5 ms·

The more times someone feels the need to tell you a particular technology is inevitable, the more sceptical you should be. Translation: "we need more money, an

The more times someone feels the need to tell you a particular technology is inevitable, the more sceptical you should be. Translation: "we need more money, and will continue to do so for the foreseeable future". People working on technologies which are truly "when not if" don't go around advertising it - they are too busy trying to be the one to profit off it before the next guy does.

aksss 7y ago
Not in this case - if we want to move off of hydrocarbons (let's assume we want to, let alone need to), nuclear is the only real candidate at this point. Renewables can't do it. What other options are there? Thus, nuclear is inevitable. Not saying this warrants giving these specific people money, but as a technology I think it's safe to say that it's inevitable.
- pfdietz 7y ago
  This meme needs to die. Moving off fossil fuels is easier and cheaper with renewables than it would be with nuclear, and the disparity is widening with time as renewables (and storage) crash down their experience curves.
mywittyname 7y ago
I agree. If a technology (especially one as significant as fusion) is viable and the production kinks need to be worked out, then VC-backers are going to dump money into it in an effort to be first-to-market. The less viable a technology currently is, the more governments will be paying for the research.
- jasonwatkinspdx 7y ago
  It's not just about viability, it's about complexity and timeline. There's very little appetite for projects with a 30+ year timeline among private investment. You can't expect VC backing for stuff like ITER.
  - tompccs 7y ago
    What technologies can you think of which didn't have profitable waypoints? - Internet - packet-switching routers for LANs - Microchips - ballistic missile guidance - Nuclear - radioactive tracers, radiotherapy, watch faces, ... - Microbiology - vaccines, insulin, ... Who on earth is making money out of controlled nuclear fusion? Where are the waypoints? Where can we make something useful in the short to medium time horizon? I also dispute your point of "little appetite for....30+ year timeline" projects, as the MEMS industry has routinely funded 30 year development cycles before technology maturation. The difference is, as I said, there are usually useful waypoints before the technology has fully matured.
  - perl4ever 7y ago
    That's an odd thing to say. Ever hear of a thing called 30 year bonds?
    - jasonwatkinspdx 7y ago
      Bond's ain't VC.
- raverbashing 7y ago
  Well VCs today are not known for taking risks in foreign areas (or knowing much about Physics or other natural sciences) Would something like Intel or MOS technology get vc money today?
  - rwallace 7y ago
    > Would something like Intel or MOS technology get vc money today? Yes, because when they were founded, it was already established that you could make money making semiconductors. The more concerning question is whether the invention of the transistor could be funded today. Historically, it was not funded by VC money, but out of the financial surplus of a big monopoly (AT&T). Are today's big, wealthy corporations doing as much of that sort of long-term research?
- chmod775 7y ago
  Assuming we are talking about generating electricity at scale, you can probably count the number of private entities with enough cash to finance a project like ITER on one hand, and it would be a massive loss to them if it failed. And even if private-ITER worked, you still only have a prototype, not an actual product you can sell. This is not something any VC-investor I know of would risk or can afford.
- trhway 7y ago
  > If a technology (especially one as significant as fusion) is viable and the production kinks need to be worked out, then VC-backers are going to dump money into it in an effort to be first-to-market. case in point - SpaceX. Before Musk private space on a big scale (not the laughable Pegasus) was may be just a couple notches below fusion. Once Musk shown the way for everybody, now we have VCs going all in. VCs are really very risk averse - they aren't going to dump money into being the first to make fusion because it is risky, they are going to dump money into the 20 startups of the second wave to make money for sure even if most of those startups fail.
AriaMinaei 7y ago
> they are too busy trying to be the one to profit off it before the next guy does That's not how innovation works. Moonshot projects take time. And they are risky and expensive. The risk/expense is too high for one or even a few investors to bear. That's why innovators can't get "too busy to profit off it." These efforts need buy-in from a large crowd, from charities, states, institutional investors, etc. Convincing such a large crowd needs a PR effort, which is why such articles get written. Are all of these efforts genuine? Of course not. But a project being long-term and "needing more money for the foreseeable future" is not a good heuristic to tell that it's not.
- tompccs 7y ago
  The only "moonshot" projects which ever worked were military ones. Ie, moon landing, the Manhattan project, etc. In these situations you have a state which can throw virtually unlimited resources at a project (even if it fails! see Star Wars and many other failed military "moonshot" projects) because it poses an existential threat for your opponent to develop it first. Of course, if the technology turns out to be impossible, then it's no big deal, because your rivals can't benefit from it either (in fact this was one of the main motivators for the Manhattan project - to see if it could be done, and make sure the Allies got there first if it could).
  - beerandt 7y ago
    While I don't necessarily disagree about SDI being a overall failure, we did get our current missle defence program and technology directly from the program. SDI also funded a vast amount of scientific research, although I don't know of any specific success stories to result. Regardless, I wouldn't classify it as a textbook example of a moonshot failure.
  - willis936 7y ago
    Many are of the mind that a crash program like the Manhattan Project is exactly what fusion needs in order to actually be “30 years away”.
    - 7y ago
Shivetya 7y ago
why? it is no different than any prediction ten or more years off. guess it depends on the subject matter, if you already believe in something they any number of years is acceptable
pasabagi 7y ago
There's this interview on omega-tau of a couple of plasma physicists(?) working on a fusion project in the UK, and there was this funny moment where one of them said that fusion was almost definitely going to work at some point, simply because the problem was so complex. The interviewer, a software engineer by trade, was extremely surprised, and asked him how complexity could be an advantage. The physicist replied that it means that there's a lot of variables you can fiddle before you start coming up against real physical limits. It sticks in the mind because you can sort of see why a scientist would think of fusion as a definite, next-few-decades thing, and at the same time, that doesn't mean it in the engineering sense of, these reactors will actually be just like a nuclear or coal plant.
- pfdietz 7y ago
  Indeed, the complexity is anathema to an engineering mindset.

