Follow-up of Kepler data yields more than 100 confirmed exoplanets | Hacker News Reader

Follow-up of Kepler data yields more than 100 confirmed exoplanets

news.ucsc.edu

54 pointsHooke16 comments

Can anyone explain what they mean by the word "confirm"? I looked at the paper to find out and ended up even more confused:

"HAT-P-56b (EPIC 202126852b) is a hot Jupiter confirmed by measuring the planet’s mass with Doppler spectroscopy (Huang et al. 2015). Our analysis indicates that the planetary hypothesis is the most probable explanation for the signal detected, with the next-most-likely scenario being an eclipsing binary (FPP=65%; see Table 9). However, the radial velocity measurements of Huang et al. (2015) rule out the eclipsing binary scenario favored by vespa and so confirm the planetary nature of this system." http://www.lpl.arizona.edu/~ianc/docs/crossfield_K2s_new_pla...

So this object was previously "confirmed", yet they are still considering that it may not be a planet at all. Then they say some other measurements do "confirm" it is a planet...

Either an object is confirmed to be a planet or not (presumably according to some gold standard measurement method). I don't understand using that word when they may waffle back and forth as to whether the object is a planet.

There is no gold standard for confirming the existence of an exoplanet. Very different techniques are used by different groups based on the type and source of experimental data. Every method has its strengths and weaknesses, therefore, often planets detected by one method are then reconfirmed using another method applied to the same or new data. Here is a quote from Huang et a. (2015) https://arxiv.org/abs/1506.01776

"This planet was originally identified as a HATNet (Bakos et al. 2004) planetary candidate, was followed up by the TRES spectrograph on the FLWO 1.5 m telescope, and also by the KeplerCam imager on the FLWO 1.2 m telescope. Encouraged by these initial results, all pointing toward a bona fide planet orbiting the host star HAT-P-56, the target was proposed for K2 observations. Indeed, the very high quality photometric observations of K2 confirmed the transit, and also eliminated most of the possible blend scenarios."

Then it sounds like these are not really 100 "confirmed" planets (for the usual usage of confirmed), instead they are 100 "likely" planets.

Yes. There is no notion of confirmation in physics (or science). All you can do is perform different experiments till all explanations except one have been falsified. The one has been "confirmed" is the lazy way of saying it. Somebody could think of new theory tomorrow that explains the data just as well as the theory that planets exist, and then the planets will be "deconfirmed" till further experiments falsify one of the theories.

Personally I wouldn't use "confirmed" or "proved" to discuss any hypothesis. However, I think confirmed could be acceptable in some cases. Usually it would be some kind of cheap screening test followed up by a more expensive (but more reliable) gold standard test. Sure, the gold standard may still be wrong, but at least everyone knows what you mean by confirmed. I find the use of confirmed here to be highly misleading.

Here's a good short 10 minute documentary on this: https://vimeo.com/174313049

Note: the bulk of this video is a scientific funding pitch for "Mission Centaur", which is a small satellite proposal to look for planets around just Alpha Centauri A and B. Nothing wrong with that, but this doesn't attempt to be a dispassionate review of Kepler.

> Mission Centaur is a Silicon Valley-based nonprofit corporation dedicated to studying Alpha Centauri. We support investigations that advance the state of the art knowledge and technology for exoplanet exploration, the ultimate goal being the discovery and characterization of Earth-like planets around Alpha Centauri. We envision a new way of engaging with science where experts, institutions and the general public can be part of humankind’s most ambitious and transformational quest: the search for life beyond our solar system.

http://missioncentaur.org/about/

I recently went to a talk where the researcher was able to derive the location and size of starspots (a.k.a sunspots here on the sun) from the Kepler data [1]. It just amazes me the amount of information you can derive from a single pixel of light dimming in and out.

[1] https://www.youtube.com/watch?v=Gf7jsYA6oJ8

Do you think mankind will ever reach one of them? Robots or humans? Interstellar ark or von Neumann machines?

Or will these planets be simulated in VR?

What kinds of science and technology advancements will be needed to make a Starship Enterprise a reality? The first step is getting more humans living in space or on planetary colonies.

We're already living in VR. Each of us are just threads running on the Universe VM. All we need to do is find a use after free bug in reality, and we'll be able to go wherever we want.

Using what humans have learnt from developing software as an analogy for how the universe operates is fine until you start expecting human-type exploits to also be present in the rules of physics.

At that point you're basically hammering a square peg into a round hole that may not exist, expecting the analogy to hold true.

If there are bugs in the universal algorithm, they'll not be the kind of bugs you find in C, but the kind you find in the ideal programming language. So research into better programming languages is crucial to human transcendence!

I think humans will eventually visit a few planets and maybe live on the Moon or Mars. To go further will require either radical bioengineering or AI / brain upload. (The distinction gets blurry there.)

We are just too fragile and we don't live long enough for interstellar trips. Even the outer solar system may be prohibitively hard for us.

I guess we could upload our personalities to a computer then fly around in space for thousands of years. Then grow a body on whatever planet we feel like.

Even if we don't, the observation that all these potentially-life-hosting worlds are (or aren't) dead would tell us something important.

Reply on news.ycombinator.com