what do you mean? the bag is useful for going around, but it's frail and needs replacing every few decades.

a $1000 drone is much more useful for going around

Something tells me the bag might change its mind about the whole operation once the upload is finished and it realises what's up next.

So yeah not sure I like the idea.

So far the only idea that I've read about and might feasibly result in uploading instead of copying is gradual replacement of each cell by a nanobot simulating that cell. So at the end of it, there'd be no bag to change its mind.

You can simplify it a bit. Yes, gradual replacement is likely the way to go, but you probably don't need to replace individual neurons one at a time. Individual neurons don't really matter or meaningfully contribute to our consciousness.

You can likely replace the large "functional groups" of neurons instead, with the group size threshold being the maximum that doesn't meaningfully affect our consciousness. This might well be many billions of neurons at a time.

Greg Egan's Sci-Fi jewel (dual) idea: https://philosophy.williams.edu/files/Egan-Learning-to-Be-Me...

Hence the "gradual replacement" part of my comment :)

Scan and copy never made sense to begin with, as you point out. Not sure why it's the default when people think of mind uploading.

People seem to imagine it like plugging a cable or getting an EEG cap and then they can't imagine the next step. Gradual replacement is a very radical idea to most people - a friend of mine recently described is as a horror movie.

In this book, I think: https://www.ebay.co.uk/itm/393899945799 - 108 Tips for Time Travellers by inventor and Professor Peter Cochrane, 1999, one of the essays is him asking his wife if she would still love him if he had false teeth, a false leg, etc. bit by bit until she stops the conversation saying "I'm not having you dying by installments!".

When you replace a heart with a pump, you don't get a human heart. When you replace a kidney with a dialysis machine, you doon't get a human kidney. Why expect that when you replace neurons with simulations you get a human brain or a human consciousness, or when you've replaced everything, a human?

Biological replacement, your body growing more new neurons, maybe, but it won't be mind uploading. And it won't get you you-at-age-twenty back.

To be honest, I don't really care if I get a human brain or not as long as I stay conscious.

I – and I'd wager most people – would be fine with some level of alteration to their being if it meant they wouldn't literally die, and that they would get to spend some more time with their loved ones. Change is a part of life, and this would just be another one.

It's arguable that you would literally die and be replaced by a machine-puppet pretending to be you; see the short story about that I linked in another comment in this thread exploring this topic: https://news.ycombinator.com/item?id=40407969

There is no technology I can see which would convince me that a mechanical replacement hand was my hand and my hand was not literally dead. Only Prof. Michael Levin's work of regrowing real limbs might do that - see https://www.popsci.com/body-electrician-whos-rewiring-bodies... and he has given many video talks, I saw one of him presenting at Facebook about his work triggering regrowing of frog's legs, but there appear to be many more videos of him - TED and Lex Fridman interviews.

That's pretty cool! At the same time, I think I would take the risk if I was on my way out anyways.

Let's assume that the replacements are perfect replicas from the outside, only the inside is different. Why would it not work?

The examples you listed are current technologies that don't remotely approach the primary nor secondary functions of the originals.

> "Why would it not work?"

As a thought experiment I'll go with "atom by atom replacement would still be you". But I won't easily go with "therefore we modern humans could actually do it in reality, and replace the parts with fakes, and then with simulated fakes, and then extract the simulation to run elsewhere for the uploading part, and even God couldn't tell the difference" just by asserting all those extra bits to be true. When have we ever made perfect replicas with the inside the same, of anything non-trivial at any scale?

Physical impossibility (it would work but we can't do it) argument, see here for long discussion of how hard the engineering problems of just Neuralink are: https://waitbutwhy.com/2017/04/neuralink.html

Physical "doesn't make sense" argument; we model brains as 80 billion neurons but bodies are not neatly separate organs and neatly separate pieces. The brain isn't separate from the body - nerves are basically brain matter spreading down to your spine, guts, arms, legs. Are 80 billion fake neurons surrounded by layers of scar-tissue reconnections going to be the same? What sense does it make to suggest a simulated sperm could mate with a real egg?

The physical engineering problems may be overcome with time. Billions of neurons - well, arguably you could replace more than one at a time, up to the largest "functional group" or threshold that doesn't comprise you.