Hacker News new | past | comments | ask | show | jobs | submit login
The 100 Year Starship (wikipedia.org)
39 points by mmhsieh on Jan 17, 2020 | hide | past | favorite | 57 comments

What if Country X starts building a starship in, say, 40 years and it works and off they go.

Then Country Y invests in new technology and starts building their new ship in, say, 100 years.

It's possible and likely that Country Y's starship is more capable and faster than Country X's ship as they would have waited for and capitalized on scientific breakthroughs that make interstellar travel better.

This is the same idea as waiting to buy a computer until they're faster, on the assumption that the faster computer will give you the answer sooner. About twenty years ago, this idea was quantified and published in the paper "The Effects of Moore's Law and Slacking on Large Computations", which you can read here:


Country X will have 40 years of experience of actually being in space and might learn much more valuable knowledge than the theoretical advances made by country Y.

You're assuming the 60 years after Country X's starship is launched is going to be a period of progress.

If it's a period of collapse, perhaps complete collapse, humanity will have lost an opportunity by delaying the launch.

> It has been argued that an interstellar mission that cannot be completed within 50 years should not be started at all. Instead, assuming that a civilization is still on an increasing curve of propulsion system velocity and not yet having reached the limit, the resources should be invested in designing a better propulsion system. This is because a slow spacecraft would probably be passed by another mission sent later with more advanced propulsion (the incessant obsolescence postulate).


There is a fair amount of thought that has gone in to that, often referred to as the Wait Calculation.

An interesting example was "Interstellar Travel - The Wait Calculation and the Incentive Trap of Progress"¹, which somehow isn't on libgen [yet?].

1. http://www.jbis.org.uk/paper.php?p=2006.59.239 - The incentive trap of linking to a £5 download for fifteen year old paper is another paper entirely.

The "incentive trap" has also been referred to as the "incessant obsolescence postulate", in this paper from 2011 titled "Energy, incessant obsolescence, and the first interstellar missions":


The author also uploaded that paper to Researchgate, where it's available for free:


(edit: I presume the downvotes are for linking to Researchgate? I know they're not held in high regard like Arxiv, they're more like a sleazy Linkedin for academics, but if there's valuable content there why not use it?)

Are you asking for a value judgment? Or what exactly is your question? This can happen for example in some games like Master of Orion. You send a ship to the border, and before it gets there, you've already researched, built, and deployed a faster version. Sometimes it changes things, sometimes it doesn't. If your sole purpose is simply to arrive then it hasn't changed a thing. If you are planning to fight aliens when you get there, however...

This is a plot point in various SF works.

For example, in the Honor Harrington books by David Weber, the Manticore system colonists traveled to their target a couple hundred years in hibernation. But as it was likely propulsion technology will advance before they reach their destination, they also left part of their money in a trust fund back on Earth, to make sure the Manticore system won't get squatted in the meantime.

This proved to be a sound idea, as practical FTL drives have been invented after the colony ship has departed and they have been greeted by the Manticore Trust navy upon arrival, that has secured the system a couple years earlier, arriving by the newfangled hyper drive. :)

Very nice. Alastair Reynolds touches on this concept as well in Chasm City.

This idea is in the game Elite Dangerous, and you can go and look at them: https://elite-dangerous.fandom.com/wiki/Generation_Ship

The lore is quite amusing and poignant. Many of them met a sad end in one way or another, reminiscent of Fallout.

Interesting. 100 years is a long time. Would this be a generation ship [0]? If so, I'm curious about their discussion of the moral and ethical considerations of committing children to such a journey.

[0] https://en.wikipedia.org/wiki/Generation_ship

The project aims to study how we can develop interstellar travel technologies in 100 years.

Not to develop a ship that can travel for 100 years. (Even though that might be the case)

Is this program dead? I had a look through the website and couldn't see much to suggest otherwise other than their writing prize in July last year. The list of partners too was a bit weird, all branding consultants and design agencies.

> the endeavor was meant to excite several generations to commit to the research and development of breakthrough technologies to advance the eventual goal of interstellar space travel

It doesn't seem that the purpose is to come up with a working concept but rather a concept that's interesting enough to push the topic forward and tickle the imagination of scientist, engineers, etc. for generations until the practical implementation is achievable. So I expect right now branding and design are more useful in popularizing this.

So, where is that? That branding and design they made?

Seems like this was a big scam.

Sometimes when humans try to get a hand-wavy initiative started, all you get is hand waving.

