Hacker News Reader

Comment by cromwellian | Hacker News Reader

I love the humility, skepticism of belief Linde professes at the end. Despite a deep desire to believe in the beautiful, he is well aware of how you can be seduced into believing things because you want to, not because the universe is that way.

View on HN

There's a Ben Franklin quote: "Believe none of what you hear and half of what you see." Seems like Linde subscribes to that thought, as all scientifically-minded people do.

This is what separates the scientists from ... others.

We can only wish that is true, we are all just humans in the end:

http://www.economist.com/news/briefing/21588057-scientists-t...

Also, I vaguely remember a study in 'some journal' that looked at all it's P values[0] and found an alarming number of them parked right at the limit of acceptance (.05), more so than chance could assume. If anyone remembers this study and can provide actual proof (not my terrible memory), I would be very thankful.

[0]https://en.wikipedia.org/wiki/P-value

I don't think he's commenting on statistical analysis, but rather saying that a scientist should follow the clues where ever it may take them, instead of looking for the clues that will get them where they want to be. And of course they fuck up, too.

http://en.m.wikipedia.org/wiki/John_P._A._Ioannidis

"Most published findings are probably false."

I can't claim to understand much of the statistics, but according to [1], p-values are pretty sketchy as a measure of "is this likely to be true?"

[1] https://www.youtube.com/watch?v=ez4DgdurRPg

Or rather, the scientific method from. . . other ways of "knowing"

Reply on news.ycombinator.com