in CERN a long time ago and still keep in touch with a few physicists. All of them are very skeptical about commercial nuclear fusion before 2050 - 2070. It's going to take huge advances in science and engineering and a lot of money (on a scale similar to Manhattan or Apollo projects).
VCs funding a company has nothing to do with innovation
Their goal is to make a profit
That's why we don't have fusion energy yet
And that's why Nuclear plants have stagnated for many decades already
Education is where innovation comes from, it's been like this for centuries (scientific, health, industry) and yet it's been neglected for decades, go look at the salaries of teachers, even in private schools
VC backed startups prevent innovation, look at today, we living a recession since 2008 because the market invested and moved their wealth in for profit pockets
If you invest in schools, you don't expect profit in return, you expect innovation in return
If you invest in a startup, you don't expect innovation in return, you expect profit in return
It is very simple
Galileo wasn't a VC and he wasn't backed by a VC
Even Tesla and SpaceX couldn't be where they are with only just VCs, they had to get massive help from the fed and the government to stay afloat, as well as a shit ton of internships from all around the world to sustain their R&D efforts
The Manhattan project cost $23 billion in 2020 dollars. The ITER project alone is spending more than that. Helion is spending a couple billion, but draws on decades of published research.
Nothing like it. Nearly all the best physicists and mathematicians in the free world were employed in Manhattan, a minuscule amount of the top scientists in the world today work on fusion.
You said "a lot of money (on a scale similar to Manhattan", so I've been arguing that specific point. Money on the scale of Manhattan is in fact being spent.
I don't think there's any way to quantify whether the "best" physicists are working on fusion so I'm not going to argue that one.
As I tried to explain, what I meant was the Manhattan project cost 0.8% of 1945 US GDP. To spend the same percentage today would be 184 billion. Obviously nothing like that is being spent.
If you read physics journals, you'll know there is very little research being done and published on fusion today.
To be fair, Q > 1 in a research system and “commercial nuclear fusion” are two different things. I’m a lay person but it feels to me like SPARC has a good chance of Q > 1 this decade and reasonable that they could build a commercial system before 2040.
I’m reminded of all the NASA people that were highly skeptical of SpaceX when they were getting started and I feel like these big institutions end up with a bias against new companies with new ideas.
Sure, I'm not an expert either and future is hard to predict. It's possible that someone will solve the outstanding technical challenges in the next few years.
Regarding SpaceX, I believe when they started all their top rocket engineers where ex-NASA.
Please ask your physicists if they’ve heard of the Alcubierre drive. Not if they believe in it or if they have a debunk or it; but if they’ve even heard of it. I’ve talked to multiple ex cERNs and none of them have even heard of it but will be happy to say FTL is stupid.
(I’m perfectly aware of the outrageous requirements of the various Alcubierre configurations, but the point is before this it was considered mathematically impossible to do FTL. Now it’s just physical. The point is modern day academics are quite good at being very confident that everything except what they do is crap.)
The Sam Altman article and the Helion web site have some major red flags, but this[1] item from the FAQ in particular jumped out at me:
> Where is Helion going to get helium-3?
> Helion produces helium-3 by fusing deuterium in its plasma accelerator utilizing a patented high-efficiency closed-fuel cycle.
That sounds like it might be relevant for the overall efficiency of the system. Are there any outside fusion experts who can comment on the legitimacy of this company and their process?
From what I've seen, it's not a separate accelerator. The fusion reactor itself is a plasma accelerator and will run a hybrid reaction of D-D and D-He3. The combination will release 6% of its energy as neutron radiation.
For all the cynicism in this thread, I can't really deep down dislike sama for investing his money like this. There's plenty of ways to get more money and this isn't one of the most secure ones so money going to physics and smoothing out the engineering here may or may not get us all the way but it's nice that he's spending money to try to get us closer.
I agree. A simply b2b sass startup is hard enough to make work. This is probably 10x as hard to pull off. I still think it's great he and others are trying.
Probably closer to 100x or 1000x. I don't know how to build reactors, but I do know how to build SaaS, and as long as you solve a problem, implementation matters less than people think. You can really screw the pooch in so many ways and still won't kill people and you can be relatively "successful".
I'd image building reactors require you to really nail everything or you won't be anything close to "successful". The requirements and stakes seems to be unfathomably higher.
I think the way good investors look at it is based on the potential impact vs investment size.
Yes, investment in Helion is likely huge, 100x the cost of developing a SaaS app, but IF it works, the payoff is likely 1000x the return of a SaaS app.
