Tens of billions of potentially habitable, Earth-size planets in our galaxy | Hacker News Reader

Tens of billions of potentially habitable, Earth-size planets in our galaxy

kurzweilai.net

63 pointsatpaino69 comments

>One in five stars in our galaxy like the Sun have planets about the size of Earth and a surface temperature conducive to life

That is a considerable claim that jars with most of the astronomical data I've seen. 1 in 5 is far more probable, perhaps an order of magnitude more probable than I've previously heard.

Does anyone have studies that corroborate their claim?

[EDIT] Later in the article data is cited that more closely matches what I've heard.

>Independently, Petigura, Howard and Marcy focused on the 42,000 stars that are like the sun or slightly cooler and smaller, and found 603 candidate planets orbiting them. Only 10 of these were Earth-size, that is, one to two times the diameter of Earth and orbiting their star at a distance where they are heated to lukewarm temperatures suitable for life.

It makes for a much less sensationalist title, but it seems to have more research behind it.

Over all, I would question the method that caused them to arrive at the 22% idea. I'm not sure that they are accounting for the missed planets correctly; if they are, I would like to see their estimates corroborated.

[EDIT 2] Thank you, Tuna-Fish. Typo corrected.

> astrological

UGHH. Astronomical.

The previous studies have mostly been lower bounds, not attempts to estimate the most likely count. We are still essentially terrible at locating earth-sized planets in earth-like orbits, we are much better at finding larger planets in closer orbits. Because of this, most of the planets that we do find are larger and/or closer, however, this obviously should not be used to deduce that most planets are larger and closer.

This study is using honest statistical analysis to estimate the likelihood of earth-like planets from the data available.

Psychics at Cal Tech...

http://www.youtube.com/watch?v=M3DoDMfDa_Q

I think a key point is that, unless planets rotate in a plane that nearly osculates our line of sight vector they will not be observed by current methods. Basically any planet whose orbit from our perspective looks like an oval rather than a left-right wobble will not be observed by current methods which check for light flickers as the planets cross their star. There are complex factors at work here, but I'd think a reasonable approximation would be that solar systems plane of rotation are oriented randomly.

So, given random rotation, if we would expect to observe 1 out of 100 plants using this method and then find that there are planets around 1 out of 200, we can extrapolate that (1/100)/(1/200)= 1/2 of suns have planets. That explains the discrepancy anyway; since so many don't rotate in the right plane the fact that we observe them at the frequency we do, rare as it is, means they are in fact quite common. The universe just got a lot more exciting!

Our current methods of observation are resolution limited and have huge, huge blind spots. When you correct for that, you get the "1 in 5" number.

If you're looking for the astrobites treatment, I think this is Howard and Marcy's paper, or at least Howard and Marcy are authors on this exoplanet paper discussing this topic:

http://astrobites.org/2012/09/02/looking-for-planets-around-...

Although I think the reported 22% number comes from a different exoplanet paper on the topic of moon formation, probably on the assumption you need tidal forces for ... something:

http://astrobites.org/2011/06/16/searching-for-pandora/

I highly recommend the astrobites RSS feed. Formerly in my google reader, now in my newsblur.

In my mind, Fermi's Paradox looms very large.

http://en.wikipedia.org/wiki/Fermi_paradox

Honestly, I've come to disagree with the "paradox". The ability to send physical evidence of other intelligent life is based on a certain speed limit of the universe. So that doesn't serve as much of a lack of evidence. Habitable planets are too far apart for much interstellar travel to occur readily given that the speed limit is c.

Beyond that, the power of a radio signal is inversely proportional to the square of the distance, and we have things like magnetic eddies at the edge of our heliosphere, so it might just be incredibly improbable for any civilization to have generated the types of signal creation that we would pick up.

In the hypothetical/fantastical, other advanced civilizations might have long distance, superluminal communication technology, but that's not something our civilization could actively intercept.

tldr; I don't buy Fermi's paradox.

The Fermi Paradox doesn't necessarily say that intelligent life doesn't exist. The impossibility of interstellar travel is compatible with the Fermi paradox.

The point is that if interstellar colonization is possible, the age of the universe practically requires galaxy-wide colonization as an inevitability. Yes, it would take millions of years to colonize an entire galaxy. Tens of millions, even.

