4 ms·

The answer is always "30 years away"

pseudolus 7y ago
Looking through the history of technology the same could have been said at one time for the automobile, aviation, genetic engineering, etc. The timescale during which nuclear fusion has been pursued is relatively short and declaring it a dead end at this early stage is premature.
- pfdietz 7y ago
  It was also said about innumerable technologies that went on to fail. Focusing just on those success stories is survivorship bias.
- dredmorbius 7y ago
  Let's look at each of those problems. Automobiles required a suitable fuel and a lightweight powerplant capable of highly dynamic power output. The first powered carriages date to the early 19th century, on rails, though coal and steam were not responsive, dynamic, or efficient. Good for railroads, not for untracked vehicles (not that this wasn't tried). Discovery and refinement of petroleum (1835 - 1865), creation of the reciprocating engine (Otto, Deisel, 1860s - 1880s), suitable tyres (vulcanised rubber, 1844), and high-quality steelmaking (Bessemer process, 1856 - 1860s). With the basic parts together, the first automobiles appeared in the 1880s, but further expansion required both mass-production assembly-line practices (Ford, 1901), and expanded and reliable fuel supplies (major oilfield discoveries including Spindletop, 1901, Lakeview, 1910, and East Texas -- "Pappy" Joiner's "Daisy Bradford No. 3", 1930). Well into the 1930s, it wasn't entirely clear that petroleum or petrol would be the fuels of choice, with other petroleum distilates (NGLs, condensate, and deisel) and ethanol among contenders. Once the automobile was developed, a period of rapid innovation and patent filing followed ... and largely tapered off by the 1930s. There's been incremental improvement since. Even in areas such as safety, mortality per billion passenger miles has improved at a remarkably constant rate of halving about every two decades, since the 1920s (the rate was twice that in the decade 1910-1920). Heavier-than-air craft were also largely dependent on the same factors: high power-to-weight powerplants, a high energy-density fuel, and a basic knowledge of aeronautics. Development of aircraft trailed automobiles only slightly, with similar trends of patent filings. The DC-3, described as a perfect aircraft design in Robert J. Gordon's The Rise and Fall of American Growth saw first flight in 1935, only 29 years after the Wrights on first first flight. It remains in active commercial use. Gordon on the DC-3: The very aircraft, the “Flagship Detroit,” that operated the first American flight from Chicago to New York is owned by the Flagship Detroit Foundation and often flies around the country to visit air shows. DC-3s are used everywhere because they are cheap; one can be purchased for $100,000. Their popularity eight decades after the first flight demonstrates that the DC-3 is the best-designed aircraft in aviation history. Its ruggedness is legendary; a common saying among pilots is that “the only replacement for a DC-3 is another DC-3.” Aircraft designed in the 1950s, and built in the 1960s, not only still fly, but serve as the backbone of strategic military bombing capabilities of the United States, in the form of the B-52, specific airframes of which will continue in combat missions through the 2040s if not beyond, and hold status as the fastest operational manned aircraft, in the form of the SR-71. That is: there are inventions for which, once the prerequisite requirements are met, rapidly emerge and attain levels of maximum development. DNA's existence was suggested by Darwin, the basic mechanics of genetics by Mendel, its structure by Watson and Crick, based on the work of others including Rosalind Franklin, and relying largely on the insights of X-ray crystallography. Direct manipulation and reading of DNA was achieved in the 1970s through enzymatic processes, though not well-developed for several decades further -- the problem is a fundamentally difficult one. Even now, the benefits of direct genetic manipulation are scant, another point covered by Gorden, in his general assessment of medical advances, or largely lack thereof since the 1970s. Still, at least there's been demonstrable progress. And then there's practical nuclear fusion.
gadders 7y ago
Nuclear fusion and aeroplanes that will get you from London to Sydney in two hours.
swebs 7y ago
Nah, we have a clear timeline today. Right now, we have fusion reactors that consume more power than they produce. Now we're developing ITER, which will produce more power than it consumes, but will only run for 20 seconds at a time. Once that's complete, we will move on to DEMO, which will continuously produce as much power as a small power plant. After that, projected to be after 2050, there will be PROTO, which should be the reference model for worldwide mass produced fusion power plants. https://en.wikipedia.org/wiki/ITER https://en.wikipedia.org/wiki/ITER https://en.wikipedia.org/wiki/DEMOnstration_Power_Station https://en.wikipedia.org/wiki/DEMOnstration_Power_Station https://en.wikipedia.org/wiki/PROTO_(fusion_reactor) https://en.wikipedia.org/wiki/PROTO_(fusion_reactor)
- beerandt 7y ago
  >projected to be after 2050 So about 30 years away?
  - swebs 7y ago
    Yes. The difference it that it's "30 years away" and not "probably 30 years away"
    - Udik 7y ago
      I'll mark it on the calendar.
    - robotresearcher 7y ago
      Assuming stable project funding, which is probable at best.
    - mrkstu 7y ago
      "This scenario assumes that ITER will deliver on the promise of tokamak-based fusion. That isn’t guaranteed. Magnetic confinement fusion remains a scientific research problem rather than an engineering problem, and the engineering issues have barely been touched." Quoted from the first link on your PROTO wikipedia page: https://community.dur.ac.uk/superconductivity.durham/The%20roadmap%20to%20magnetic%20confinement.pdf https://community.dur.ac.uk/superconductivity.durham/The%20r...