aksss 7y ago

Not in this case - if we want to move off of hydrocarbons (let's assume we want to, let alone need to), nuclear is the only real candidate at this point. Renewables can't do it. What other options are there? Thus, nuclear is inevitable. Not saying this warrants giving these specific people money, but as a technology I think it's safe to say that it's inevitable.

pfdietz 7y ago
This meme needs to die. Moving off fossil fuels is easier and cheaper with renewables than it would be with nuclear, and the disparity is widening with time as renewables (and storage) crash down their experience curves.

mywittyname 7y ago

I agree. If a technology (especially one as significant as fusion) is viable and the production kinks need to be worked out, then VC-backers are going to dump money into it in an effort to be first-to-market. The less viable a technology currently is, the more governments will be paying for the research.

jasonwatkinspdx 7y ago
It's not just about viability, it's about complexity and timeline. There's very little appetite for projects with a 30+ year timeline among private investment. You can't expect VC backing for stuff like ITER.
- tompccs 7y ago
  What technologies can you think of which didn't have profitable waypoints? - Internet - packet-switching routers for LANs - Microchips - ballistic missile guidance - Nuclear - radioactive tracers, radiotherapy, watch faces, ... - Microbiology - vaccines, insulin, ... Who on earth is making money out of controlled nuclear fusion? Where are the waypoints? Where can we make something useful in the short to medium time horizon? I also dispute your point of "little appetite for....30+ year timeline" projects, as the MEMS industry has routinely funded 30 year development cycles before technology maturation. The difference is, as I said, there are usually useful waypoints before the technology has fully matured.
- perl4ever 7y ago
  That's an odd thing to say. Ever hear of a thing called 30 year bonds?
  - jasonwatkinspdx 7y ago
    Bond's ain't VC.
raverbashing 7y ago
Well VCs today are not known for taking risks in foreign areas (or knowing much about Physics or other natural sciences) Would something like Intel or MOS technology get vc money today?
- rwallace 7y ago
  > Would something like Intel or MOS technology get vc money today? Yes, because when they were founded, it was already established that you could make money making semiconductors. The more concerning question is whether the invention of the transistor could be funded today. Historically, it was not funded by VC money, but out of the financial surplus of a big monopoly (AT&T). Are today's big, wealthy corporations doing as much of that sort of long-term research?
chmod775 7y ago
Assuming we are talking about generating electricity at scale, you can probably count the number of private entities with enough cash to finance a project like ITER on one hand, and it would be a massive loss to them if it failed. And even if private-ITER worked, you still only have a prototype, not an actual product you can sell. This is not something any VC-investor I know of would risk or can afford.
trhway 7y ago
> If a technology (especially one as significant as fusion) is viable and the production kinks need to be worked out, then VC-backers are going to dump money into it in an effort to be first-to-market. case in point - SpaceX. Before Musk private space on a big scale (not the laughable Pegasus) was may be just a couple notches below fusion. Once Musk shown the way for everybody, now we have VCs going all in. VCs are really very risk averse - they aren't going to dump money into being the first to make fusion because it is risky, they are going to dump money into the 20 startups of the second wave to make money for sure even if most of those startups fail.

AriaMinaei 7y ago

> they are too busy trying to be the one to profit off it before the next guy does That's not how innovation works. Moonshot projects take time. And they are risky and expensive. The risk/expense is too high for one or even a few investors to bear. That's why innovators can't get "too busy to profit off it." These efforts need buy-in from a large crowd, from charities, states, institutional investors, etc. Convincing such a large crowd needs a PR effort, which is why such articles get written. Are all of these efforts genuine? Of course not. But a project being long-term and "needing more money for the foreseeable future" is not a good heuristic to tell that it's not.

tompccs 7y ago
The only "moonshot" projects which ever worked were military ones. Ie, moon landing, the Manhattan project, etc. In these situations you have a state which can throw virtually unlimited resources at a project (even if it fails! see Star Wars and many other failed military "moonshot" projects) because it poses an existential threat for your opponent to develop it first. Of course, if the technology turns out to be impossible, then it's no big deal, because your rivals can't benefit from it either (in fact this was one of the main motivators for the Manhattan project - to see if it could be done, and make sure the Allies got there first if it could).
- beerandt 7y ago
  While I don't necessarily disagree about SDI being a overall failure, we did get our current missle defence program and technology directly from the program. SDI also funded a vast amount of scientific research, although I don't know of any specific success stories to result. Regardless, I wouldn't classify it as a textbook example of a moonshot failure.
- willis936 7y ago
  Many are of the mind that a crash program like the Manhattan Project is exactly what fusion needs in order to actually be “30 years away”.
  - 7y ago

Shivetya 7y ago

why? it is no different than any prediction ten or more years off. guess it depends on the subject matter, if you already believe in something they any number of years is acceptable

pasabagi 7y ago

There's this interview on omega-tau of a couple of plasma physicists(?) working on a fusion project in the UK, and there was this funny moment where one of them said that fusion was almost definitely going to work at some point, simply because the problem was so complex. The interviewer, a software engineer by trade, was extremely surprised, and asked him how complexity could be an advantage. The physicist replied that it means that there's a lot of variables you can fiddle before you start coming up against real physical limits. It sticks in the mind because you can sort of see why a scientist would think of fusion as a definite, next-few-decades thing, and at the same time, that doesn't mean it in the engineering sense of, these reactors will actually be just like a nuclear or coal plant.

pfdietz 7y ago
Indeed, the complexity is anathema to an engineering mindset.