Guess what? 10 million years is only 0.1% of the age of the universe. So if one civilization develops interstellar colonization, they experience geometric growth and claim the entire galaxy.

So either there is no civilization that has a 0.1% lead on us (which is statistically absurd if Earth-like planets are common) or interstellar colonization is impossible.

It goes beyond that. Even if interstellar travel is not plausible, you'd still expect to find solar systems completely transformed by intelligent life (think Dyson spheres, although maybe not so extreme). If interstellar travel were merely supremely difficult, then you'd expect regional clusters of stars exhibiting these properties. These would be bizarre astronomical oddities that we'd have likely noticed by now.

I'd still maintain we simply can't detect it.

Damn, thank you for the informative response! This makes more sense, I misunderstood the Fermi paradox.

The Fermi paradox is "why a highly developed civilization isn't behaving like a Western colonialists or tourists toward primitive natives of some remote wilderness".

The universe is old. If habitable planets are common, then it's functionally inevitable that intelligent life exists with a billion year head start on us.

Would they eschew expansionism? Maybe. Would they always eschew it 100% of the time? Definitely not. And this is a Darwinean process - the ones that decided to stay home are the ones you don't meet.

And they've had a billion years to change their minds. Consider that the likelihood is numerous billion-year-head-starts, one species would inevitably try it.

i'm not saying that they wouldn't pursue expansionism. A half-billion years ago they could have had a thriving colony on our planet. And, as you said, there is a plenty of time to change their mind, literally, and bodies, so planet surface existence isn't a key anymore, or at least it isn't a big deal to let go the planet which has potential for development of new civilization. We're just so low on the development ladder, that we're a national park, designated wilderness, off-limit for the ones who can reach us.

As far as we understand physics, actually traveling through interstellar medium is very very slow. Light itself takes next to forever to go from here to there. Most likely, intelligent species realized it was much more fun to explore inner space rather than outer space. I'd wager we won't see them: they've surrounded their local stars with Dyson Spheres, communicating with highly directed, encrypted, massively wide-band, low power communications that would look rather like background radiation unless they were pointed right at us and we were meters away.

Think about how important latency is to high frequency traders. In an information ecosystem, the laws of physics become paramount: latency and bandwidth are king. Physically being far from the action is essentially choosing to banish yourself to the hinterlands when everything exciting/valuable is happening at the hot core as close to the ball of fusion in the middle.

update: Thinking further, if anyone has in fact developed FTL, you can bet your ass they don't want anyone else to have it. In space, Sir Isaac Newton is the deadliest MotherF&@$%^# around. A baseball going 99% C is a weapon of mass destruction (http://what-if.xkcd.com/1/), much less a ship, which would essentially be a planet destroyer you literally couldn't see coming.

Yup.

I'm pretty sure we're screwed and interstellar colonization is impossible. At best, we'll get a probe that takes a thousand years to tell us what it found, and we'll only explore a handful of neighbouring systems this way before something eventually scrapes humanity down the memory hole.

Perhaps most (all?) advanced civilizations annihilate themselves (or get annihilated) before they have a chance to colonize other worlds?

That's one of many theories. One of the other popular ones is that we're in some sort of galactic natural park until we're ready to cope with joining the civilized folks.

I've seen this reported as 2 (http://www.theguardian.com/science/2013/nov/04/planets-galax...) and 8.8 (http://abcnews.go.com/Technology/wireStory/study-88-billion-...) billion this week. Anyone know why the wildly different numbers being reported for the same research?

I'm also curious about Earth-sized moons around gas giants in the habitable zone.

The research says 22% of sun-like stars have Earth-like planets in their habitable zone. Seems like they're all based on different assumptions of the number of sun-like stars.

>Guardian: There are around 100bn stars in our galaxy, of which 10% are like the sun.

>ABC: There are about 200 billion stars in our galaxy, with 40 billion of them like our sun, Marcy said. One of his co-authors put the number of sun-like stars closer to 50 billion, meaning there would be at least 11 billion planets like ours.

>Kurzweil: The estimate was based on a statistical analysis of all the Kepler observations of NASA’s Kepler space telescope of the 200 billion stars in our galaxy. Given that about 20 percent of stars are Sun-like, the researchers say, that amounts to several tens of billions of potentially habitable, Earth-size planets in the Milky Way Galaxy.

n.b. Kurzweil's math clearly sums up to 8.8. Maybe he's adding in stars unlike the Sun and not mentioning it (given that the article discussed is based on earlier work showing 15% of red dwarfs have Earth-like planets in the habitable zone: http://www.nytimes.com/2013/11/05/science/cosmic-census-find... ).