This can't be ignored when looking at investments. Most investments will fail, but the larger investments won't necessarily fail at a higher rate than smaller investments. Market size to investment size must be considered along with probability of success.
Investors know they can't judge probability of success very well. But they try to judge opportunity return size.
This is why outliers like AirBnB are so difficult for investors. Sure, many investors didn't think it would be successful, but more importantly, they also didn't realize if AirBnB was successful, just how large it could be.
When you are looking at energy, it isn't difficult to figure out just how large the opportunity is, the risk is all in the execution of the technology.
> This is why outliers like AirBnB are so difficult for investors. Sure, many investors didn't think it would be successful, but more importantly, they also didn't realize if AirBnB was successful, just how large it could be.
What a great example, and great way to think about this phenomenon.
I completely agree with your post and have known about these dynamics of VC investing for 15 years, so am rarely surprised when it comes to this topic - but this way of putting it actually makes me think differently about things.
I dislike steam. Their approach to this is to extract electrical energy direct from the reactor is what I really like. If they actually succeed, this is perfect for space travel.
207 comments
[ 4.8 ms ] story [ 255 ms ] threadThat was quite an effective way to take your credibility and shove it down the drain.
They're all scumbags.
I don't want to defend anyone, but at the same time, if I see accusations, I'd like to see substance and reference points, not just "they're scumbags", end of conversation.
Please, provide some details on why you think they're like that.
https://news.ycombinator.com/item?id=30931614
It just puzzles me when people jump into defend what is clearly unsavory and amoral behavior.
Just because its legally grey area doesn't mean its moral.
Agree with not just this comment but your comments as a new throwaway (I assume?) account in general. Though I don’t care about the CCP enough to be as anti CCP as you. Tho Ofc I am anti CCP
No, I don't need to provide details. If you don't see it, you're a lost cause.
They're collecting _some_ value from individuals, whether you understand it or not.
Your comment reminds me of Facebook's disingenuous "we we're not _selling_ the data", which just a way of distracting from the real issue. This isn't about data, it's about yet another incremental stripping down of individual's control over their intrinsics.
What I know is that they're extracting some value from people who (like me) likely don't understand the implications. This isn't complicated.
I, like you, believe it is very important to remind folks that they should clearly understand the value prop of an exchange before they participate in it.
But I don't think that calling "scumbags" the folks behind any proposed value exchange (like Worldcoin) that you yourself (self-admittedly) don't fully understand the implications of, is a way to achieve that.
You asked me exactly what "instrinsics" are people being stripped from. So according to that article, people are being robbed of the ability of proving their own identity. If that startup succeeds, people will no longer be able to prove their identity other than thru them. This is really dangerous stuff.
You see, the mistake you made here is assuming that without understanding the details you can't come to any conclusions. But that is wrong. There's some things that I don't need to know the details of to accurately determine they're nefarious.
So again, they're all scumbags.
The article doesn't at all claim that the startup seeks to monopolize world wide human authentication!
For the benefit of "bystanders", the TLDR of the article is that (1) WorldCoin is collecting biometric data (not just a cryptographic hash of it), (2) Worldcoin switched compensation of its "operators" from a stablecoin to a brand new cryptocurrency, (3) perfecting the tech to authenticate being a human while preserving anonymity itself might be very valuable, and (4) the fact that brand new cryptocurrency is distributed to everyone might make the cryptocurrency itself very valuable.
I think 1 is valid concern if true; 2, IIUC, wasn't done "behind the backs" of anyone, so it's fine, 3 and 4 are very true. Thank you for linking the article!
> The mistake you made here is assuming that without understanding the details you can't come to any conclusions.
You are free to choose whatever standard of evidence you see fit, but personally, I will always choose to reserve strong judgement (like calling someone a scumbag) until I am confident that I fully understand the details.
You'll get played in life a lot.
Example: you were repeating zero-knowledge nonsense, whereas I saw right thru it. You get played a lot.
A "one way function" (ie, a hash) is not non-sense. It is entirely technically feasible (eg, Apple's NeuralHash)[0]. WorldCoin still maintains that participants can elect to submit only the hash of their iris scans.
> You'll get played in life a lot [if you reserve strong judgement (like calling someone a scumbag) until you're confident that you understand the details].
I mean, I can just refuse to play games I don't understand? For example, I don't know enough about the web3 space to cast strong judgement (positive or negative) on it. So I simply don't participate in web3. (But I keep on trying to learn about it).
