Hacker News new | past | comments | ask | show | jobs | submit login

The Milky Way lacked rocky planets early on, making that 12.5 Billion year age largely meaningless.

Which is just one of many bad assumptions in how people do these calculations. At a minimum you don’t get a straight path from A to B. Assuming any kind of limitation in distance traveled and your at the mercy of the distribution of whatever resource you need. Aka need rocky planets with liquid water, that’s likely going to be an very indirect path. Even just non binary star systems is significant. Further, assuming every trip is successful is again unlikely. How long you need to wait before the next trip is again a major qualifier.

For example, what happens if they first send a probe to verify habitability? Suddenly travel times more than double.

I have seen plenty of optimistic calculations that still add up to over a billion years before total colonization and sometimes much much higher than that. And that’s a billion years where expansion is a major priority at the frontier. Look at successful colonizations that don’t expand rapidly and again things keep slowing down.




One effect not considered in early considerations was the differential radial speed. So even the ability to seed planets 10 light years closer or further from the galactic core over a few 100M years results in expansion much more quickly than earlier models.


If you’re talking 10 light year colonization jumps that takes very deliberate and resource intensive expansion efforts. If you instead model things assuming generation ships as harvesting kepler Belt objects, moving between stars only incidentally or as local resources are exhausted. And only occasionally reproducing. The relative motion between stars becomes the primary driver of net expansion, and that’s quite slow.


> For example, what happens if they first send a probe to verify habitability? Suddenly travel times more than double.

This doesn't seem like a good idea if their method of traveling is generation ships. What would be the point?


One obvious consideration is to see if anyone is living there. Even just the basic economic consideration of spending a low fraction of the cost of a habitable ship to collect more data. Generation ships also implies different people will be arriving, however having a X thousands of year lifespan is another possibility. Which is kind of the thing, sharing physics is a given but biology and especially culture could be very different.


If you used the Von Neumann machine approach wouldn’t it not really matter if the planet was habitable or not?


Let’s assume your sending a Von Neumann probe to prep the way for your ship. Now it shows up and does whatever it needs to and then sends an everything is ready signal before you send your generation ship.

I mean sure you could in theory send out an endlessly replicating probe that keeps going to other star systems, but that’s even more risky.


What’s risky about it? The risk seems in them overtaking the entire galaxy more than anything, no?


Building something with zero possibility of change across millions of years is difficult. Something as simple as reproducing endlessly means they end up wasting enormous amounts of resources. But, you also risk things like probes that don’t slow down and hit one of your planets.


Sure, but we’re now beginning to talk about infinite scale and infinite resources. Who cares if 1 planet doesn’t produce probes as fast or at all? Who cares if 1 probe slams through a planet? That being said, I think the latter case might be more controllable by a civilization capable of creating the machines?


The fear is that you destroy one or more of your inhabited planet not just a lifeless ball. Which I would hope you and most people would consider a significant downside. That risk is balanced vs the benefit of sending self replicating probes to every star system which just doesn’t seem that useful.

Now is that a large or small risk? I don’t have the technology but it’s something that could reasonably concern a civilization.




Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: