Caution! This post contains a lot of words and no pictures. Alice would not be happy.
Whenever I mention I’ve been trying to write a novel for the last six years and people ask me what it’s about, I usually answer “prevarication”. It isn’t really, but the actual plot is a bit philosophical. Actually, the actual plot is quite simple, I’m just no good at writing it.
The lady hero in this novel is/was studying for a PhD in something to do with pure mathematics and science, it being a world I am vaguely aware of. But I wasn’t sure what that would be until I reached the thinking for my third unwritten novel. Yes, there’s a second one as well.
I was thinking about the latest Mars rover landing earlier this year. The Mars Reconnaissance Orbiter is still in orbit, over 10 years after its own mission ended, and acts as a high-speed relay station between Earth and whatever Mars mission happens to be on the ground. And that got me thinking about bandwidth.
In a normal exploration mission, the satellite would send back a relatively low-powered signal to Earth, which would use really large ears to listen out for the signal. Low-powered transmitters use, obviously, less power and also saves weight. But the amount of data that can be transmitted is relatively small.
I say “relatively” here because satellite transmissions are not the same as the terrestrial ones we all know and love. The theory is the same though; a lower frequency signal will travel further but can carry less information that a higher frequency signal. That’s why speech-based radio stations tend to use Long Wave (BBC Radio 4) while music stations use VHF. See also mobile phones.
Interplanetary network
So I had this idea. What if we could put a series of satellites into geosynchronous orbit around the outer planets, plus a few in the asteroid belt? Then if we have a satellite in orbit of, say, Triton (Neptune) or Oberon (Uranus) doing some scientific study, it could transmit at a higher bandwidth because it would ‘only’ have to send the data to the nearest satellite in the interplanetary network (IPN). That satellite would then store and forward to the next nearest towards Earth, and so on until Earth was reached. Or the Moon, or a similar satellite in Earth orbit, either of which would then downlink to wherever the data needed to be for analysis.
That’s quite clever, actually.
Another option – which I only thought of while writing this post, so I haven’t fully thought it through – is to create a series of relay stations at key points along the ecliptic, rather than in specific orbit around one of the outer planets. The problem would be maintaining position relative to Earth and the rest of the satellites… but I can fix that later.
Of course, we still have to get those satellites out there… which is why I thought of a seed ship. Unmanned (and unwomanned), piloted by a machine-learning artificial intelligence that could fly the craft roughly where it needed to go, before jettisoning a module containing three IPN satellites. Think of three cans of beer on a plastic container (why not) but with an engine and a small amount of propellant to slow the mini constellation down so that it achieves orbit. The three IPN satellites would then be released and use their on-board jets to position themselves.
A question of narrative
So far so simple. The science is fairly stable. The IPN satellites would be 12-sided; six with receiver-transmitters, the other six with positioning nozzles. They would in the form of regular dodecahedrons (or duodecahedrons), which is a handy word to abbreviate if necessary.
The narrative problem I set myself was that I wanted to have the seed ship’s AI take part in regular mission update meetings. It would learn about the people back in Darmstadt, and their lives, and react accordingly.
This is, of course, impossible.
The whole point of having an AI on the seed ship is that messages from Earth would take a long, long time to travel there and back, even at the speed of light. It takes 1.2 seconds between Earth and the Moon; 8 minutes and 17 seconds from the Sun’s corona to Earth. Earth to Neptune would be around 246 minutes – over four hours.
So, I need a way of speeding up communications between the seed ship and Earth and, guess what, it can’t be done. Not in any sensible way, at least. I wanted to avoid the regular scifi ideas of subspace and hyperspace. These things are too much like magic for my liking, and have no scientific grounding. I mean, if in Star Trek people can communicate at near instantaneous speeds via subspace… and if you are effectively turned into a radio signal when you use the transporter… why can you be transported via subspace? Oh, there’ll be a made-up reason… power consumption or suchlike. I mean, the whole thing’s made up, right?
Tachyons are another suggestion, but they have been misused in so many different ways it’s hard to know what a tachyon actually is or does any more in science fiction. In science fact it is a theoretical particle that has yet to be measured as so cannot be proven to exist.
So long, free will!
It was while pondering how FTL or superluminal communications would work that I hit upon the following thought experiment.