Personally, I think I can avoid "being played" without calling folks whose work I don't understand scumbags.
[0] https://www.apple.com/child-safety/pdf/CSAM_Detection_Techni...
> Why defend [Sam Altman]?
That's not my intention! For all I know, he might actually be a scumbag (to borrow the term from our friend in this thread). I'm genuinely trying to understand the web3 space (you can look at my comment history), which is why I started this thread by asking about whether our friend was aware of the zero-knowledge proof thing.
The quibble over the "scumbag" name-calling is because I was surprised that our friend would feel comfortable calling someone a scumbag, when they, self-admittedly, don't understand how WorldCoin even works.
No one said that’s the only thing that makes him a scumbag. Do we now need to dutifully lay out all our arguments for why someone might be a scumbag before saying so? That seems like a really high bar to push for. It would take far too long to do that each time especially as proponents for the person can just demand more arguments and proof.
To be clear, I am not invested in anything Sam Altman is (financially, academically, or any otherwise). This maybe what you're trying to get at?
> Do we now need to dutifully lay out all our arguments for why someone might be a scumbag before calling someone a scumbag?
Yes, I 100% think so.
Edit: Good job with your arguments for the rest of the thread too.
https://www.helionenergy.com/faq/
Which kind of raises the question why they are not aiming to sell said Helium-3...
I call BS on basically all of it. Except their cookie banner, that has an "opt-out" button right next to accept.
So if this works great, if not it's not the only shot we have.
Also the most important price point is "less than fossil fuels" which renewables have reached already in most of the world. That's when you start displacing large amounts of carbon.
But more new cheap, clean energy sources are always welcome news.
80 or 85% is the figure that people usually quote as being when solar and wind will start having issues (the exact division between the two varies by latitude mostly), but at that point you're saving money on 85% of your energy expenditure so things have to go pretty drastically wrong for you to end up in the negative.
As this post notes, with cheap power you can make liquid fuels for less than it costs to dig them up.
This web site here will give you example renewable only grids for most of the earth and how much battery and power2x renewable gas you'd need. It's generally not a lot, wind and solar do the heavy lifting.
https://www.wartsila.com/energy/towards-100-renewable-energy...
You get lots of batteries in sunny places, but that's just because it's cheap and we know it works well with solar. But yes anywhere with hydro gets to take advantage of cheap renewables much more easily.
If you're effectively getting paid to release carbon into the atmosphere, then you're going to release a lot of carbon into the atmosphere. You're also going to get really pissed if someone wants to start charging you for it rather than paying you.
Luckily there were enough people who pushed for renewable subsidies to get them over the hump where even supervillains would use them for their own selfish profit.
Still to be seen whether we left it too late, how long disinformation and lies will hold us back still and so on, and whether any petro-states will start a big war to distract from the new reality but I'm generally positive (or as positive as you can be when you realise some of the largest corporations in the world and their pet politicians were literally gambling the future of the human race just so they'd get a fat bonus while making everyone else sicker and poorer, again).
The general consensus among him and his peers is that the hardest challenge to overcome is the energy capture. As interesting as it is, it's not efficient enough to be useful yet.
The company is by no means snake-oil, but there's a reason the energy-capture part of the video is a bit vague.
Take all that with a grain of salt.
But yc definitely also has solar startups, and Sam Altman appears to believe in those as well[3], as one would expect.
Conclusion: It's unclear how "all-in" he is (but it's a pretty big bet), and online "net worth" estimators are even more terrible than I'd assumed.
[0] https://www.cnbc.com/2021/11/05/sam-altman-puts-375-million-... [1] https://finty.com/us/net-worth/sam-altman/ [2] https://www.celebritynetworth.com/richest-businessmen/ceos/s... [3] https://techcrunch.com/2015/02/25/y-combinator-backed-bright...
Plenty of people are tackling the political side of things, with a notable lack of success. I'm glad there are people working hard on the technological breakthroughs that could make the task a lot easier.
I have made plenty of money, and live a very comfortable life, but I would really love to be able to work on projects like this.
I guess one option is to invest in companies like this.
1) I picked up low voltage electrical engineering along the way, just never did a degree. As in: PCB design, tons of robotics and very small run (<100) bespoke devices.
2) I’m oldish (mid 30s) and have a family and am semi retired. So it would be burdensome to do a college degree, especially with the additional liberal arts requirements.
My point is: if I had the degree, I would “come out of retirement” and apply for jobs like this one, but I don’t, and it would take several years to get it even at an accelerated pace.