My uninformed take on it: if the initiative was more on the level of "Here's my amazing ship design, prove me wrong" they would have gotten way more involvement. Even if Version 0 was a Jules Verne design...

I hope the future of interstellar travel is not these gigantic starships. It's like starting a project of building a gigantic Zeppelin in 1900 in order to get able to get to moon in 100 years. Teleportation seems more convinient to me and much less financially and socially devastating.

>Teleportation seems more convinient to me and much less financially and socially devastating.

Sure, but there is that pesky problem where the solutions need to be theoretically possible.

Both teleportation and Death Star are impossible with current technology, but the folks anyway are building the Death Star?

Let me introduce you to the Nicol-Dyson Beams: https://www.orionsarm.com/eg-article/48fe49fe47202

This actually fits nicely with the name "Death Star", as you take all the energy or a star (via an orbiting Dyson swarm) and turn it into a concentrated beam of death.

a.) Who is building a Death Star?

b.) Death Star isn't theoretically impossible. We know you just need to direct a massive amount of energy at a planet to blow it up, the only question is figuring out how to get that kind of energy. Teleportation though, we literally have no idea how it would work or what it even means.

Engeneration is an interesting very high tech "teleportation" alternative:


You basically send your digitized DNA & personality copy to the target (needs to have reasonably high tech infrastructure already in place), where a blank body is grown & "flashed" by your personality. And voila, you are now walking the surface of another world couple light years away from your destination. (And you likely also now effectively exist at least twice, but oh well.)

There are serious ethical and security issues with this technique, but it might still be useful none the less. :)

>And voila, you are now walking the surface of another world couple light years away from your destination

No I'm not, my clone is. My consciousness is right where it started, making this an extremely limited form of "travel"

Altered Carbon, gotcha.

"... a couple light years away from your transmitter"?

There's a lot more people with money willing to invest in a planet-destroying weapon than space exploration, unfortunately...

There should only be one Manhattan style project now: figure out how to download consciousness into a machine. Besides conferring immorality, this would also make interstellar travel time scales irrelevant.

If we could do this wouldn't we spin up minds-in-machines to do all sorts of cognitive tasks on demand and shut them off when the task is done (or they abandon their task) like a virtual machine.

Imagine making a copy of someone like Fabrice Bellard [1] and spinning up 100 of them to write your flappy bird clone for you and then shutting them off.

Would this be a terrible crime, or a paradise? This is an idea known as Hansons "ems" [2].

- https://en.m.wikipedia.org/wiki/Fabrice_Bellard

- https://ideas.ted.com/are-you-ready-for-the-impending-age-of...

This is a pretty serious question and something that really should be answered. Is a digital conscious alive, and if so what rights does it have?

Ian Banks' Culture novels look at this a few times.

Appreciate the article links, will read them after work today.

It's not at all a serious question. It's entirely speculative and there is zero hard scientific evidence that such a thing is even possible. Science fiction.

Philosophy definitely can contemplate theoretical scenarios

That sounds like the "Fast Folk" from Ken MacLeod's Fall Revolution books:


>this would also make interstellar travel time scales irrelevant.

Not really. You're assuming this "machine storage" mechanism can preserve state forever. Even if you manage to upload consciousness losslessly and maintain all the ineffable qualities of "life" or "soul" or whatever (that we don't even really have solid philosophical definitions for), you'd still need to contend with data corruption and bit rot over cosmic time-scales.

There are many error correction techniques suitable for this, such as for example fountain codes:


You basically encode a message (say, it's 10 some data storage unit unit long) into a much bigger lot of chunks (say 1000 0000 units), which you then transmit. As long as you receive a pre-determined amount of the chunks (1000 units might be enough), regardless of order, you can reconstruct the original message.

With speed of light lag, you can't really expect re-sends & will have to use some encoding working on a similar principle to this one.

Although I think "immorality" was a typo and you meant "immortality", it's a funny typo, because ethics are definitely relevant.

Frank Herbert's take on this had humans enslaved for a few centuries before the Butlerian Jihad got humans back in control of themselves and instituted a ten thousand year taboo on 'thinking machines'.

Sci-fi is great for postulating dystopian outcomes so we can at least anticipate bad choices and guard against them.

I can probably train a simple AI to be "close enough" to lots of people. Does that count?

More seriously, if we can send software, why care about this muddy consciousness stuff?

I fear that the rush to integrate humans with machines will miss something essential and end up replacing true consciousnesses with philosophical zombies.

Presumably there will always be some culture of holdouts who resist the upload process kind of like how the Amish resisted the industrial revolution.

