You will die. Get used to it.
jacquesmattheij.com
jacquesmattheij.com
* Increasing the human lifespan past 130 using means anything resembling modern medicine seems pretty hard. Especially for keeping the brain going strong.
* People have a fixed identity point somewhere in their lifetime, and should they deviate too far from that later in their life, they shall be considered different persons and as good as dead.
* Going insane is going to be a big problem if the brain keeps going and changing state for centuries or millennia.
* People would go into decision-theoretic ruts if they knew they lived forever. Apparently people also want to go into permanent retirement after a while, regardless of how healthy they are.
* People living forever keep consuming resources. This is somehow worse than the current situation where people give birth to more resource-consuming people and then die.
* Unlimited lifespans would cause massive social stratification.
* Since we don't know exactly how minds work, they probably run on magic instead of atoms and electrons, so you can't get them in a computer.
* Although human minds are perfectly fine being trapped for decades in a dark liter-sized bone shell in nature, they would quickly go insane from claustrophobia in a similar computer substrate. I'm not entirely sure how this works, it might be related to the part about computers not running on magic again.
The article also includes some non-physicalist ideas that simulated brains can't feel emotion or feelings won't be "real" etc.
In essence it raises no new arguments or points of interest.
The word 'bored' is not in the article so I don't know where you got that.
That said (in support of the grandparent comment), I did think it was an interesting point to at least raise in your piece for those that might not have previously considered it.
It's only decent to be able to say that enough is enough. The case of Terri Schiavo shows that some parts of the world still have a long way to go in this.
The 'pro life' movement is probably going to transform into a 'do not delete' movement in the future :)
A relatively-recent example of how unexpected use can be made of pre-existing data - look at how search engines worked in 1997. The state of the art didn't know what signals were most important to look at, so they paid attention to things like keyword density. Then Google came along and illuminated a way more important signal, blowing away the state of the art for search engine ranking. Who's to say there is no analog for any kind of 'consciousness recording' that we have right now? Is it entirely implausible that a recording of my thoughts today won't provide significantly more information 50 years from now than it does today?
The problem seems akin to compressed sensing (http://en.wikipedia.org/wiki/Compressed_sensing), posted in a few articles on HN about a month ago.
Not only would you have to perform a 'near perfect' programming of such an intelligence, you'd also have to record it, so that is two big problems to solve.
The '50 years from now' has been around just about forever, make that 50 years. But in reality, in spite of all our amazing computer power we are not much closer to being able to do this than we were 20 years ago, and I don't expect it to be much different 20 years from now.
The article is addressed to those alive today, and it is possible that 100, 200 or 1,000 years in to the future the situation will be dramatically different. It's hard to imagine that it would not. But barring a 'singularity' or the development of something really technologically advanced (say large scale quantum computing) I have a hard time seeing how it could be sooner than that. So that's where the premise comes from, not that it is a complete impossibility, just that it probably is an impossibility in the lifetime of those alive today.
And, like the first link I included, it's interesting to consider the possibility that you consciousness could be recreated sometime even after you die based upon a recording of your activity, whether a blog post, a 'brain dump', or something else entirely.
I guess stepping back from the details, the questions is: why shut down the sense of wonder and possibility?
It's very interesting, that's why it makes for great science fiction. Science fiction can be an awesome inspiration for real science and plenty of things in real life today were science fiction not that long ago.
But the link between knowing that our brains store and process information and storing a copy of that state and bringing it back to life again in some other substrate (another brain or a digital simulation of one) is a very large distance away from our current capabilities.
> why shut down the sense of wonder and possibility?
That's not my intention. The thing that bothers me is that it seems as though a fairly large number of people are living under the impression that these things are 'just around the corner'. I'm all for trying to achieve some or all of this, but I think we should be realistic about our expectations.
The 'anti aging' market is the modern equivalent of snake oil and there are plenty of salesmen and more than a few of the gullible.