So I guess regret is meant literally. I wish I had just done the degree in my 20s, because I actually don’t want to do it now, and even if I do just decide to do it online or something, it will be more annoying at this point in my life than it would have been when I was 20.
Also, you might not really need a degree depending on your body of work. I imagine most startups these days will happily ignore that requirement if the candidate has demonstrable experience in the field.
On the flip side, a company like this probably has low pay, long hours, and it will be a decade or more before you see your work come to fruition (if at all). So if you are financially secure enough already, the work might not be for you. Probably best to pick up some hobby projects that scratch the itch.
Anyway, you're young, already semi-retired, with "plenty of money, and live a very comfortable life" "spending time with [your] wife and children", and your regret is not being able to help create fusion directly (as opposed to "merely" investing)?
Not to be that guy, but uh, "must be nice".
Seriously though, congratulations! You seem to have won life. :)
I'm curious about your motivation for your original comment, do you not want encouragement to go back to school?
No I don't want encouragement to go back to school. My point was that helion is a really cool company, and anybody with the credentials to work there should should be happy about it.
And congrats for achieving that position where you are, that's brilliant. A lot of people is unable to form healthy functioning families. But you in the inside will want more, you'll eventually be hungry for things that aren't transient.
You're in a great position to study classic philosophers and ask the big questions in life and see what comes to you from that.
You are in such a priveleged position, if you want that engineering degree, go get it.
My post literally meant: I am an engineer who never got a degree. Oh man it would be so cool to work specifically for helion, but a degree is a requirement. I wish I had gotten it back in the day, because I would apply for one of the positions.
So maybe if you have relevant experience and can show it, apply anyways?
Have it next week boss
Charging without wires! /s
Where does the thermal power go?
I understand the theory behind this, but when it is used in an analogy like this it really blows my mind.
Here's a paper that explains it better than the wikipedia
[1] https://en.wikipedia.org/wiki/Magnetohydrodynamic_generator
Pulsed fusion systems have been around for a while. There's another startup, First Light Fusion, also claiming to have a commercial energy source Real Soon Now.[1] People have been claiming that for pulsed fusion since at least 1971, when laser confinement was a big thing.
[1] https://firstlightfusion.com/
> Helion now needs to figure out how to engineer machines that don’t break
This is actually honest. It encompasses all the past, current, and future researches. The rest doesn't matter as much. Wish them the best.
1. Theory is in place. Need to work out some some physics.
2. Physics is solid. Engineering needs work.
https://astralcodexten.substack.com/p/your-book-review-the-f...
Fusion has been theoretically viable for decades, but we've never produced a sustainable fusion reaction that produces more energy than we put into it. We know it's possible because we can see the sun. But we haven't proved that any of our theoretical reactor designs can actually work.
How??
When a neutron is created, it has a lot of energy (heat) and it has 0 charge. This means it cannot be contained by electric/magnetic fields, and it cannot create an electric potential in metal directly. It must run into something else, transfer it's heat energy, and that something else must then transfer energy into electricity (usually this means heat up some water and run a turbine).
That's hard, lossy, and messy. And it's how all major fusion projects have always worked (including ITER). If instead, someone got an aneutronic fusion reaction to work (no theoretical issues, just engineering problems), they could produce pure protons and electrons, trap them with an electric field, and use the high speed motion of the charged particles to induce a current/voltage in a nearby metal.
This would solve the following problems: 1. Easy to contain high temperature outputs 2. Easy to protect the inner wall of a reactor (minimal interaction due to electric repulsion) 3. Easy to generate useful (electric) energy from heat energy
This would definitely be a gamechanger. It's like going from "use the sun to boil this water and turn the turbine" to "use the sun to induce a photoelectric effect." Categorical difference.
Then their plans are to have a factory to build 20 50MW-generators per day. Each generator is shipping-container size.
Link to the interview in local press here: https://news.ycombinator.com/item?id=32088113
Isn't that like a medium-sized nuclear reactor every day?
If they can actually produce a 50 megawatt generator the size of a shipping container and requiring inexpensive, easy to source fuel, with sane emissions, every power producer in the world will be chomping at the bit to buy them.
My state is currently in the process of putting up solar panel fields on the sites of coal power plants that are being decommissioned, because it is getting harder to find suitable locations with cooperative local governments. Many see them as eyesores, along with the added trouble of putting in new HVDC lines out to the fields.
