What I’ve been reading

The Age of Spiritual Machines by Ray Kurzweil – Wherein the author argues from 60 years of historical data about increases in computing power that commodity computers will exceed not only the processing power of the human brain but that of all living humans by the end of the 21st century.

Coupled with advances in sensing technologies such as MRI, we should soon be able to (either destructively or not) transfer human minds into computer simulations of working brains. Eventually the majority of human civilization will be in digital form and a staggering array of new kinds of interaction will become possible.

Naturally I hope he’s right, but I think his predictions are a little too optimistic in terms of timing.


S. Andrew Swann: The Hostile Takeover trilogy (Profiteer, Partisan and Revolutionary) – Pretty good as political science fiction goes. An eclectic cast of non-stereotypical characters, set in an interesting universe that makes me want to check out some of his other books in the same setting. There are a couple of places between the halfway point and the end where things get a little too deus ex for my taste, but thankfully all that gets out of the way in time for the more conventional resolution.


The Grand Design by Leonard Mlodinow (with Stephen Hawking’s name on the cover to boost sales).  Basically it’s a science popularization book specifically about quantum theory, what it does and doesn’t do, and what we’ve learned from it.  Short and easy to understand.


The Complete Maus – My long relationship with this book has finally come to a close.  I first noticed the series in the the college library when I was in my mid-teens.  Leafing through it, I could tell that it contained good storytelling and told an important tale, but at the time I had little interest in history, so I shelved it with a note to read it through later.  More than 20 years later, that “later” has finally arrived thanks to someone at work selling the compilation at a good price.  Well, having read it, I have to say it was rewarding.  It’s riveting in the way all tales told by survivors of major events are riveting.  Not much more to say than that; go read it.


Trouble with Lichen, by John Wyndham – How rare, a pro-immortalism story with a happy ending.  It’s a short and easy read, and well done.  It mostly concerns means of manipulating the masses into accepting longevity advances, something individuals are strangely resistant to and institutions actively resistant to.  Recommended.


Fleet of Worlds by Edward M. Lerner and Larry Niven – a very enjoyable Known Space story that reads like a classic Niven.  There are a couple of flaws with the setup and teardown; the ending, first off, seems to conflict timing-wise with established canon about the fleet in other Known Space stories.  Secondly, I have trouble reconciling the setting for the story (a human colony on the fleet) with known behavior of the Puppeteers – it’d be a real stretch for them to do what the backstory claims they did.


Year Million edited by Damien Broderick.  It’s a collection of essays by a variety of authors, most of them scientists, ostensibly about what they think our destiny over the next million years is likely to be like.  Many of them go straight for the big picture though and talk about all of the future.

My favorite part is this quote by Steven B. Harris, who articulates something I sort of unconsciously knew all along but am excited to see revealed clearly: “Homo Sapiens Sapiens is now largely a software species, perhaps the first, governed mainly by epigenic factors (outside the genome), some of which are extrasomatic (outside the body).  Much of what makes us special as a species is stored not in genes or brains, but in libraries, laws, traditions, and songs.”

This is so true.  Obviously our genes are somewhat important because they give us the hardware capable of running the minds we’re so proud of, but the hardware by itself is useless; a child by itself does not grow into a human being, nor can it be made into one after it has grown on its own past a certain point (as I believe is demonstrated by studies of feral children).  Everything we consider important about ourselves as individuals and as a species is programmed into us after we’re born by those who teach us.  And the continual improvement of that programming is what has enabled us to improve our lot at an exponential rate.

The rest of the book I found rather depressing though, for two reasons.

First, many of the authors keep coming back to Matrioshka Brains, and collectively they make the case that building such devices is the logical way to squeeze the most living out of the universe in its present form (ie while there are still stars), and that the drive to maximize life will ultimately compel all sentients to build such things.  I can’t disagree, but I don’t like it either – it’ll make the universe a boring place.  Suppose we were to build an M-Brain for ourselves.  We could have trillions and trillions of humans living it in happily for billions of years or more, but to build it we’d have to disassemble the rest of the solar system – there’d be no other places to live or to explore.  We would basically turn inward, living in simulated universes of our own devising instead.  That’s not for me – I want to explore the real universe.  So we’d probably end up with an expanding shell of people like me, moving outward to explore real places while they still exist, and behind us the stars would be going dark as more and more of them are wrapped in minds or their energy redirected to power minds elsewhere.

But then again, stars (and matter itself) won’t last for ever, which brings me to the second depressing part.  Recent discoveries suggest that not only is the universe expanding, not only is it not likely to contract again, but the rate of expansion is actually accelerating.  These measurements still need to be confirmed, and the baffling question of how this can be happening could use a good answer, but for now let’s assume it’s true.

One consequence of this is that the observable universe will shrink over time – the furthest galaxies will fade away as they accelerate to the point where even the light they shine directly at us isn’t fast enough to outrace the expansion of the space in between.  Superclusters of galaxies may stick together, but the gulfs between them will get larger, and that means the limit of the amount of interesting places we can explore gets smaller with time, even though the universe itself is getting larger.

Another is that the majority of time in this universe – basically all of it – will be spent in an uninsteresting state where there is no matter and almost no energy, and less energy all the time.  We’re living on the slope of an exponential downward curve in the abundance of energy gradients (which are necessary for any kind of thought).  Eventually the stars will go out and stop being born.  Then the black holes will evaporate.  Then, assuming we’re correct in our belief that protons can decay, all matter will evaporate into subatomic particles and the universe will contain only energy, radiating out in all directions and thus becoming more diffuse.  The temperature of the universe will become more and more uniform and will continue to drop towards absolute zero – but will never actually get there.  It’s impossible.  So there will always be some energy gradient around – always less and less, but some.  And this process will continue for ever.

Imagine then that we can devise some way to survive the death of matter – say we can create a cloud of particles that can compute thought and do so using any available energy gradient (though of course it’ll have to compute more slowly as less energy becomes available).  Now we could live forever, but it would be, to me, a very claustrophobic existence because there would be nothing to do but live in simulations – navel-gazing in our memories, or trying out randomized simulations for all eternity.  (Perhaps we’re actually living in one of these right now and have deliberately forgotten, to make it interesting).  Because this environment would have to compute more slowly as the available energy dissipated, the ratio of real time to simulation time would increase.  At first perhaps we might think faster than real time, but after a time it would get to the point where the components of our immaterial computer might only exchange a photon of information every trillion years, because there isn’t that much to go around.  A hypothetical person inside the simulation watching a clock count off the real-time years would see the number rising faster and faster all the time, then the number of digits in the year would start rising faster and faster, and so on.  But other than that one indication, the rest of us would never know.

I find this depressing and really hope we can engineer a better solution for our long-term destiny, though in a way this is still better than a Big Crunch because it means an infinite existence of a sort is possible.

But, all of this is still a ways off.  We’ve got half a billion years or more to enjoy Earth, then perhaps five billion years with the solar system in its current form, and hundreds of billions until the stars in general start to fade.  I hope that’s enough time to come up with some answers to these pressing problems.


Comments are closed.