4 ms·

In her TED talk [1] Tabby explains how comets could be an explanation for this phenomenon, she even says that it is consistent with their observations, but then

by soheil 10y ago

In her TED talk [1] Tabby explains how comets could be an explanation for this phenomenon, she even says that it is consistent with their observations, but then she goes on to say "but it 'feels' a little contrived". Sorry what did I miss here? I thought we were discussing science and coming up with theories that support our evidence using data, how does feeling fit into this picture? Why would it be so unlikely that a big comet swarm happens to be between us and Tabby's star, a distance of 1500 light years! Comets seem like a pretty good explanation, I haven't heard an argument against them. [1] https://youtu.be/IyIAjUuxNKw?t=7m17s https://youtu.be/IyIAjUuxNKw?t=7m17s

Retric 10y ago
Thousands of comets in just the right spot or tens of thousands of comets over a small area just for that star which seems contrived. If you can create a credible reason why there might be a swarm of comets like this say as remnants from a collision between a large body with just the right orbit* then that's just 1 event which is more plausible. *And I don't mean me saying it, actually create a model that would result in it based on you know math.
throw_away_777 10y ago
It probably is comets, but why the comets are in the position they are in is not well described. Science is done by people, and it is often very challenging or impossible to rigorously quantify exactly how unlikely something is. This is especially true in observational science. Furthermore, in communicating results to the public it isn't necessarily helpful to speak with quantitative rigor because the audience doesn't care. If someone says this has a 1 in a 10000 chance of being due to aliens, the headlines will scream "Scientists say alien life could be found!"
MichaelGG 10y ago
A lot of evaluating a theory means making subjective judgements on the probability of things if there's not perfect priors established. Like debugging: You end up with things that could have a few ways of happening. Nothing seems to line up perfectly. Different team members can make arguments on what the most likely culprit is. If something doesn't feel right, then you perhaps lean towards another hypothesis.
JonnieCache 10y ago
You don't come up with theories using data, you test them using data. You can generate hypotheses however you like, "feeling" is a pretty common way. Semi-random guesswork is another.
mcv 10y ago
I believe the comet swarm theory also rested on this being a fairly short as well as really massive and rare one-time event, which makes it rather unlikely (though obviously not impossible) that we just happen to be looking just as it happens. It also doesn't explain the long-term dimming of the star. The alien megastructure kinda does. It's not a perfect explanation either, but from what I understand, slightly more convincing than the giant comet swarm.

by soheil 10y ago

Retric 10y ago

Thousands of comets in just the right spot or tens of thousands of comets over a small area just for that star which seems contrived. If you can create a credible reason why there might be a swarm of comets like this say as remnants from a collision between a large body with just the right orbit* then that's just 1 event which is more plausible. *And I don't mean me saying it, actually create a model that would result in it based on you know math.

throw_away_777 10y ago

It probably is comets, but why the comets are in the position they are in is not well described. Science is done by people, and it is often very challenging or impossible to rigorously quantify exactly how unlikely something is. This is especially true in observational science. Furthermore, in communicating results to the public it isn't necessarily helpful to speak with quantitative rigor because the audience doesn't care. If someone says this has a 1 in a 10000 chance of being due to aliens, the headlines will scream "Scientists say alien life could be found!"

MichaelGG 10y ago

A lot of evaluating a theory means making subjective judgements on the probability of things if there's not perfect priors established. Like debugging: You end up with things that could have a few ways of happening. Nothing seems to line up perfectly. Different team members can make arguments on what the most likely culprit is. If something doesn't feel right, then you perhaps lean towards another hypothesis.

JonnieCache 10y ago

You don't come up with theories using data, you test them using data. You can generate hypotheses however you like, "feeling" is a pretty common way. Semi-random guesswork is another.

mcv 10y ago

I believe the comet swarm theory also rested on this being a fairly short as well as really massive and rare one-time event, which makes it rather unlikely (though obviously not impossible) that we just happen to be looking just as it happens. It also doesn't explain the long-term dimming of the star. The alien megastructure kinda does. It's not a perfect explanation either, but from what I understand, slightly more convincing than the giant comet swarm.