If instead you could get significantly more power, and equally clean, from the same site, power companies will ditch solar in a heartbeat. No need to buy more land or run new lines, just decommission your coal and nat gas power plants and put up a few of these.
Their big problem is that they depend for fuel on 3He, and there isn't any. Not on Earth, and not even, usefully, in the lunar regolith. (By the way, there should be a better name for 3He as there is for 3H, tritium. "Helionium"?)
There is one source for 3He, and that is decay of 3H, tritium. Tritium can be made in high-neutron-flux processes such as in a CANDU reactor, slowing the neutrons with heavy water until they fuse with the 2H deuterium atoms in it. Then, keep it for 12+ years waiting for it to decay. We actually do manufacture a noticeable amount of tritium, and then store it in missile silos for the requisite 12 years, purifying out the 3He half periodically and topping up the tanks. Tritium is stored there ready to be funneled into the middle of a hydrogen bomb shortly before lofting to its unlucky recipients.
Operating these things is supposed to produce a little bit of 3H, which in principle could be collected, banked, and harvested. But each 3He they burn is very unlikely to make the 3H they will need to get another 3He some years on. They have to extract, anyway, the tritium they make, to avoid fusing those by accident, producing useless and harmful hot neutrons.
There is a competing technology, that has been worked on at Princeton and generically labeled "field reversed configuration", FRC, also burning D-3He, optimized for spacecraft propulsion. They hope to loft a test article by 2035.
So there are a few practical uses for this technology, if it can be made to work at all: to power spacecraft exploring the outer solar system, and maybe even for a moon base; but not for practical energy here on Earth.
For energy here, solar + wind + storage are the key to forestalling civilizational collapse. It might still be possible to prevent worldwide collapse if we radically increase the investment rate in these known-practical technologies. But not if we don't, and we haven't, enough. If civilization can be saved from collapsing, this tech could become interesting someday. Otherwise it will stall when civilization collapses out from under it. It is safe to say that would be a shame.
Android autocorrect thinks there is a word "trillium" already. So, maybe trelium or terlium. Or live with it, as we do lama and llama. Dreilium? Internet would probably go all in for threelium. Or heliumy-mc-heliumface.
D+D fusion reaction yields 50% of T+p (tritium and a proton) and 50% of He3 + neutron
So in total their fusion reaction does not consume He3. He3 is here like a catalyzer.
They still need some He3. The D-D reaction produces tritium, and the tritium decays into He3. Half life of T is 12.5 years.
Done, no issues with the helium 3.
Actually they will probably produce T in excess and could sell it.
These are not the fast neutrons of the DT fusion (14.06MeV)
They claim it won't be worse than waste neutrons of radiation therapy medical devices. To be seen...
But yes, agree this is not exactly aneutronic fusion. Around 10% of energy is lost as slow neutrons and this is definitely a problem to address.
According to the head of MIT's fusion department, even D-T neutrons only mean you'll have to bury the activated materials for a few decades. A much lower amount of lower-energy neutrons isn't going to cause any major problems.
The problem is more with neutron embrittlement that weakens the materials exposed to fast neutron radiations. For DT fusion in tokamaks the idea is to replace the inner core every year or so.
Helion claim embrittlement is not much an issue for them. First, because their neutrons are relatively slower and second, because they can choose neutron resistant materials since they do not fear plasma contamination thanks to the pulsed nature of their operation: in between pulses they vacuum the vessel before injecting the exact mix needed.
In tokamaks, the materials facing the plasma are designed primarily to avoid emissions that might contaminate the plasma, embrittlement comes as an afterthought. Also in DT fusion, it's not 6-10% of the energy released as slow neutrons, but 100% as fast neutrons.
Conclusion: there is a neutron problem but it seem minor and fairly solvable
The waste heat from a 2 MW plant, ~200 kW, would not be negligible, but capturing and handling 200 kW in warm (guessing?) lithium-deuteride saturated heavy water is different in kind from extracting 20 GW (probably the minimum size of a the D-T plant) from thousands of tons of molten metal.
You forgot the expensive but easiest solution to this, building more regular fission power plants.
Coal producers love fission nowadays, because it means another decade of dependence on their product. They see the writing on the wall, so they only hope to put off the end. Promoting nukes is their lifeline.
Not to mention, we have a land use problem. Far too much land is given over to grossly inefficient livestock rearing, but solar installations are hardly better in land use terms, much as I recognize their necessity.
In any case, sites for solar installations are hard to find. Same deal for fission plants, for obviously different reasons, and in spite of their impeccable safety record by the numbers.