Don't Earth-sized moons of gas-giants have a few others problems? Less consistent weather patterns due to a more complicated relationship with the sun, stripping of the atmosphere due to gas giants gravity, and crazy huge tidal forces from the same.

I'm not an expert, so take these with a grain of salt.

There are plenty of spots on Earth with unusual weather patterns (not to mention extremophiles). I'm not sure what sorts of weather changes a gas giant moon would see but I'd imagine they're surmountable via evolution.

Titan has a nice atmosphere, so it's possible for a gas giant moon to retain one. Gravitational force diminishes pretty quickly with distance, so if a moon's a reasonable distance from the giant it'll likely be fine there.

Tidal forces give Io troubles, but the other moons of Jupiter seem to be fairly stable geologically.

Tidal force can be a blessing instead. Some says that plate tectonics plays a vital role in making Earth a habitable planet. It requires continuous source of heat from inside, which is supplied by radioactive elements[1] on Earth, but in a gas giant moon, tidal force could play the same role.

[1]: (Off-topic) The fact that our whole planet's life basically survived thanks to the gigantic amount of radioactivity buried inside, hot enough to keep the whole planet's core molten for 4.6 billion years, gives me endless amusement whenever I see someone ranting about how radioactivity has no place in nature and is a deadly scourge we're inflicting on Earth.

Titan is a good existence proof that moons can retain dense atmospheres, but the exact reasons for it are complicated. Why don't any of Jupiter's large moons have dense atmospheres? Maybe radiation around Jupiter causes more atmosphere loss. Maybe it's just random. There's actually an argument that gas giant moons might be more capable of retaining atmospheres, all other things being equal, than equivalent free bodies, because gas molecules they lose will remain in orbit about the gas giant and eventually re-join the moon.

So, moons with atmospheres are clearly possible, but it's very unclear how frequent they might be, or how their prospects compare to those of regular planets.

I submitted the actual study for this yesterday and got no traction :\ https://news.ycombinator.com/item?id=6679443

Unfortunately every single one of them is unreachable within several lifetimes. We're going to be pretty much trapped in our solar system until we try to solve some of the more difficult problems.

We're (other people, not me) at least working on it.

http://www.nasa.gov/centers/glenn/technology/warp/warp.html

http://en.wikipedia.org/wiki/100_Year_Starship

http://en.wikipedia.org/wiki/Faster-than-light

If they're successful and you warp to a planet that's 100 light years away then warp back, I don't know how much time would have passed on Earth though.

Depends on their velocity. Assuming a warp of 95% the speed of light, then ~640 years (~320 years one direction). See:

http://www.fourmilab.ch/cship/timedial.html

http://physics.stackexchange.com/questions/31105/how-to-calc...

http://www.phy.olemiss.edu/HEP/QuarkNet/time.html

t / 2 = 100 / (1 - (0.95 * c)^2 / c^2)^(1/2)

Where t is the time for a one-way trip and c is the speed of light.

That's not how Alcubierre-derived warp fields work. The time inside a warp bubble is predicted to remain synchronized with the time in the originating frame. Thus, if your warp bubble goes at .95c, it'll take you 105.26 years one way, 210.52 going there and back.

And I'm sceptical of your calculation, anyway. From Earth's unchanging perspective, if your are going at .95c in a conventional manner relative to Earth, Earthlings will see your trip as taking the same 210.52 years. Time dilation will only affect the time that passes for the people on board, which will make it seem shorter, not longer. So, really, if your warp drive can't break c, it's better from the passengers' perspective not to use one. This is explained in the StackExchange article you linked to- you put the factor of gamma in the wrong spot (http://physics.stackexchange.com/questions/31105/how-to-calc...).

I was assuming conventional propulsion at 95% c. An Alcubierre warp field would be ideal, assuming the problems can be overcome:

http://en.wikipedia.org/wiki/Alcubierre_drive#Difficulties

http://arxiv.org/abs/1001.4960

http://arxiv.org/abs/1202.5708

I'd enjoy seeing your calculation of the number of years that would have passed on Earth for a round-trip of 100 light-years out.