Looking through the history of technology the same could have been said at one time for the automobile, aviation, genetic engineering, etc. The timescale during which nuclear fusion has been pursued is relatively short and declaring it a dead end at this early stage is premature.

pfdietz 7y ago
It was also said about innumerable technologies that went on to fail. Focusing just on those success stories is survivorship bias.
dredmorbius 7y ago
Let's look at each of those problems. Automobiles required a suitable fuel and a lightweight powerplant capable of highly dynamic power output. The first powered carriages date to the early 19th century, on rails, though coal and steam were not responsive, dynamic, or efficient. Good for railroads, not for untracked vehicles (not that this wasn't tried). Discovery and refinement of petroleum (1835 - 1865), creation of the reciprocating engine (Otto, Deisel, 1860s - 1880s), suitable tyres (vulcanised rubber, 1844), and high-quality steelmaking (Bessemer process, 1856 - 1860s). With the basic parts together, the first automobiles appeared in the 1880s, but further expansion required both mass-production assembly-line practices (Ford, 1901), and expanded and reliable fuel supplies (major oilfield discoveries including Spindletop, 1901, Lakeview, 1910, and East Texas -- "Pappy" Joiner's "Daisy Bradford No. 3", 1930). Well into the 1930s, it wasn't entirely clear that petroleum or petrol would be the fuels of choice, with other petroleum distilates (NGLs, condensate, and deisel) and ethanol among contenders. Once the automobile was developed, a period of rapid innovation and patent filing followed ... and largely tapered off by the 1930s. There's been incremental improvement since. Even in areas such as safety, mortality per billion passenger miles has improved at a remarkably constant rate of halving about every two decades, since the 1920s (the rate was twice that in the decade 1910-1920). Heavier-than-air craft were also largely dependent on the same factors: high power-to-weight powerplants, a high energy-density fuel, and a basic knowledge of aeronautics. Development of aircraft trailed automobiles only slightly, with similar trends of patent filings. The DC-3, described as a perfect aircraft design in Robert J. Gordon's The Rise and Fall of American Growth saw first flight in 1935, only 29 years after the Wrights on first first flight. It remains in active commercial use. Gordon on the DC-3: The very aircraft, the “Flagship Detroit,” that operated the first American flight from Chicago to New York is owned by the Flagship Detroit Foundation and often flies around the country to visit air shows. DC-3s are used everywhere because they are cheap; one can be purchased for $100,000. Their popularity eight decades after the first flight demonstrates that the DC-3 is the best-designed aircraft in aviation history. Its ruggedness is legendary; a common saying among pilots is that “the only replacement for a DC-3 is another DC-3.” Aircraft designed in the 1950s, and built in the 1960s, not only still fly, but serve as the backbone of strategic military bombing capabilities of the United States, in the form of the B-52, specific airframes of which will continue in combat missions through the 2040s if not beyond, and hold status as the fastest operational manned aircraft, in the form of the SR-71. That is: there are inventions for which, once the prerequisite requirements are met, rapidly emerge and attain levels of maximum development. DNA's existence was suggested by Darwin, the basic mechanics of genetics by Mendel, its structure by Watson and Crick, based on the work of others including Rosalind Franklin, and relying largely on the insights of X-ray crystallography. Direct manipulation and reading of DNA was achieved in the 1970s through enzymatic processes, though not well-developed for several decades further -- the problem is a fundamentally difficult one. Even now, the benefits of direct genetic manipulation are scant, another point covered by Gorden, in his general assessment of medical advances, or largely lack thereof since the 1970s. Still, at least there's been demonstrable progress. And then there's practical nuclear fusion.

gadders 7y ago

Nuclear fusion and aeroplanes that will get you from London to Sydney in two hours.

swebs 7y ago

Nah, we have a clear timeline today. Right now, we have fusion reactors that consume more power than they produce. Now we're developing ITER, which will produce more power than it consumes, but will only run for 20 seconds at a time. Once that's complete, we will move on to DEMO, which will continuously produce as much power as a small power plant. After that, projected to be after 2050, there will be PROTO, which should be the reference model for worldwide mass produced fusion power plants. https://en.wikipedia.org/wiki/ITER https://en.wikipedia.org/wiki/ITER https://en.wikipedia.org/wiki/DEMOnstration_Power_Station https://en.wikipedia.org/wiki/DEMOnstration_Power_Station https://en.wikipedia.org/wiki/PROTO_(fusion_reactor) https://en.wikipedia.org/wiki/PROTO_(fusion_reactor)

beerandt 7y ago
>projected to be after 2050 So about 30 years away?
- swebs 7y ago
  Yes. The difference it that it's "30 years away" and not "probably 30 years away"
  - Udik 7y ago
    I'll mark it on the calendar.
  - robotresearcher 7y ago
    Assuming stable project funding, which is probable at best.
  - mrkstu 7y ago
    "This scenario assumes that ITER will deliver on the promise of tokamak-based fusion. That isn’t guaranteed. Magnetic confinement fusion remains a scientific research problem rather than an engineering problem, and the engineering issues have barely been touched." Quoted from the first link on your PROTO wikipedia page: https://community.dur.ac.uk/superconductivity.durham/The%20roadmap%20to%20magnetic%20confinement.pdf https://community.dur.ac.uk/superconductivity.durham/The%20r...