And 'uploading/storing/downloading' is another order of magnitude more complicated than simple preservation of what we have.
No you don't. You need to perform a near perfect simulation of the physics of a brain. It's not nearly as difficult as intelligence, since you only need a low level understanding of what is going on.
We are considerably closer to this than we were in the past. We can already simulate nerve clusters and some parts of vision processing.
We can't even simulate a single drop of water with any chance of fidelity, and we probably won't be able to do so in the near future.
These problems are hard, very hard. Hundreds of millions have been spent, and many billions more will be spent before 'a low level understanding' of what is going on will be available. And then you can begin to solve these problems.
A famous scientist is on the record for having said that within a decade they'd have a working simulacrum of a human brain, meanwhile they are slogging it out with others that claim to be simulating the brain of a cat: (http://news.techworld.com/personal-tech/3207325/scientist-ca... and http://www.brainmaya.com/neuroscience/neuroscience-videos/su...) but meanwhile we can't even vat-grow a cockroach from scratch.
I'm all with Dijkstra on this one, whether computers can think is just as meaningless as asking whether a submarine can swim.
A molecular simulation of a brain is extremely interesting but I doubt that that is the shortest route to AI, especially a simplified simulation.
Also, you can probably get away with neuron-level simulation rather than molecular simulation. As far as I know, most nerve cluster simulations do this.
[1] I learned stat mech as a senior undergrad, and didn't really get it until grad school. On the other hand, I built newton solvers when I was 13.
"Almost everybody that is past the age of 50 has had one or more life-extending operations"
Is this an American thing? In Australia it is pretty rare to have had any operation, yet alone a life-extending one, by the age of 50.
"But, let's assume that somehow - unlikely, but maybe it's possible - we manage to tweak these mechanism enough that we manage to survive and keep our bodies in the state of say a healthy 30 year old, forever. What are you going to do ? Sit indoors all day long ?"
The options open to someone who only faces accidental mortality are beyond our imagining. Imagine what a Jefferson or an Edison would achieve in 500 years. Imagine if Einstein had a 100 more to work on unified theories. Besides, with all that time to master knowledge I'm sure some bright spark could come up with a way to heal stab victims.
"The human brain has limited storage capacity, and even though we're fairly sure that we do not use all of it we also can't fill it up forever"
I'm not sure this is proven. Even if that's the case, I have a lot of useless knowledge about the maps of Quake 3 levels or the characters in obscure TV shows that could be jettisoned without great loss.
"That seems to be the problem with most of these far fetched ideas. 'Uploading' your mind, means measuring in great detail the state of your mind. And we do not even know what 'mind' is"
We have preliminary technology that can read thoughts and imperatives. Doesn't seem that far-fetched to map the brain given time. Besides, you seem to rule out human/brain interfaces - why can't I have a constant link between brain and computer and in a worse case death scenario I simply lose the experiences since my last link time.
However, you do raise some interesting societal issues. Nevertheless, "how will people deal with it" hardly seems a valid reason to prevent research that could let people live healthily to 500+. The alternative is we are saying people deserve to die because they are old.
That's ok, PG disagrees with me too, so you're in good company :)
> Is this an American thing?
No, just statistics. Most people have one operation in their lifetime, sometimes more than one, usually those are to treat conditions that would otherwise cause a substantial decrease in lifespan.
Appendicitis, gal bladder removal, heart stents and so on. I'm Dutch by the way, not American.
> The options open to someone who only faces accidental mortality are beyond our imagining. Imagine what a Jefferson or an Edison would achieve in 500 years. Imagine if Einstein had a 100 more to work on unified theories. Besides, with all that time to master knowledge I'm sure some bright spark could come up with a way to heal stab victims.
That would have been great, but it presumes that you can keep the brain functioning for that long as well, and in 'top shape'. As it is most 'brilliant' work has been done by scientists when they are relatively young, late bloomers are the exception.