You've seen it. 210.52 years. The passengers on the ship should only experience 65.7 years- (210.52)sqrt(1-.95^2). Hence the famous twin paradox- the twin who goes on a relativistic trip comes back younger than the one who stayed behind. What you calculated was how much time would pass on Earth for a trip that took 100 years and 100 light-years from the perspective of the people on the ship, which is not generally what people mean when they say that a star is 100ly away.

I found the source of my confusion.

The post I replied to stated, "I don't know how much time would have passed on Earth though." You stated I calculated "how much time would pass on Earth."

The OP was wondering, "if a person on a ship visited a star 100ly away, how much time would have passed on Earth when they returned?"

However, the OP was wondering about time frames in terms of a warp drive, rather than a conventional almost-light-speed drive. Take my upvotes. :-)

I wonder how much of a perception you have of this trip though... if you're just radiation floating through space you might just was well think you just "instantly jumped" as soon as you arrive and "assemble" (whatever that process may consist of)

A lot sure will have changed. Imagine going back to the year 1370 and arriving today. Before Kepler, Galileo, before the discovery of America, or the printing press. Heck, a lot can go wrong in 640 years too.

We can visit up to a few tens of solar systems in a few hundred years, if we could develop a space-ship powered by a black hole, antimatter, or some advanced or innovative hybrid fusion engine (there were talks about one from the 60's around here).

Colonizing a few solar systems out of the 100 or so we can visit with such a ship that reaches a large fraction of the speed of light, should be "enough" for the next 1,000 years or so.

By year 3,000, we'll probably figure out something faster than that, to really begin exploring the whole galaxy.

Black holes naively look like a great way to get 100% mass-to-energy conversion, but when you really look into it, things aren't that great. To get more energy output, you want the black hole to be small (to have a high curvature in the event horizon); but, black holes that are small enough to give reasonable output are also too small (smaller than an atom) to reasonably feed new matter into.

Meanwhile, black holes big enough to feed fuel into have an incredibly terrible power-to-weight ratio.

And that's assuming that black holes even work the same way when they get that small. Quantum gravity could easily screw things up even more (or make them more convenient; who knows?)

1/5 seems to be the accepted number doing the rounds, but there must be other factors to consider too - I'm sure this number is still overoptimistic?

For example..how many of these habitable exoplanets also do the following earth things

1. Rotate around their axis as opposed to having one side facing the star.

2. have a stabilising moon to stop them from wobbling about.

3. Also how many of these exo solar sytems similarly have heavier elements and have evolved from spent star dust - using spent supernova explosions? To have enough water you need H and enough O

Some other important considerations that should perhaps be factored in?

The Earth went through a couple of "snowball-earth events" which resulted in a healthy % of oxygen being made available in the atmosphere.

We also have an outerlying and right-sized Jupiter to sweep away most of the asteroids and space debris.

Even with all the luck in the world, intelligent life only involved in the last million years of our 4 billion year plus Earth history... and we needed an asteroid, the right size again, to wipe out the dominant dinosaurs so that the evolutionary stage was set for our homo sapien species.

How much of this is repeatable?

We may indeed be the only intelligent life in our galaxy.

I'm curious as to what is next, now that Kepler is dead what do we do to substantiate this data further?

My impression is that there is not much more we can do to confirm outside of the statistical models giving us probabilities, it would take some much more advanced detection tech (or traveling there) to get further, which means we might be stuck at this point for a little while.

I guess pointing even more antennas to a star which is detected to have earth-like size and location (SETI like programs). With even larger telescopes - checking the light spectrum of the planet reflected light for chemical check of the planet (reflected light spectrum can show what the source is made from, to some extent).

We'd have to find them first.. Kepler was aiming at a distant part of the galaxy.

But let's say what we really hope for is true. What if we somehow discover that a large number of these planets are teaming with life ? What then ? How would that help us ? Spiritually, economically ? Wouldn't that make us feel even less significant ?

I'm just wondering, why are we hoping there is life out there and what would we do if we find it ?

Hopefully, we'd get a little attitude adjustment, question religion more and work together as a human race in the newfound awareness of our place in the universe. Perhaps also work towards actually surviving as a race for a while instead of selfishly exploiting every resource before others do.

It'll revolve around the Native American Experience, hopefully minus the conquest and genocide, more likely as a pattern than an anti-pattern.

