/sci/ - Science & Math

[Anonymous No.10769515]

Why do you think Drexler's nanomachines aren't possible?
If it's not possible, what are the biggest technical hurdles preventing it? How close to it do you think we could get?

[Anonymous No.10769519]

If it's not possible, then how would you respond to those who argue that life and cells are evidence that it should work?

If a molecular machine made of proteins can work, why not a non-biological one made of, say, carbon or silicon?

[Anonymous No.10769542]

>>10769515
Citation needed
https://www.youtube.com/watch?v=Cg--UVL9xCc

[Anonymous No.10769555]

>>10769542
Those medical "nanobots" that swim through the bloodstream are impressive but might instead be something on the micro scale, different from Drexler's vision of assemblers that can manipulate atoms and molecules

[Anonymous No.10769656]

>>10769519
If anything, cells show that it's not possible, because they already work at a speed that is very close to what is physically possible.

[Anonymous No.10769669]

>>10769519
This is exactly the argument against it. They're great examples of the limited potential of ultra advanced machines at those scales.

[Anonymous No.10769701]

>>10769656
>>10769669
Reading through this right now and he doesn't actually propose swarms of machines that would be able to move as fast or look as impressive as what's seen in movies (so perhaps the OP gif I used was a bit misleading)

He acknowledges that atoms and molecules are so small that it limits the speed and it might take days or weeks for assemblers to do or create anything visible.
Even with that limitation, however, it would still be a massively impactful tech breakthrough with endless potential uses

[Anonymous No.10769721]

>>10769701
At that point why not just genetically engineer existing living organism? It's not going to be easy but it sure beats building them from scratch.
Although I am talking about the stuff that self replicates. Purely artificial ones that accomplish simple tasks might be more effective, don't know.

[Anonymous No.10769829]

>>10769515
we don't have any way to make them today and it might not be practical to make them at all. The fat and sticky fingers problems still apply. Mechanosynthesis, the suggested process for making them, might not be practical.
>>10769542
that's not a robot at all. It's basically a molecular bear trap.
>>10769555
>>swim
worthless. They're externally controlled so there's not much you can do with them.
>>10769721
>>why not just genetically engineer existing living organism?
because it is exceptionally difficult to do so. Drexler makes the case that it's easier to engineer things made of hard matter rather wiggly polymers. It's harder to computationally model wiggly polymers because they move around so much. The other thing is that if we could build such things we can achieve higher performance than biology. For example Drexler shows that it's possible to make electric motors with ridiculous power densities. In addition, we could make very compact and powerful computers.

[Anonymous No.10769972]

>>10769721
That's sort of where the progress is at right now, building them out of cell structures like DNA strands.
Idea is that once you get good enough at building "soft" nanomachines then they could provide a path to "hard" nanomachines and mechanosynthesis >>10769829
>worthless because they're externally controlled
What? Wouldn't most nanomachines NEED to be externally controlled, if they're going to work together in large numbers, so that they can coordinate and accomplish macro-scale tasks?

[Anonymous No.10770025]

>>10769972
>>building them out of cell structures like DNA strands.
DNA nanotech is a dead end. DNA has a very low stiffness. It's has about the structural properties of jello.
Those swimming things are puppets. Magnetically controlled puppets. The kinds of control you can do is very limited. It is a developmental dead end.
>>nanomachines NEED to be externally controlled
There are tasks where external control is impractical. Hunting down bacteria is one of them. It is impractical to cover literally every part of the human body with microscopes so we can track where the bacteria and micropuppets are so that we may maneuver the micropuppets toward the bacteria to kill them.

[Anonymous No.10771495]

>>10770025
If those medical bots are actually on the micro scale then it should be possible to put sensors and computers in them so they can detect pathogens on their own, right? Why would they need to be externally controlled?
Difficult, but you wouldn't need mechanosynthesis or other still-completely-hypothetical tech to do it

[Anonymous No.10772240]

>>10771495
>>but you wouldn't need mechanosynthesis
then how else are you going to make sensors, computers, and actuators that can fit in a 1-4 micron in diameter robot?
www.rfreitas.com/Nano/Microbivores.htm
https://foresight.org/Nanomedicine/Respirocytes.php