And Einstein was pretty much stuck in a rut at the end of his days, no disrespect there, it's fairly well documented, he was such a formidable force to oppose that he managed to hold back black hole research until 10 years after his death.
> I'm not sure this is proven. Even if that's the case, I have a lot of useless knowledge about the maps of Quake 3 levels or the characters in obscure TV shows that could be jettisoned without great loss.
But how to tell those apart from the ones that make up your identity ? A 'garbage collection' algorithm would have to be pretty precise! Do you want to compress your memory Yes / No is probably best answered with 'No'.
> We have preliminary technology that can read thoughts and imperatives.
True
> Doesn't seem that far-fetched to map the brain given time.
Also true, but how much time. There are plenty of people that seem to believe this will happen in our lifetime and I think that's not even close.
> Besides, you seem to rule out human/brain interfaces - why can't I have a constant link between brain and computer and in a worse case death scenario I simply lose the experiences since my last link time.
Interesting possibility, but it combines all of the problems associated with AI, recording and programming through a 'link' the structure of some new brain. And where does that come from ? (that new brain, not the link)
> However, you do raise some interesting societal issues. Nevertheless, "how will people deal with it" hardly seems a valid reason to prevent research that could let people live healthily to 500+.
No, I'm all for the research. I just have an issue with the belief that this is just around the corner.
> The alternative is we are saying people deserve to die because they are old.
For the time being, that's the way it will be. Research is to be encouraged, taking in to account the social consequences in parallel. But this - if it ever comes to pass - will upset our society like nothing else ever has. But it won't be our worry. I think :)
Also, all those proposed new technologies to make you live a long (but finite) time sound really complicated.
I don't think that there is a difference between a simulation of a brain and a simulation of the mechanics of a brain. I believe that's the whole point of the Turing Test thought experiment. If it talks like a mind then it is a mind.
Actually, you're argument seemed pretty weak up until then, but Penrose was the last straw. The only thing worse you could have done is brought up the Chinese Room. Re continuity of the self, that's an interesting problem for philosophers, but it isn't a practical problem. You aren't your memories; more than your memories, you are your patterns and habits. But those also change, though not all at once. At 30, I am not the same person I was at 25, much less 20. For identity, there is just an apparent continuity of existence.
Personally, I think cryogenics is the key, as I don't think the technology for proper life extension will appear within my immediate lifetime.
Looks like we agree after all then.
Penrose's point is that the brain might be quantum mechanical in nature.
TL;DR: The smarter you get, the more interesting things you can work on. We're already smart enough to do interesting things for a long long time.
Are you suggesting the current system is flawless? I think we're pretty good at dealing with systems with problems.
I'd say it's a miracle it works even as is. I don't see how tweaking the miracle is any less realistic than the miracle itself.
:)
Best point in this article is the one about working memory. This is something I didn't understand at 20 but know too well at 40. The human mind is made to forget things -- lots of things. In fact, the ability to forget things is probably what keeps us sane and able to focus. Otherwise one thing would trigger another and we would just spin off on tangents all the time.
Having read the article, and agreeing with the points and sentiments, I still think the Great Singularity Cometh. Just nowhere near 20 years. More like 500. And I think by that time we will have a much different idea of what is human and what is not. So it all works out.
I've been meaning to set up getting frozen when I die. Not for immortality -- I think the problems Jacques mentions will work out but take a while -- but simply for life extension. I think the definition of "die" is a lot more flexible than we currently imagine and I think there's a chance that sometime in my lifetime we'll work out cryogenics to the point where there's a shot at getting another 20-40 years at some point in the distant future after you die the first time. (Although coming back to life with everybody else you know gone for hundreds of years has to be a very sobering thing.)
But the main, critical point here is that yes, you will die. And, apologies for the existentialist plug, but how you deal with your impending death, in a real and immediate fashion, is what separates you from a lawn slug. The denial of one's death through whatever fashion is a great travesty on what it means to be human.