In the long run, some high school sports teams controversially named after the aliens, some B-list movies, some academic interest, and that's about it?

"... and what would we do if we find it ?"

Grill it.

Even if it can talk to us.

In all sincerity, that's what I honestly think will happen.

Humans are the apex predators of Earth. Give us starships and we would be the apex predators of the galaxy in short order.

For tens to hundreds of thousands of years, there have been no creatures on Earth about which some human didn't think, "I wonder what that tastes like?" For Pete's sake, our ancestors ate giant, super-fast cave bears to extinction using little more than sharp sticks. Eskimos hunt huge multi-ton whales in the open ocean in flimsy canoes with hooked spears.

Sure alien life will be examined in labs and some will go to zoos. But, eventually, Monsanto would sell you Venusian steaks with a side of Martian scrapple. And, the most telling thing is, I would buy some and eat it. My mouth is already watering.

>Humans are the apex predators of Earth.

Probably most advance life would have evolved from predators. If you look at earth, the most intelligent animals are either predators or some sort of scavenger-predator hybrid. Which makes sense, predators need intelligence, cunning, sometimes pack awareness, to bring down prey. They need to be at the top or near the top of the food chain to devote some time to pursuits other than not being eaten.

Probably, nothing changes untill we start communicating with them, what may take a while. If we start feeling less significant, well, list that in the benefits... But I doubt it'll happen.

Rationaly, we should start looking hard for the explanation to Fermi's Paradox. Since it's not in our past, it must be in our future, what's very bad. But we are not rational, so we probably won't.

So we meet the aliens, we communicate with them, we become friends, they share a bunch of technological discoveries with us.. Then what ? We travel some more ?

If we can travel parsecs to meet them, wouldn't we be supremely technologicall advanced ?

I mean, the human race is pretty technologically advanced today, too advanced actually and it's hurting us as much as it's helping.

Will we some day realise that it is of no use to pursue higher mountains or more distant stars, because at the end we're still stuck within ourselves ?

What if the supremely advanced aliens realised this a long time ago and renounced technology all together, because it ended up destroying them ?

I wonder..

Renouncing technology you say... Also Medicine? If not so, overpopulation would soon become a problem and if by then they still don't know how to colonize further into space they face a serious problem.

Assuming this life is not techological civilizations that were hiding in plain sight, such a discovery would be bad news for us because it put us before the great filter.

cf http://www.technologyreview.com/article/409936/where-are-the...

I think it's more dangerous to seek alien life, spiritually and economically then it is to actually discover it.

Well, SETI isn't free, but it doesn't seem too expensive compared to some of the things we invest resources in. As for the spiritual danger of the seeking, I can't say it seems too fraught - finding, on the other hand, would at least require some adjustments to a few creation myths...

"Spiritually, economically?"

Community norms would be an interesting dynamic. Some groups of humans like to conform more than others. What does equal rights for beings with different numbers of tentacles mean to us? Humans think its hilarious good time to fight and kill each other over multinational corporations "owning" teams of grown men playing childrens ball games, so if you think humans aren't going to kill each other over "equal rights for unequal tentacles" from some alien culture, you're a little over optimistic.

Looking at politics, many humans behave as quislings and support groups that basically exist to ruin the groups the quislings are members of. What happens when the same type of quisling susceptible personality meets space aliens is so weird I can't even think of a sci fi story about it.

Also assuming light speed communication is more realistic than light speed transportation, something along the lines of internet trolling behavior is unfortunately highly likely. If you thought your mom browsing 4chan would be uncomfortable, wait till the space alien sociologists start reading /b/ and related. Or even worse, watching televangelist TV shows. Or reality TV. If they discover us by watching honey boo boo and decide to photon torpedo the whole planet, can we really blame them?

Does venture capitalism work across light speed delays? It would probably screw up exponential growth if the bubble has already popped before the IPO news even reached the other side. Can you do a bitcoin like protocol as a concept with multi year light speed delays between 3+ civilizations? What if latency is so long that it increases the likelihood of factoring attacks?

I would be most interested in learning about space alien UI fads. Our own fads are somewhat slow paced compared to faster paced ladies clothes/shoes fashions.

Humans think its hilarious good time to fight and kill each other over multinational corporations "owning" teams of grown men playing childrens ball games...