Imagine a single light source projecting onto a surface. This could be a fibre optic cable, a torch, whatever; the point is that the beam of light is projected onto the surface (which could be an OHP screen, blank wall, whatever). Flick a switch and the light comes on. Flick the switch and the light goes off. You can see the light appear and disappear on the surface.
Simple enough, no? The light travels at the speed of light and hits the surface.
Now imagine you have two light sources and two switches. Flick the switches on and off at the same time, the lights come on and off at the same time. That’s a bit difficult to coordinate, so wire the two switches together. Flick one switch and two lights come on. Nothing controversial there.
Now image there is a tube between the light source and the surface. One per light, with closed ends. The tube is made of a material with a very low refractive index, so that the beam of light doesn’t vary or slow down much. Flick the switch, and the lights appear as before – it’s just that there are now two tubes between the source and surface.
Now: imagine that one of the tubes is filled with a super-cooled sodium gas. This is because physicists have a little party trick, by which firing a beam of light through a super-cooled sodium gas slows down the apparent speed of light. If you don’t believe me, here’s a write-up of such an experiment.
What happens now when you flick the switch is that the light source passing through the gas is appreciably slower than the light source passing through the empty tube.
Let’s move on a bit.
Instead of flicking a switch, let’s automate the process with a sequence of coloured lights running red-amber-green-amber-red. When the ‘box’ runs through the sequence it should be much more obvious that the lights are not hitting the surface at the same time.
Now: logically – and people such as Einstein agree on this point – if I can vary a constant such as the speed of light so that it is slower than it should be, then I can speed up the speed of light so that it’s faster than it should be. Logically. Until proven otherwise, of course, and the Second Law of Thermodynamics is a good illustration of this point. Entropy increases or stays the same but it never reduces, which can be proven.
Let us assume, then, that there is a certain combination of gas or gases at a specific temperature that increases the speed of the beam of light as it passes through. I don’t know what those gases and temperatures might be any more than you do but, logically, they must exist (until we can prove that they don’t). Drain the gas out of the tube and replace it with our new combination.
Et voila! As the coloured lights cycle through we can now see that the light passing through the tube with our special gas hits the surface before the one that doesn’t.
Now for the fun part (he says, after 1,400 words).
Replace that automated switch with three buttons; red, amber and green. Instead of a microcircuit doing the switching, you’ll be doing it yourself.
Congratulations! You’ve lost free will.
When you press, for example, the green button the light that doesn’t pass through the gas appears at the speed of the light. But the other beam is travelling faster than the speed of light, so you see it appear before you’re pressed the button. And you can’t change that; if you see a green light appear you must have always been about to have pressed the green button (where’s Dr Dan Streetmentioner when you need him) and you can’t change that. Not without ripping a hole in the space-time continuum, anyway (probably).
On the one hand, effect (the green light) follows cause (you pressing the button). On the other hand, effect precedes cause.
It’s a cracker, isn’t it?
Satellite delays
One other bugbear of mine is when programmes such as Star Trek show ‘instantaneous’ communication (via subspace, of course) taking place.
There is no such thing as instantaneous communication. Even when you talk to someone next to you there is an imperceptibly small delay as the sound of your voice reaches the ears of the recipient. It’s just that the delay is beyond the ability of people to perceive. This delay exists at the speed of light as well.
Consider Alice and Bob. There are 186,282 miles apart, so when Alice communicates with Bob her messages take one second to reach him (the speed of light being 186,282 miles per second in a vacuum). Sam is a spy. They are halfway between Alice and Bob. If it takes one second for the end-to-end communication then Sam intercepts the signal at 0.5 seconds. You might remember that distance = speed x time from your schooldays.
If Alice can somehow speed her signal up to 10 times the speed of light, it will take 0.1 seconds to reach Bob, and Sam will intercept in half of that time. However fast Alice communicates to Bob, Sam will intercept in t/2 seconds.
So if Alice communicates literally instantaneously with Bob… when does Sam intercept the message? You can’t have half of instantaneous. Using our d=st formula above, we’ve divided by zero or divided by infinity (depending on how you arrange the equation).
This is, of course, impossible (unless you change your maths. Which we haven’t).
So, that’s that then
The funny thing is, while writing this blog and laying out the problem on paper – instead of keeping in my head, where it’s been for months – I’ve come up with a solution. Not to superluminal communication, I mean the problem of having the AI somehow invested in the daily lives of the Flight Controllers and mission team.
All I have to do now is write it.