Difference being, though, that for every fission plant that overcomes NIMBYism, you get a lot more generation, albeit with considerable delay before that generation comes online, as you note.
All of which is to say that, the realities are not straightforward, it is not a binary choice. We should build as much new, non-carbon-emitting generation as possible.
Energy is the principal civilizational force multiplier. Given the damage we’ve already done that we need to undo, we need as much energy as we can get.
While we’re on the subject, let’s drop the facile pretense that any of these generation technologies are actually clean. They all have environmental externalities. I think it’s incumbent upon us to fairly represent that, and to make a whole-system analysis of this problem.
There is absolutely no hint of shortage of land to site overabundant solar on. Solar coexists synergetically with numerous existing uses, including reservoirs and canals, pastures and farmland, and on industrial, warehouse, and even residential roofs and parking lots.
On water, solar farms operate more efficiently and cut evaporation and biofouling. In pasture and farmland, they operate more efficiently, cut water demand and heat stress, and for many crops even increase yield. They extend the life of roofs under and cars parked under them.
The area now devoted to domestic oil extraction is well over the most that would ever be needed for overbuilt solar, even without dual use.
Wind extraction, of course, has long coexisted nicely with agriculture and navigation.
So, the most effective way to get the overabundant energy we will need is to overbuild renewables.
The "environmental externalities" of renewable generation are orders of magnitude less than for others, so for purposes of planning really are negligible. Most of that is carbon emission occasioned by fabrication before the power used can be generated renewably.
Agrivoltaics looks like a very clear win, especially for shade-loving crops [2].
[1]: https://meta.m.wikimedia.org/wiki/Cunningham%27s_Law [2]: https://youtu.be/lgZBlD-TCFE
It's interesting that talking with rocket people, they point out that getting enough - enough - He3 from the Moon regolith isn't that big of a problem, and then the phrase is practically repeated - "their big problem is that 3He isn't needed by anybody on Earth, since there are no fusion plants to utilize it". Classical catch-22.
https://en.wikipedia.org/wiki/Explosively_pumped_flux_compre...
> Electricity Recapture
> As the plasma expands, it pushes back on the magnetic field. By Faraday's law, the change in field induces current, which is directly recaptured as electricity. This clean fusion electricity is used to power homes and communities, efficiently and affordably.
Probably "free energy" is necessary to achieve a utopian, post scarcity society, but with all our prior experience with gluts of resource mostly being hoarded and used for profit-seeking, my enthusiasm for the nature of any fusion revolution is more measured than wide eyed.
More or less everyone has a 5-day week. Most jobs offer 25-30 days/year + unlimited sick days
An allotted amount of "sick days" doesn't really make sense anyway, not like you can choose to be sick next year.
(Only exception is for long term sicknesses, if you are sick for more than 6 consecutive weeks, then the next week your health insurance will pay you 60% of your normal wage instead. Or if there is no hope of you ever being able to work, again insurance.)
Oh and if you are sick on vacation you get your vacation days back as you were not really able to take advantage of them.
I really disagree strongly with trying to push things onto companies.
The government want to pay basic income? Great! The government want to provide health care? Great!
This to me is the main advantage I like about basic income, is it really seems like we can get rid of all these suboptimal laws with horrible abuses and bad tradeoffs, including: - min wage - overtime - vacation requirements - healthcare requirements
So businesses can just focus on what they do best and the government provides a safety net.
In Germany it's just more structured. There's multiple special rules, like the employee must be working at the company for a while, it has different rules for companies with fewer employees, companies can't be in an emergency situation, there might be the possibility tax rebates.
This is mostly handled by careful accounting. You need to provision future money for it rather than living in a reverse "paycheck to paycheck" situation. I never had a German company but I had a company in a country with similar arrangements and my accountant made me learn it all.
It also has the side effect of having businesses to care about employee quality of life. Someone working 12 hours a day will probably burn out quickly.
As someone else pointed out, some companies still offer vacation time and sick leave separately. None of the companies I've worked for still do, all have combined those into a single Paid Time Off pool.
It is better than calling it "Unlimited Vacation" and then leaving it up at discretion. They're being up-front and honest about it. It should be totally OK to advertise a job with 0 vacation days. We're talking about fully functional and grown up adults who have great college education who can make that decision if they would like to take on a high-responsibility and high-risk job.
In fact, I would argue the complete opposite. Doing anything else is interference and forceful coercion.
