Hacker News Reader

Hacker News

Comment by sega_sai | Hacker News Reader

Full thread

sega_sai

"In the end the Party would announce that two and two made five, and you would have to believe it."

View on HN

It wasn't the party that made the replication crisis.

Right now universities have all the trappings of science with none of the substance.

Do I particularly care than they are going to be bent to the ends of those in power?

Kind of.

The same way I'm a bit upset if someone pisses in the holy water at a church.

Universities have produced medical breakthrough after medical breakthrough.

Peer review isn't perfect, but it has gotten us incredibly far, and it's way better than political appointees who don't believe that AIDS is caused by HIV making decisions based on culture war considerations.

Clinical trails have produced medical breakthroughs, peer review gave us leeches.

Clinical trials don’t happen until there’s a treatment to try.

Like leeches.

False dichotomy

How do you think medical treatments make it to the stage of doing clinical trials?

The basic research that leads to these treatments is all selected and evaluated using peer review. Even the results of the clinical trials are analyzed using peer review.

You're happy to send us back to the Middle Ages, when people actually did think leeches were the solution to everything, because you've got some weird chip on your shoulder.

It also wasnt The Party who identified the replication crisis, and started the work of fixing it.

Science is a self correcting mechanism, monarchism is not.

The crisis has been ongoing since the 80s and isn't slowing down.

If the cure for it is to wait till the researchers die so new better ones take their place you just reinvented feudalism with tenure.

Maybe because the crisis isn't a crisis. It's a 1) examining tough things with a lot of noise and low funding prevents adequate sampling

You should care at least as much as if someone pisses in the cup you're drinking from, because regardless of the merit of the 'science' being produced in certain fields, the findings of bad studies are nonetheless breathlessly reported as objectively true Science, becoming "facts" that influence who your fellow less-discerning citizens vote for and influence how those voted in approach policy decisions. The state of scientific research is bad enough, but it can become much worse when actively applied to propagandistic ends (moreso than it already is).

Scientists have been pissing in all our cups for longer than I've been alive. That you're arguing what concentration of urea is acceptable is part of the problem.

There comes a point at which an institution is unsalvageable and the only solution is to burn it down.

Universities are far past that point.

yes yes the solution to some debatably valid social science research is to burn the entire epistemological method and practice to the ground. We'll surely find something better to replace it in its absence.

"Burn it down" is not on the menu. Your options are "bad science, used to mildly propagandistic ends, mostly in the form of lobbying public policy via bought-and-paid-for studies" or "bad science, used to extremely propagandistic ends directly by the state". Even if the option you would prefer is "good science" or "no bad science", you live in the real world and are currently being presented with a society making the choice between the actual options on the table rather than fantasy options.

What we have right now is, "Mostly good science, which produces a constant stream of major advances, but which is subject to the same pressures and failure modes as any human endeavor, such as bias and financial interests with ulterior motives."

If you want to improve that system, relying on people like RFK Jr. (a crackpot who rejects basically all of modern medical science, right down to the germ theory of disease) and Trump (the most corrupt president in US history) is crazy.

The Party certainly isn't going to solve the replication crisis. It's just going to pressure researchers to publish a different set of unreplicable results.

Reply on news.ycombinator.com