Hacker News Reader

Comment by giantg2 | Hacker News Reader

"Yet, people can be harmed indirectly. For example, research may — inadvertently — stigmatize individuals or human groups. It may be discriminatory, racist, sexist, ableist or homophobic."

If it's good research, then it shouldn't have any of these biases. Stigmatization and restrictions of rights is a policy issue. Science is not policy. I think we need to promote this separation more. Too often I see "but the study says this thing". Sure, that may be a scientific fact, but that doesn't mean it's the best thing for society, or even that it provides a complete picture of the issue.

View on HN

This infuriated me during the Covid pandemic. “The science says we need to close schools”. No, the science at most says that Covid will spread more slowly if we close schools. Whether that’s worth it or not is a political question, not a scientific one.

They are using the word science differently, to mean Science, as in "the group of recognized (by us) authorities on a subject".

They're including policy people in with the scientific people though, as "the science".

This. That editorial and its new policies might be justified, but it is definitely a watershed.

The old model was: scientists discover facts about the world, politicians/society decide what to do with these facts.

This new policy undermines the separation of descriptive (purely factual) and prescriptive, normative (infused with value judgements) statements.

Should physicists and engineers worry about whether a discovery they make could be used for new weapons?

> Should physicists and engineers worry about whether a discovery they make could be used for new weapons?

Szilard and Einstein thought it was worth worrying about.

https://en.wikipedia.org/wiki/Einstein–Szilard_letter

You are too generous. In practice, silencing science is not about good of the people, it is about censorship.

This is basically saying "Scientific research that supports my politics should be allowed, and those that refute it should be banned (because of course anyone who opposes my politics will be hurting people)"

I suspect that it's peoples perception of bias that will drive this, not actual measurable bias. Often people infer bias when they see an opinion they disagree with.

So how do we decide what to study, and who funds what studies? How will the results be communicated to people who lack scientific literacy? What will we do with the results and how will they be published? With or without a paywall? In studies that are impossible to double-blind, how do we make sure that the researchers and participants are not allowing their biases to get into the research?

You can do your best, but you can never completely remove bias from the scientific process

Reply on news.ycombinator.com