We use the word exploitative to capture that sort of scenario. It suggests that the employer is exploiting candidates' extenuating circumstances. It's generally not sustainable for either party and the side effect for the employer is high churn and low productivity/quality.
The entitlement is simply astonishing and incongruent, completely untenable even in the most charitable interpretation.
Furthermore, I believe that in jobs like this, companies cannot afford to hire someone that is going to take 8 months parental leave, for example. That would be consider exploitation of the employer by the employee in my view.
I prefer to give people plenty of holidays. Enjoy life, recharge. I often come up with some of my best work ideas while or after a holiday.
(I'm often concerned about the productivity of people who don't take time off. I work with independently contracting software engineers and some of the most productive ones take plenty of time off)
But since that's not the case, there are alternatives.
Go and ask any PhD student at a top 10 CS university what they think of 20 days vacation and 30 hour work weeks. I think you'll find the answer expands your worldview.
I've had interviews where they were pretty open about the company culture not being one where you take vacation.
There is something to be said for working people a ton of hours to compensate for talent/skill. A few notable Canadian startups are built heavily on evening and weekend labour of immigrants and young people.
In a way, being open about it actually makes sure you get exactky those candidates. As I said, exploitative.
Yeah, just saving the world, no big deal.
Usually those claims are bullshit, but not in this case. If I had the qualifications to meaningfully contribute to this, I'd jump at the chance and minimize my vacations for the next several years.
I wonder whether Sam really believes in all these ventures, or if he's just building hype? I mean according to his timeline the whole world will run on Helion fusion energy in 15 years, A.I. will write all of our software code, we'll all pay in cryptocurrency using our biometric data and quantum computers will be a proven, practical technology (among other things). Would probably be amazing if these things came to pass (at least for some of us), but if I would go around and tell people stuff like that everyone would think I'm crazy. Then again I don't live in SV, so maybe that's on me.
> I wonder whether Sam really believes in all these ventures, or if he's just building hype?
Part of being a VC is having many moonshots in your portfolio; if only some of them work out, that's still a huge win in total, even if the others go bust.
So he likely "believes" all these things in the same way that he believes in his portfolio: not that literally all of them turn out to be true, but some of them will, if you just believe in enough of them.
This is incorrect — Worldcoin orbs[1] use iris scans to create privacy-preserving proofs (these are called zero-knowledge proofs[2]) of uniqueness client side, which are just metadata for creating social graphs. The data is never used beyond the construction of the proof.[3]
If you're going to be critical you should at least understand why the idea is exciting to some. :)
[1]: https://www.youtube.com/watch?v=MA2ttYtUbF8
[2]: https://en.wikipedia.org/wiki/Zero-knowledge_proof
[3]: https://worldcoin.org/privacy
If you take this approach with only one bet than you don't minimize your risk. Probabilistic reasoning doesn't have any guarantee for a particular sample. You can get unlucky. In order to safely pursue low probability high utility gains you have to make multiple uncorrelated bets such that your winners pay for your losers.
So the perception of this approach being crazy is exactly backwards. Rather than being crazy in his bold vision of the future, he is actually a cautious person who is trying to minimize the riskiness of his beliefs.
Write yourself a thoughtful post once per month about y2038 issues and you'll have a lucrative set of consulting jobs starting in 15 years and ending about one year later.
LOL.
Is Helion backed by the same VCs that back all those shitcoins?
I read the web site. It's all web glitz, not much content.
Are there peer-reviewed papers on the technology?
https://iopscience.iop.org/article/10.1088/0029-5515/51/5/05...
> Helion has demonstrated the concept[4] in a D–D reactor with plasmoids that fire once every three minutes, and it is now seeking $15 million in private financing over the next five years to develop a full- scale machine that could use D–T fuel to reach the break-even point, when it generates as much energy as it takes to run. The company hopes that its reactor could eventually reach the hotter conditions needed to fuse deuterium with helium-3, another combination that produces only α-particles and protons, with no neutron by-products.
https://www.researchgate.net/profile/Mitch-Waldrop-2/publica...
- [4] https://iopscience.iop.org/article/10.1088/0029-5515/51/5/05... (https://sci-hub.se/https://iopscience.iop.org/article/10.108...)
Magnificent use of weasel wording there. D-He3 is aneutronic, but of course a mixed plasma hot enough for that reaction is going to have tons of fusions among deuterons, producing plenty of neutrons. I'm not sure if the sky high price of helium-3 is worth just reducing the neutron output of a reactor, rather than eliminating it.