And I'm almost certain the Easter Bunny isn't coming for somebody here, but I'm not going to mention any names. Starts with a J, though.
To some degree, both of these beliefs are true. But in other important ways they are not. However as hackers we have subconsciously and deeply bought into these two beliefs as part of our worldview, so in our hearts we "know" Jacques is just a sourpussed Luddite -- even if we can't exactly work out why.
I'll go a step further and say that our natural response to being confronted with the limits of our belief system is to thrash around quite a bit. Kind of like a bug getting pinned in a display book.
That makes the comments here very interesting reading!
Wow - imagine if we could repair machines! We could have vintage cars that were actually drivable instead of mounds of rust. We could have planes from WWII that, if anything, fly better today than the day they were built. We wouldn't have to throw things away the first instant they exhibited any sign of trouble.
Oh, wait. We do have all those things.
Sarcasm aside, you know what, give me the indefinite lifespan and I'll take my chances with the boredom and accidental death and brain overflow and collapsing society. Cause I'm sure all those things are bad but guess what, it's better than being dead.
However bodies are not even in the same few orders of magnitude of complexity as an old car, spending 200 years plugged into Facebook probably isn't the most productive thing for the rest of society, and your brain still can only hold so much and you still be you.
So -- points for pointing out the flaws, but no credit for failing to see the larger truth. Yes, being alive forever beats dying. Now on to step 2, where you start critically thinking about the odds of you personally living forever and the real trade-offs with not dying. Which is Jacques point.
The main point where Jacques blew it? He used our current definition of human as a working definition for the future. This species is evolving, and evolving at a super-fast rate. I don't think we would recognize our descendants ten thousand years from now as being human in any way.
your brain still can only hold so much and you still be you
Well by that logic I've died several times already. I'm very far from the same person I was at age 5 or even 15. I don't mind changing and evolving, even if you want to call that a kind of death. It's so gradual that I doubt I'd even notice.
on to step 2, where you start critically thinking about the odds of you personally living forever
I do not pretend that the odds are favourable, but that is orthogonal to all this justification about why dying is good and healthy. Dying sucks, man, no matter how much you try to dress it up as a good, socially beneficial natural part of life.
the real trade-offs with not dying. Which is Jacques point.
Well, I guess my point is that I could not care less what the trade-off is, anything is better than dying, because any problem can be solved, but you can never come back from death.
> give me the indefinite lifespan and I'll take my chances with the boredom and accidental death and brain overflow and collapsing society. Cause I'm sure all those things are bad but guess what, it's better than being dead.
The reality is though, that chances are much better than even that those things will not come to pass in our lifetimes, or even in our grandchildrens lifetimes.
So in spite of it being 'better' than being dead, chances are that you will die. What's so hard about it?
There are billions of data points (all the creatures that ever lived have died, with the exception of those that have not yet lived long enough to outlive their natural life span) and there isn't a single one that lived longer than 1500 years (an Olive tree, if I'm not mistaken).
Well, I think you're underestimating the pace of technological progress. Yes, I might not personally live to see the day, but I don't think it will be that far away. There are a lot of rapidly aging rich people. Research is increasing in pace, not decreasing, and efforts like SENS are only just getting started.
Those billions of data points are largely irrelevant on a timeline where technology only increases in potency. Historical death rates offer almost no guidance to a future where the causes of those billions of deaths are being isolated and eliminated faster than ever. You do know that according to Moore's law, computers will overtake the human brain in the 2020s, right? Where's the historical precedent for that?
You might have a point that immortality boosters are motivated somewhat by hope and self-interest, but you seem to be biased in the other direction, for no reason I can discern.
It is rational for any living being to seek to extend their life. Admitting it is possible and desirable is merely the first step. Why not considering taking it? There's still space in the 300. http://www.mfoundation.org/index.php?pagename=mj_donations_t...
I think you have the possibility to see it. We may not discover how to stop death in the next 30 years, but I'm sure we will have found a way to extend it enough so that we are alive when someone does.