An essay describing my best understanding of who I am, going from trivial details like name, date of birth and standard biographical details to all sorts of very subjective psychological self-assessments. The cover page would instruct the reader to go write a similar essay from memory before reading further, and then to compare notes with what’s written in the old essay.
Bonus points for figuring out a way to store a cryptographic checksum of the original essay in a future-proof way, so I could tell if the essay contents had been altered after I deposited it.
A variety of brain scans from various sources, different scanners, different techniques and technologies. Massive bonus points if I can get active scanners and run them constantly during “typical” and “optimal” days, recording all scans and saving them to whichever are the best available storage media (preferably multiple ones).
A large collection of 3d (or that could be reconstructed in 3d) videos of brain activity under various scanners would be best, since presumably it could be very useful for extrapolating the gritty details of the brain when reconstructing the patient, or at least for using it myself to learn about my own thought patterns.
I’d need to know how to relate the scan data to anything resembling high-level thought for them to be useful to me after being woken up.
Also, these scanners would need to be some serious future technology. Anything like today’s brain scans would be about as much help for reconstructing personality as low-resolution satellite photograph of a library building would be for transcribing the works of Shakespeare.
I’m not sure the analogy is appropriate. The sat-photos of the library give no information on the works of Shakespeare because the library’s rooftop (and the bookshelves, and the book cover, and the pages themselves) obscures the text, and there’s no traceable causal relation between the rooftop and what Shakespeare wrote.
However, even with current “off-the-shelf” brain scanners, you can with some training and machine-learning have your scanner recognize specific thoughts, so that whenever you think “Browser” in just the right mental pattern the headset will detect it (and fire up your web browser or something). So there is some correlation somewhere.
So a more appropriate analogy would be if you tore out all the pages of Shakespeare, and laid them out next to eachother on the rooftop, and then had a bunch of low-res satellite photographs, but also more importantly, of various types of video recordings from moving satellites all pointing at that text. With enough pictures and videos, emphasis on number and diversity of viewpoints / moving pictures / different imaging techniques, you’ll be able to eventually reconstruct quite a significant portion of the text if you had correspondingly amazing image-reconstruction technology.
(we can already do some pretty amazing things in that area from one single blurry picture to readable text, so imagine with the kind of future tech that reanimates dead people and massive visual datasets with varying angles and recording technologies...)
Still, it’s true that it might not provide much information. But it also might provide more than enough. It also might provide a helpful little bit more. It’s something that’s pretty hard to estimate, and I would stake my chances on more data rather than less if I’ve got the money available and am going to get frozen anyway.
That’d be help for the people doing the reanimation, not the reanimated you?
Being a bit weird, I might actually prefer a cryonic preservation where the people reanimating me get basically zero information beyond my physical remains, and will need to bring me back and ask me if they want to know my name. That way I’d know that whoever gets brought back will probably have their mind-state pretty closely causally connected to the one I had going in the suspension, assuming they will have a mind at all. Having lots and lots of lifelog information seems like it might increase the chances of the reanimators producing scrambled actors who are good at parroting my surface mannerisms to match the recordings, but aren’t internally much more of a continuation of me than a very capable actor doing Napoleon is the continuation of the actual Napoleon.
I’d need to re-write the essay every couple years, though, otherwise it might get out of synch with my memory even if they managed to exactly reconstruct my brain state from right before I was preserved.
Bonus points for figuring out a way to store a cryptographic checksum of the original essay in a future-proof way, so I could tell if the essay contents had been altered after I deposited it.
Bruce Schneier designed the ‘Solitaire’ cryptographic algorithm to use a deck of cards—if you asked politely, he might be able to refer you to some system which can provide the checksum you describe. The reasons and specifications of what you’re trying to accomplish might need to be made a bit more explicit.
I don’t see why I’d need Solitaire. There is probably going to be something much more seriously wrong than someone tampering with my diary if I end up in a future where I am unable to get my hands on any simple computer I can make run a contemporary crypto algorithm.
Basic scenario is, I don’t completely trust the safety deposit box, so I want to put something in there that can’t be easily changed and rebuilt (maybe the reanimators botch something bringing me back and then go mess with my personal effects trying to alter the evidence that could tip me off to something being wrong).
I could just use a standard digital signature with a huge private key which I would then destroy. But I’d need to store the public key somewhere outside the deposit box, or else the attackers could just re-sign the forgery and replace my public key with theirs in the box. I could commit the public key to memory, but might forget it with brain damage. I could also try to leave it in some public archives, since the key wouldn’t contain any information I might want to keep private, like the deposit box would.
