[jerf]

The low rate of success achieved to date would be my first clue. I try to avoid the trap of letting today's glittering promises of future success distract from the fact that yesterday's glittering promises of future success' dates have actually come and gone (and come and gone and come and gone again).

Cards-on-the-table, my definition of success is full-on commercially-viable fusion, because little else justifies the enormous investments made into this technology. Any other "incidental" benefit you might want to cite, such as "better understanding of plasma physics", could probably have been obtained for orders of magnitude (plural and no exaggeration) less money.

(I also try to avoid the trap of seeing the word "science" and just shutting off my brain and throwing money at the problem because "SCIENCE!". Nonsense. Science is not immune to cost/benefit analysis just because it's shiny, and cost/benefit analysis is always probabilistic so yes, it can perfectly well deal with uncertainties of discovery too. See my previous comment about the value of keeping some cheap longshots in the portfolio. Just think, what other science could we have done with the billions of dollars it has consumed?)

[DennisP]

The fusion triple product has improved exponentially, and is 10,000 times better now than it was in 1970. It's a hard problem, but it's not so much "we keep trying and failing" as "we're consistently improving, but the bar is really high."

It's as if computers were useless until we had a 6th-generation Core i7, and people were saying "bah, computers, haven't succeeded so far." We're not there yet, but we're getting pretty close.

It is true that early researchers made some over-optimistic predictions. But they conditioned those on a certain level of funding. For the funding they got, they said it would never happen.

[jerf]

Again, I just can't help but think you've been blinded by SCIENCE! What I see is a curve that is slowing, not speeding, and no particular reason to believe that the curve ever gets over the "commercially-viable fusion" break-even point, ever.

We are not obligated to stop thinking about things because of how awesome the alternate reality where this all works like gangbusters would be. This is science, not politics.

Or, more accurately, ITER is politics, and not science, which is another reason I'm not holding my breath for this. It's not an experiment we're doing because the results are just so darned promising we had to carry on... it's an experiment we're doing because politicians have decided this is the way forward. We pour billions into this, and other approaches have a hard time getting single-digit millions. I do not care to follow along with the everybody-get-happy politics of ITER, I want fusion.

Might I further add that as this is science, should ITER succeed and produce a commercially-viable reactor, I will celebrate them all the harder for doing something I thought very unlikely, not try to argue it away. However, that is my bar for success, and I will not accept something sneaking under it under a cloak of pretty words.

[mahranch]

"What I see is a curve that is slowing, not speeding, and no particular reason to believe that the curve ever gets over the "commercially-viable fusion" break-even point, ever."

I think even educated people have a few misconceptions about fusion energy. We have already created fusion, we can do it in our test facilities on command. The problem researchers are currently facing is getting the efficiency to a commercially viable level. Getting out much more energy than they are putting in.

The research they are doing requires 2 things; time and money (and lots of it). They're trying to get the plasma models "just right" to achieve the desired efficiency. And they're not shooting for the stars, their goals are incremental. That's important because there will never be a time where researchers will announce out of the blue "We've done it!". It will be a slow build up due to how their research is taking place.

Harnessing fusion power isn't just theoretically possible, the models are completely mathematically sound. Unfortunately for the scientists, superheated plasma is incredibly unstable, hard to predict, and hard to control. To figure out the best way to stabilize it to maximize efficiency, they have to run countless tests, review the data, then try a countless more tests. It's not impossible, we WILL have fusion power eventually, that much is not in dispute, what is up for dispute is the "when".

If fusion researchers had unlimited funds, we could have a commercially viable fusion plant supplying power to a public grid in less than 10 years. Fusion is likely the only area of science where you can literally throw money at it and achieve your goal of a viable working product. The problem is that nobody wants to spend billions of dollars every year for the next 20 years. Only governments are footing the bill, and they're notoriously unreliable when it comes to scientific research due to the shifting political landscape

There was a great thread on reddit right after the fukushima disaster where several experts were talking about this very issue. I wish I could find it, it was a fantastic read.

[DennisP]

I agree that we should support alternate approaches. I actually got that 10,000x number from a TED talk by the CEO of General Fusion.

Other worthy projects: Sandia's MagLIF, MIT's levitated dipole and the Alcator C-Mod, UW's spheromak and FRC, polywell, focus fusion, fast laser, Helion, Tri-Alpha, and probably more that I'm forgetting.

If you want a cheap fusion experiment, the ultimate is focus fusion, which needs a million bucks to finish their breakeven attempt. They've spent $3 million so far, and right now have an indiegogo campaign to raise $200K for a new reactor part made of beryllium.

[bsder]

> Cards-on-the-table, my definition of success is full-on commercially-viable fusion, because little else justifies the enormous investments made into this technology. Any other "incidental" benefit you might want to cite, such as "better understanding of plasma physics", could probably have been obtained for orders of magnitude (plural and no exaggeration) less money.

Perhaps. Perhaps not. The Supeconducting Supercollider drove superconducting magnet technology which then made its way into all manner of other devices. Would anybody have put that amount of money into superconducting magnets without the SSC funding? Doubtful.

If you want to see fusion work, it's quite simple--convince China to announce that they are funding fusion at billion dollar levels. At that point the US will fund it at billion dollar levels and we'll have fusion before 2020.

[jerf]

We are funding it at billion-dollar levels! This is the result of funding it at billion-dollar levels!