I've seen Helion and other say that the combined reaction will release 6% of its energy as neutron radiation. That's low enough to skip a heat cycle, which is the real benefit since it substantially lowers your capital cost. That's why Helion estimates a cost starting at $0.01/kWh, and less as they get into mass production.
Curious where Helion would put themselves on the nτE vs ion temperature plot he uses to evaluate how feasible a technology is.
[1] https://www.youtube.com/watch?v=L0KuAx1COEk [2] https://en.wikipedia.org/wiki/Fusion_energy_gain_factor
Scroll down for table of lectures.
https://suli.pppl.gov/2022/course/
Slides
https://suli.pppl.gov/2022/course/IntroductionToFusionEnergy...
Video
https://mediacentral.princeton.edu/id/1_u3eelkrl
Helion - https://news.ycombinator.com/item?id=29119180 - Nov 2021 (250 comments)
Helion Energy achieves fusion milestone - https://news.ycombinator.com/item?id=27915833 - July 2021 (80 comments)
Y Combinator And Mithril Invest In Helion, A Nuclear Fusion Startup - https://news.ycombinator.com/item?id=8178450 - Aug 2014 (192 comments)
You wish :)
I used to work in CERN a long time ago and still keep in touch with a few physicists. All of them are very skeptical about commercial nuclear fusion before 2050 - 2070. It's going to take huge advances in science and engineering and a lot of money (on a scale similar to Manhattan or Apollo projects).
Their goal is to make a profit
That's why we don't have fusion energy yet
And that's why Nuclear plants have stagnated for many decades already
Education is where innovation comes from, it's been like this for centuries (scientific, health, industry) and yet it's been neglected for decades, go look at the salaries of teachers, even in private schools
VC backed startups prevent innovation, look at today, we living a recession since 2008 because the market invested and moved their wealth in for profit pockets
If you invest in schools, you don't expect profit in return, you expect innovation in return
If you invest in a startup, you don't expect innovation in return, you expect profit in return
It is very simple
Galileo wasn't a VC and he wasn't backed by a VC
Even Tesla and SpaceX couldn't be where they are with only just VCs, they had to get massive help from the fed and the government to stay afloat, as well as a shit ton of internships from all around the world to sustain their R&D efforts
Also, the Manhattan Project took three years. Fusion has been actively researched for seventy years.
We can spend all day quibbling but it seems clear that fusion research is at least roughly "on a scale similar to Manhattan."
I don't think there's any way to quantify whether the "best" physicists are working on fusion so I'm not going to argue that one.
If you read physics journals, you'll know there is very little research being done and published on fusion today.
I’m reminded of all the NASA people that were highly skeptical of SpaceX when they were getting started and I feel like these big institutions end up with a bias against new companies with new ideas.
Regarding SpaceX, I believe when they started all their top rocket engineers where ex-NASA.
(I’m perfectly aware of the outrageous requirements of the various Alcubierre configurations, but the point is before this it was considered mathematically impossible to do FTL. Now it’s just physical. The point is modern day academics are quite good at being very confident that everything except what they do is crap.)
> Where is Helion going to get helium-3?
> Helion produces helium-3 by fusing deuterium in its plasma accelerator utilizing a patented high-efficiency closed-fuel cycle.
That sounds like it might be relevant for the overall efficiency of the system. Are there any outside fusion experts who can comment on the legitimacy of this company and their process?
[1] https://www.helionenergy.com/faq/
I'd image building reactors require you to really nail everything or you won't be anything close to "successful". The requirements and stakes seems to be unfathomably higher.
Yes, investment in Helion is likely huge, 100x the cost of developing a SaaS app, but IF it works, the payoff is likely 1000x the return of a SaaS app.
This can't be ignored when looking at investments. Most investments will fail, but the larger investments won't necessarily fail at a higher rate than smaller investments. Market size to investment size must be considered along with probability of success.
Investors know they can't judge probability of success very well. But they try to judge opportunity return size.
This is why outliers like AirBnB are so difficult for investors. Sure, many investors didn't think it would be successful, but more importantly, they also didn't realize if AirBnB was successful, just how large it could be.
When you are looking at energy, it isn't difficult to figure out just how large the opportunity is, the risk is all in the execution of the technology.
What a great example, and great way to think about this phenomenon.
I completely agree with your post and have known about these dynamics of VC investing for 15 years, so am rarely surprised when it comes to this topic - but this way of putting it actually makes me think differently about things.
Thank you!