This is obviously tangential to your point, but it doesn't make you sound smarter or more evolved when you talk this way about sports. This sort of stuff is just tiring. As a software developer in one of the cities that sent a team to the World Series this year, I assure you I've heard every "sportsball" joke a dozen times and each time delivered as though the person saying it thinks they're the first person to realize some cultural blind-spot.

It doesn't advance your point, in other words.

In general, I agree, but I think it does serve a valid rhetorical point in this case- i.e., that is the way in which we are likely to view a lot of the weird, different, unexpected practices of an alien culture. And yet, there will be people who embrace those things in all seriousness.

I look forward to the discussion revolving around the concept of ten billion inhabited worlds colliding with the prevailing HN wisdom that the only place to do tech in the known universe is SV.

That's nice that you want to do tech, but if you stay on flyover Tatooine you'll never work on anything more modern than a moisture vaporator, better boot up the x-wing and fly to SV back on old Earth if you want to get paid to do stuff with jquery.

Yes, Tatooine under Pootine really sucks, doing even jquery in the Valley beats it by orders of magnitude.

This raises the obvious question .... Where is everybody?

It's like we were born on the Marie Celeste; a place that should clearly be brimming with intelligent life, yet appears (on the face of it) to be an empty desert.

The more we learn that the extent to which the galaxy is chock-a-block with habitable worlds, yet apparently devoid of intelligent life, the more I get a truly creeped-out feeling that we are living in some horror-movie galaxy where something is killing everything.

I sometimes think the answer is both boring and sad: everybody is just really, really, unimaginably, far away.

Even if we manage to continue for another 10,000 years without causing our own extinction, will we come up with anything capable of moving people even 10% of the speed of light? If we do, will we bother to go to the stars?

That. Why did I have to scroll down so far to find exactly what I was thinking all the time? I mean, I absolutely love Asimov and Mass Effect but they are entertainment stuff. The distances we are talking about here seem to be completely beyond everyone's mental reach. Even if we could upload ourselves into spacepods like Kurzweil suggests (or for that matter could just UDP our souls out into space) we would not break the speed of light - which makes interstellar travel possible but only in a very constraint range feasible. Plus the problem that during the four years it had taken me to get to Proxima Centauri, humanity on earth will probably have seen centuries pass...

I think we just have to modify our brains to think a little bit longer term than we do right now.

Perhaps life is common, but sentient space-fairing species aren't

This news reminded me of Carl Sagan's Cosmos episode about the Drake Equation (http://en.wikipedia.org/wiki/Drake_equation) that I watched as a kid. It is really an improved estimate for a couple of the terms.

That segment of Cosmos is online: http://www.youtube.com/watch?v=MlikCebQSlY

"Where is everybody?"

http://en.wikipedia.org/wiki/Fermi_paradox

re: my previous post - I realize that 1/5 is for "potentially habitable" ... but I was thinking too far ahead. What I was trying to get at is what factor would we then need to multiply it (ie., this 1/5th) with to make intelligent life a mathematical/chemical/biological certainty - as we know it here on earth, for example.

Was what happened on earth - us - a unique fluke - would intelligent life still come about if reran the whole thing 1000 times, even here on earth, where it was almost an after-thought?

Are there other pathways - sliding doors - that would still have brought about intelligent life here on earth?

What if we didn't have a magnetic field or if there was no Ozone? - that was another thing I left out before.

We may be a fluke - the end result of a set of rare and unrepeatable circumstances?

I'm only surprised that astronomers haven't concluded this long ago... Speaking of which, there has been a lot of these new astro sightings this year. Did they do something to the Kepler? I thought that spacetelescope was abandoned due to technical problems. What am I missing?

Confirming a Kepler planet requires observing multiple transits of the planet, so it takes several years if it's an Earth-like orbit. On top of that, the analysis and confirmation and peer review take time.

Incidentally, Kepler's been partially revived - they're going to point it somewhere else that doesn't require the reaction wheels to stay stable.

The data from these observatories can take some time to analyze. As for the spacecraft, they are currently investigating whether there is other science, or other methods to hunt for exoplanets, that can be performed despite the loss of half of the reaction wheels.

data mining, paper writing, and peer review are inherently slow. We'll be seeing new Kepler papers for quite a few more years, I think.

Reply on news.ycombinator.com