Of course, if the philosophical zombies end up being evangelistic transhumanist zealots, we'd be screwed.

Conferring immortality?

Are you under the impression that hardware never breaks down, or that media never rots, or that software never fails?

I guess if the transfer process is perfected then you don't have to worry that much about HW failure or rot. Very little of today's data actually disappears due to these causes.

Imagine today's equivalent of having your data on 2-3 hard drives, in the cloud, on 3 email addresses, and in a printed copy. You will backup a brain dump every evening to a satellite backup storage and 3 data centers around the world, maybe even to a cold or hot spare body. Someone stealing or altering a copy of you, or altering you (lots of philosophical questions around who is you anymore) would be the real worry.

If anyone told you today your life depends on a piece of electronic data in your possession I'm sure you wouldn't let HW failure or data rot become a problem. Your birth certificate is in a single paper copy and it's damn hard to get a replacement and yet most people have no problem keeping it safe for decades.

> Your birth certificate is in a single paper copy and it's damn hard to get a replacement and yet most people have no problem keeping it safe for decades.

Decades might as well be seconds if we're talking about interstellar travel times.

Sure. But pieces of paper might as well be nothing compared to the technology we have available once we're capable of downloading consciousness into a machine and interstellar travel. So the challenge may be bigger but also the tools at your disposal.

You are projecting today's problems on tomorrow's technology. And I'm being generous here, we're actually talking about multiple generations probably. We're just as close to understanding such future technology as people (scientists) 100-150 years ago were to understanding ours. Saying "in 200 years we'd surely lose the data to HW failure and bit rot" is no different from them saying "you can't fly without flapping the wings" 150 years ago.

Those future generations face challenges we can't even imagine today. Like if I whip up 5 copies of myself who is "me"? And if 4 died will it be a loss of data or just loss of redundancy? Will it actually be just one combined copy running on multiple machines (Borg style)?

>You are projecting today's problems on tomorrow's technology.

Entropy is just a fundamental aspect of how time progresses. It's basically axiomatic that anything we have degrades without actively recreating/reinforcing it with additional work.

You can't have a perfect transfer process, thermodynamics prohibits that, and you have to have some kind of physical media which will fail over time, and thus require regular upkeep and maintenance, as well as a constant supply of power (which is also not infinitely available or free.)


But there is such a thing as "good enough for all intents and purposes". I didn't say any transfer has to be perfect, or that the media is absolutely perfect, or that it has to be done for free, with no energy consumption, etc. just that if we achieve practical interstellar travel and can transfer consciousness to a machine I'd say the tech will be advanced enough for these issues to not be the major concern. You're introducing additional constraints and expecting my previous comment to account for them.

Theseus's ship, theoretically you are not the same you from 10 years ago, cells have changed, DNA was altered, shape, consciousness, etc. are all slightly different. But for all intents and purposes you are you. So having a "good enough" copy of yourself may just be enough. Once it's in a machine it does not matter anymore. Thermodynamics allowed this comment you are reading now to be transferred from my mind to your mind (in its literal form, regardless of personal interpretation) or at least from my keyboard to your screen exactly how it was intended, with no discernible loss of fidelity.

You're taking the purely theoretical stance where it's not the same photon of information. I'm taking the practical stance where it doesn't have to be. And even losing some bits of your data might not be a practical concern.

Presuming technology has advanced to the point where transferring a consciousness to a machine form is possible, it can be presupposed that the power and maintenance requirements would have been worked out in parallel.

The main thread of discussion is interesting enough to set aside quibbles about support technology that would have to be developed in tandem for it to be a viable process.

We definitely need to get our environmental problems and ressource consumption under control first.

That sounds more like a recipe for missing the window of opportunity, that should over time fix all our environmental & resource issues once and for all thanks to the (in comparison to Earth) limitless resources of Earth.

What about reproduction?

We can do both. One doesn't limit the other does it?

cp -R me/ me2/

Yep, I would fork myself.

Merging and deconflicting if you want to reunite forks could be a fun problem to solve in the future

You might enjoy the Bobiverse trilogy: https://www.amazon.com/gp/product/B0753LBFQ7/

Future person: "Who are you?"

Future forks: "One of us."

(Thanks to I, Robot for a cute scene.)

"Did jesus die for Klingons?" can be answered the way that Mormons answer the question for Americans.

That is, Quetzalcoatl came to the middle east to offer the good word.

(I always tell Mormon missionaries about Quetzalcoatl and they haven't heard of him, but the Quetzalcoatl cult did start around the same time as when Jesus came +/- 100 years or so.)

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