There is not enough data in a vague orbit and vague size (in a dataset that until recently contained a spurious planet) to feed a simulation like this, which means that the simulation was, mathematically necessarily (and I do not use this term lightly), made up from nearly whole cloth. And then we have no way to verify anything that it did say, assuming it even started from the correct place, especially with regards to stability on geologic time. Even an 99.9999999%-per-hour perfect simulation will still diverge on geologic time scales. For all we know, an atmosphere with a mountain range here will freeze out in a century, whereas a mountain range there could in fact result in a stable atmosphere over geologic time. This is all information-theoretic necessity. This is entertainment, not science.
That may be all the scientists claimed. It's still a fun result, even if I wouldn't put my grandchildren on a flight to Gliese on the strength of it. I'm reacting against the certainty expressed in the article.
Ok, I think I see what you mean now, and I agree with you. They've shown that the distribution assuming some set of priors includes some particular, interesting outcome. This is the first planet we've ever seen where we can say that given how much information we have about it, though, which means that it's still cool.
Yeah, we can still look at some extremely broad possibilities how how such a planet atmosphere could behave based on the little we know about the planet. Then the simulation gives some guide onto how each of these would play out.
Sure it's very speculative, but one step up from no data at all, which is the best we have now. At least have an idea of the size, the orbit and the sunlight intensity that would be hitting the planet.
That may be all the scientists claimed. It's still a fun result, even if I wouldn't put my grandchildren on a flight to Gliese on the strength of it. I'm reacting against the certainty expressed in the article.