News for hackers. Stuff that matters.
Re:Telescope Sees Star Dance Around Supermassive Black Hole,... by mannykannot on Friday April 17, 2020 @10:06AM (#22899062)
A somewhat superficial acquaintance with the philosophy of science has led some people to think it is more sophisticated - or even more correct - to talk of theories as not having been falsified rather than having evidence in their support.

Having said that, I think the concern here is exaggerated.

Re:Telescope Sees Star Dance Around Supermassive Black Hole,... by doublesCs on Friday April 17, 2020 @11:41AM (#22899974)
Even more sophisticated:

Theories have:

- an expected domain of applicability

and might have:

- applications (either in the sense of experimental physics, or even applications in the sense of theoretical physics)

So a more important question is, do we have evidence against a theory in its domain of applicability in a magnitude that invalidates some use cases?

It's perfectly ok if you find evidence against a theory outside of its domain of applicability (i.e. I don't care if it's wrong in situations I didn't expect it to be right) or if that evidence causes errors so small as to make no difference for my applications.

Re:Telescope Sees Star Dance Around Supermassive Black Hole,... by m12k on Saturday April 18, 2020 @04:39AM (#22906241)
The point isn't really whether a theory is 'not falsified' or 'has evidence in its support' (they sound roughly equivalent, at least if you formulate it 'has _only_ evidence in its support'), the point is that nothing is ever _proven_, it is only ever not yet disproven. We're collecting an ever larger body of evidence that any theory's predictions need to match in order for the theory to not be proven wrong. Of course the more predictions a theory gets right, the less our need for another theory to make better predictions with instead.