I swear, every time a new physics discrepancy is measured, everyone jumps up yelling that it's definitely not real, invites others to openly sneer at it, compares it to the infamous faster-than-light neutrinos the OPERA experiment measured, and finally points to some alternate nutty theory that's definitely right, will be proven in 10 years time, mark my words.

The biggest damage OPERA inflicted wasn't the loss of time and effort to explain it, nor the resulting confusion, but rather the idea that EVERY SINGLE EXPERIMENT IS LIKELY WRONG.
Because obviously, someone screwed up and thus everyone is also screwing up.
Which is as far away from the truth as possible. Most physics experiments are boring and just confirm exactly what we expect.

It's precisely those hints that there's something wrong with our ideas about the universe that need to be explored thoroughly, and no amount of time and effort spent doing that is wasted.

· · Web · 2 · 0 · 1

@lynne well the good thing about physics, AFAIK, is that nobody will settle on a single experiment. If they notice something wrong, they'll try to reproduce the bug in a more issue-focused environment, or come up with new experiments that should also confirm the existence of that bug, and learn how to exploit it, right?

But from what what I've heard other sciences aren't in such a good shape, and have a replication problem or sth...

I get what you mean, but it's also dangerous to jump to the other possible conclusion, that it's definitely proof of a new discovery. From what I gather, when doing science, patience is key. And patience dictates you've got to make sure you exclude everything you can think of that could have interfered with the experiment to bias the result, and then you also have to start looking for things you didn't think of that could have screwed things up. Not to mention that you've got to repeat the experiment, and have others repeat the experiment, countless times, to finally also exclude statistical flukes.

@scarlet If results are published, everything is already well thought over and accounted for. Before construction even begins, during construction, during calibration, during the actual data gathering and after all the data gathering, at least twice, by independent parties.
By plainly declaring "it must be wrong!" and discouraging further review, it's doing everyone involved a disfavor.

>If results are published, everything is already well thought over and accounted for.
Not really. You publish so that other people start looking over thing, and try to find potential mistakes that you still missed, after all the verification you did yourself.

I understand your frustration regarding "it must be wrong", but when you come up with results that contradict models that have withstood a LOT more tests than just your new experiment, the fact is that you're trying to upturn decades of evidence that these models work. The preponderance of evidence just isn't in your favor.
Sign in to participate in the conversation

A Mastodon instance for people interested in multimedia, codecs, assembly, SIMD, and the occasional weeb stuff.