5 ms·

Government-Funded Alchemy

24 points by surprisetalk 1y ago

PaulHoule 1y ago
I recently picked up a copy of the Heinlein classic https://en.wikipedia.org/wiki/The_Door_into_Summer https://en.wikipedia.org/wiki/The_Door_into_Summer where in "the future" (our past) there is a technology to transmute elements affordably and gold is cheap. The protagonist gets screwed by his business partners and sent into the future via cold sleep, buys a few kg of cheap gold, sends himself back into the past with an experimental time machine, and funds his own start-up company. The book is an interesting answer to people who wonder what eudaimonia is for an engineer, what one should do with one's talents, and the consequences of automation. In that universe, machine learning was realized with a particular type of vacuum tube in the 1950s.
- dvh 1y ago
  In Verne's Around the Moon the protagonist have to throw away dead dog out of the spaceship window, but he manages to do it quickly enough so that barely any air escaped. Writers are limited by the knowledge of their time.
godelski 1y ago
Tbh, I'm okay with this. Linked in the article it says Marathon Fusion's total investments are just under $7M. That's really nothing for government projects. We have to remember that what is a lot of money for us as people is very different than what is a lot of money for a government. Author seems to be concerned with how risky of an investment it is, but honestly, I'd be happy for the gov to make more of these types of investments. I mean it really is a very small percentage of the US national budget. Frankly, a big benefit from these types of government funded projects is that the knowledge tends to become public. Meaning that even failures lead to benefits as the next person can learn from previous work. Just like with any project. So idk, I'm happy to 10x that budget and see them fail if we learn something useful out of it. That's still a win in my book. $70M for a good step towards fusion? Deal! What I'm not okay with is the hype machines that form around these things. It is a self-inflicted wound explicitly created by our risk aversion. We try really hard to pick the best projects based on how well fleshed out their ideas are and supporting proof. But as you can easily guess from that condition, it usually means we end up moving in very small steps. Which then has the problem of being less motivating for spending money on as what, we're going to spend money for someone to research things we already know the answers to? So what happens? Either people make claims with confidence that is unfounded or they exaggerate the importance of their work. There's certainly a bias as people don't want to work on already known stuff and grant readers are willing to let some exaggerations slide as "it's just the way things are." But this compounds and gets out of hand...
- tw04 1y ago
  You’re glossing over the part where we’ve got an administration that’s anti science dumping money into what is literally snake oil. We are so strapped for cash we need to cut off people’s healthcare, slash NASA’s budget, make massive cuts to university research and terminate vaccine programs. But we’ve somehow got $7m to piss away on pretending you can turn lead into gold like it’s 600bc.
  - godelski 1y ago
    > what is literally snake oil. Well it is this part that I'm unconvinced. You can probably gather from my comment that I fully believe they are overselling, but that's different. As for the administration part, that's an orthogonal conversation. It is preying on the effects that I discussed, so they are related, but that doesn't mean they're the same thing. We can talk about the x-axis without the y-axis, even if we want to figure out where a point lies in a plane. > We are so strapped for cash I'll refer you to the first part of my comment. Personally, when you're strapped for cash I think it is best to go after the big ticket items that are costing you money rather than freaking out about losing a few pennies. You're not wrong that allocation is a problem, but allocation is a difference between lack of budget.
  - lesuorac 1y ago
    > It has been trivial since the 1980’s to produce gold alchemically, when Nobel laureate Glenn Seaborg (also of Seaborgium fame) used a particle accelerator to transmute bismuth into gold. Not sure it's literally snake oil. But also investing ~7M into a bunch of credentialed individuals to continue experiments with fusion energy seems to be a fine decision. In the past you'd need to explain how fusion energy will help diverse communities power their homes. Now you just have to explain how fusion energy will create the presidents favorite substance.
    - tw04 1y ago
      When it costs several hundred million dollars to produce gold that is measured in protons and neutrons it is functionally snake oil. The fact you can technically create gold from smashing particles together is completely irrelevant to the money being spent here. > the collision-induced reactions had removed anywhere from six to 15 neutrons, producing a range of gold isotopes from gold 190 (79 protons and 111 neutrons) to gold 199 (79 protons, 120 neutrons)
tbrownaw 1y ago
Mercury is kinda nasty, nice to see a possible method to properly destroy it like this.
- throwawaymaths 1y ago
  ironically one of the bigger uses for mercury is... gold mining.
throwawaymaths 1y ago
i mean the ostensible reason why gold is valueable is not "extrinsic", its that it's instantly verifiable using low technology (touchstone, density) whereas other metals might need at least chemistry to properly id with very high confidence. the ancient form of trustless currency, if you will.
twic 1y ago
So you synthesize gold in a bath of mercury. Gold and mercury form an amalgam. Isn't amalgam solid? The amalgam in my teeth certainly is. So will you get nuggets of it clogging up the mercury pumps? Apparently not, because the physical properties depend on the ratio of gold to mercury - when there's only a tiny bit of gold, it behaves just like mercury: https://www.youtube.com/watch?v=yAGYGGmUmUw https://www.youtube.com/watch?v=yAGYGGmUmUw I'm not sure how you would separate the gold from the mercury in this approach. The paper doesn't seem to say. I suspect the approaches used in mining assume a much higher concentration of gold, and would be impractical here.

