Timeline for Justifying why a nanoassembler can only produce inorganic things
Current License: CC BY-SA 4.0
9 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jan 18, 2023 at 17:34 | comment | added | toolforger | @LorenPechtel My point is that eating chopped-up components is not a good long-term diet; your digestion is built to to the chopping-up, and tuned towards those proteins that you find in nature. Give it a different mix, and it will be less efficient. (Look at the current wave of reports about vitamin D supplements that might actually be harmful. It's the same vitamin, but your digestion expects it to come with other foodstuff, and if it's absent, things don't work as smoothly.) | |
| Jan 17, 2023 at 16:45 | comment | added | Loren Pechtel | @toolforger The point is you can make food with the chopped-up components, there's no need to ever make the proteins in the first place. | |
| Jan 17, 2023 at 14:18 | comment | added | somebody | that said, replicating texture would likely be extremely hard - the same is true for e.g. metals though | |
| Jan 17, 2023 at 14:17 | comment | added | somebody | idk about this answer... nanoassemblers don't need to "fold proteins" - only to construct them from a given amino acid sequence, and i'd imagine a few decades from now we'd have the technology to construct arbitrary DNA very easily, quickly and compactly (basically, advanced CRISPR) | |
| Jan 17, 2023 at 9:59 | comment | added | toolforger | @LorenPechtel while yes, the body is going to chop the proteins up eventually, proteins not present in nature can have all sorts of side effects before the chopping up happens. E.g. they may poison or just hamper the intestinal microbiome, or have an awkward flavor. | |
| Jan 15, 2023 at 22:16 | comment | added | uhoh | I've piggybacked on to your answer. +1 | |
| Jan 15, 2023 at 13:17 | comment | added | TheDyingOfLight | I'm not sure that you need a super computer to run Alphafold2. Training such a Deep learning model is computationally expensive (though after Chinchilla we know that people overprioritised compute over data for optimal training results). Running something like a GPT-X model can be done on a home computer. And while Moore's Law may be dying, we are still orders of magnitudes away from the theoretical limits of computation in normal matter. | |
| Jan 15, 2023 at 2:11 | comment | added | Loren Pechtel | Nitpick: You can build nutritionally complete food without being concerned with protein folding. Despite standard nutritional advice the reality is that protein is not needed--your body is actually just going to chop it up. What's actually essential is those component amino acids and no folding is involved. Chemically, the essential fatty acids are more complex. | |
| Jan 14, 2023 at 18:33 | history | answered | KEY_ABRADE | CC BY-SA 4.0 |