An essay describing my best understanding of who I am, going from trivial details like name, date of birth and standard biographical details to all sorts of very subjective psychological self-assessments. The cover page would instruct the reader to go write a similar essay from memory before reading further, and then to compare notes with what’s written in the old essay.
Bonus points for figuring out a way to store a cryptographic checksum of the original essay in a future-proof way, so I could tell if the essay contents had been altered after I deposited it.
Could also be useful to add to that:
A variety of brain scans from various sources, different scanners, different techniques and technologies. Massive bonus points if I can get active scanners and run them constantly during “typical” and “optimal” days, recording all scans and saving them to whichever are the best available storage media (preferably multiple ones).
A large collection of 3d (or that could be reconstructed in 3d) videos of brain activity under various scanners would be best, since presumably it could be very useful for extrapolating the gritty details of the brain when reconstructing the patient, or at least for using it myself to learn about my own thought patterns.
I’d need to know how to relate the scan data to anything resembling high-level thought for them to be useful to me after being woken up.
Also, these scanners would need to be some serious future technology. Anything like today’s brain scans would be about as much help for reconstructing personality as low-resolution satellite photograph of a library building would be for transcribing the works of Shakespeare.
I’m not sure the analogy is appropriate. The sat-photos of the library give no information on the works of Shakespeare because the library’s rooftop (and the bookshelves, and the book cover, and the pages themselves) obscures the text, and there’s no traceable causal relation between the rooftop and what Shakespeare wrote.
However, even with current “off-the-shelf” brain scanners, you can with some training and machine-learning have your scanner recognize specific thoughts, so that whenever you think “Browser” in just the right mental pattern the headset will detect it (and fire up your web browser or something). So there is some correlation somewhere.
So a more appropriate analogy would be if you tore out all the pages of Shakespeare, and laid them out next to eachother on the rooftop, and then had a bunch of low-res satellite photographs, but also more importantly, of various types of video recordings from moving satellites all pointing at that text. With enough pictures and videos, emphasis on number and diversity of viewpoints / moving pictures / different imaging techniques, you’ll be able to eventually reconstruct quite a significant portion of the text if you had correspondingly amazing image-reconstruction technology.
(we can already do some pretty amazing things in that area from one single blurry picture to readable text, so imagine with the kind of future tech that reanimates dead people and massive visual datasets with varying angles and recording technologies...)
Still, it’s true that it might not provide much information. But it also might provide more than enough. It also might provide a helpful little bit more. It’s something that’s pretty hard to estimate, and I would stake my chances on more data rather than less if I’ve got the money available and am going to get frozen anyway.
That’d be help for the people doing the reanimation, not the reanimated you?
Being a bit weird, I might actually prefer a cryonic preservation where the people reanimating me get basically zero information beyond my physical remains, and will need to bring me back and ask me if they want to know my name. That way I’d know that whoever gets brought back will probably have their mind-state pretty closely causally connected to the one I had going in the suspension, assuming they will have a mind at all. Having lots and lots of lifelog information seems like it might increase the chances of the reanimators producing scrambled actors who are good at parroting my surface mannerisms to match the recordings, but aren’t internally much more of a continuation of me than a very capable actor doing Napoleon is the continuation of the actual Napoleon.
I’d need to re-write the essay every couple years, though, otherwise it might get out of synch with my memory even if they managed to exactly reconstruct my brain state from right before I was preserved.
Yes, would be good idea to make a habit of rewriting the essay every year. That way you could compare differences to a baseline yearly drift.
Bruce Schneier designed the ‘Solitaire’ cryptographic algorithm to use a deck of cards—if you asked politely, he might be able to refer you to some system which can provide the checksum you describe. The reasons and specifications of what you’re trying to accomplish might need to be made a bit more explicit.
I don’t see why I’d need Solitaire. There is probably going to be something much more seriously wrong than someone tampering with my diary if I end up in a future where I am unable to get my hands on any simple computer I can make run a contemporary crypto algorithm.
Basic scenario is, I don’t completely trust the safety deposit box, so I want to put something in there that can’t be easily changed and rebuilt (maybe the reanimators botch something bringing me back and then go mess with my personal effects trying to alter the evidence that could tip me off to something being wrong).
I could just use a standard digital signature with a huge private key which I would then destroy. But I’d need to store the public key somewhere outside the deposit box, or else the attackers could just re-sign the forgery and replace my public key with theirs in the box. I could commit the public key to memory, but might forget it with brain damage. I could also try to leave it in some public archives, since the key wouldn’t contain any information I might want to keep private, like the deposit box would.