24 points by surprisetalk 1y ago

I recently picked up a copy of the Heinlein classic https://en.wikipedia.org/wiki/The_Door_into_Summer https://en.wikipedia.org/wiki/The_Door_into_Summer where in "the future" (our past) there is a technology to transmute elements affordably and gold is cheap. The protagonist gets screwed by his business partners and sent into the future via cold sleep, buys a few kg of cheap gold, sends himself back into the past with an experimental time machine, and funds his own start-up company. The book is an interesting answer to people who wonder what eudaimonia is for an engineer, what one should do with one's talents, and the consequences of automation. In that universe, machine learning was realized with a particular type of vacuum tube in the 1950s.

dvh 1y ago
In Verne's Around the Moon the protagonist have to throw away dead dog out of the spaceship window, but he manages to do it quickly enough so that barely any air escaped. Writers are limited by the knowledge of their time.

godelski 1y ago

Tbh, I'm okay with this. Linked in the article it says Marathon Fusion's total investments are just under $7M. That's really nothing for government projects. We have to remember that what is a lot of money for us as people is very different than what is a lot of money for a government. Author seems to be concerned with how risky of an investment it is, but honestly, I'd be happy for the gov to make more of these types of investments. I mean it really is a very small percentage of the US national budget. Frankly, a big benefit from these types of government funded projects is that the knowledge tends to become public. Meaning that even failures lead to benefits as the next person can learn from previous work. Just like with any project. So idk, I'm happy to 10x that budget and see them fail if we learn something useful out of it. That's still a win in my book. $70M for a good step towards fusion? Deal! What I'm not okay with is the hype machines that form around these things. It is a self-inflicted wound explicitly created by our risk aversion. We try really hard to pick the best projects based on how well fleshed out their ideas are and supporting proof. But as you can easily guess from that condition, it usually means we end up moving in very small steps. Which then has the problem of being less motivating for spending money on as what, we're going to spend money for someone to research things we already know the answers to? So what happens? Either people make claims with confidence that is unfounded or they exaggerate the importance of their work. There's certainly a bias as people don't want to work on already known stuff and grant readers are willing to let some exaggerations slide as "it's just the way things are." But this compounds and gets out of hand...

tw04 1y ago
You’re glossing over the part where we’ve got an administration that’s anti science dumping money into what is literally snake oil. We are so strapped for cash we need to cut off people’s healthcare, slash NASA’s budget, make massive cuts to university research and terminate vaccine programs. But we’ve somehow got $7m to piss away on pretending you can turn lead into gold like it’s 600bc.
- godelski 1y ago
  > what is literally snake oil. Well it is this part that I'm unconvinced. You can probably gather from my comment that I fully believe they are overselling, but that's different. As for the administration part, that's an orthogonal conversation. It is preying on the effects that I discussed, so they are related, but that doesn't mean they're the same thing. We can talk about the x-axis without the y-axis, even if we want to figure out where a point lies in a plane. > We are so strapped for cash I'll refer you to the first part of my comment. Personally, when you're strapped for cash I think it is best to go after the big ticket items that are costing you money rather than freaking out about losing a few pennies. You're not wrong that allocation is a problem, but allocation is a difference between lack of budget.
- lesuorac 1y ago
  > It has been trivial since the 1980’s to produce gold alchemically, when Nobel laureate Glenn Seaborg (also of Seaborgium fame) used a particle accelerator to transmute bismuth into gold. Not sure it's literally snake oil. But also investing ~7M into a bunch of credentialed individuals to continue experiments with fusion energy seems to be a fine decision. In the past you'd need to explain how fusion energy will help diverse communities power their homes. Now you just have to explain how fusion energy will create the presidents favorite substance.
  - tw04 1y ago
    When it costs several hundred million dollars to produce gold that is measured in protons and neutrons it is functionally snake oil. The fact you can technically create gold from smashing particles together is completely irrelevant to the money being spent here. > the collision-induced reactions had removed anywhere from six to 15 neutrons, producing a range of gold isotopes from gold 190 (79 protons and 111 neutrons) to gold 199 (79 protons, 120 neutrons)

tbrownaw 1y ago

Mercury is kinda nasty, nice to see a possible method to properly destroy it like this.

throwawaymaths 1y ago
ironically one of the bigger uses for mercury is... gold mining.

throwawaymaths 1y ago

i mean the ostensible reason why gold is valueable is not "extrinsic", its that it's instantly verifiable using low technology (touchstone, density) whereas other metals might need at least chemistry to properly id with very high confidence. the ancient form of trustless currency, if you will.

twic 1y ago

So you synthesize gold in a bath of mercury. Gold and mercury form an amalgam. Isn't amalgam solid? The amalgam in my teeth certainly is. So will you get nuggets of it clogging up the mercury pumps? Apparently not, because the physical properties depend on the ratio of gold to mercury - when there's only a tiny bit of gold, it behaves just like mercury: https://www.youtube.com/watch?v=yAGYGGmUmUw https://www.youtube.com/watch?v=yAGYGGmUmUw I'm not sure how you would separate the gold from the mercury in this approach. The paper doesn't seem to say. I suspect the approaches used in mining assume a much higher concentration of gold, and would be impractical here.