Hacker News Reader

Comment by light_hue_1 | Hacker News Reader

This is a problem everywhere where the raw data isn't released (suitably anonymized).

In cognitive science, psychology, even computer science / ai / ml, business.

And the problem with rejected papers getting in somewhere else while being total garbage is pervasive. I've rejected a lot of papers because they were mathematically or statistically bogus only to see them get published elsewhere where reviewers were not so careful (a few times in Nature and Science).

We need an open science movement where you must release everything with your paper. The full pipeline to reconstruct every single result from the raw data. No hiding data. No hiding fmri scans. No "our code only runs on our machine". Etc.

View on HN

I'm empathetic to what you're suggesting — I've published on open science and meta-science specifically, and think open data should be the default norm. The problem with clinical research, though, is that it starts running into conflicting considerations about patient and participant privacy. Even when people aren't patients per se, the focus often involves sensitive information.

You can just say "anonymize it" but that turns out to be more difficult than it seems initially, especially with many questions of interest.

Also, there's often too many opportunities to do science that is of real public benefit that comes with privacy expectations attached for all kinds of reasons. Cases where there is legitimate consented access but an expectation of privacy without data sharing.

People have tried to solve this problem in different ways (for example, methods where someone can analyze data without having access to it directly) and maybe those solutions will lead to a good resolution. But they often have problems of their own (overhead costs associated with providing anonymized remote data analysis), and don't solve all problems (guarantees of absolutely restricted access to personal data).

We wasted decades, billions of dollars, and countless promising careers due to bad and fraudulent research in areas such as Alzheimers.

Whatever the costs and challenges are, they are not nearly as high as maintaining the status quo.

I think value-based care is the only real incentive on this. Otherwise, there is simply no reason for anyone to care enough. Even the insurers, they found a way to make money by making sure their premiums factor in these things. In the expense of the patient. As long as the drug doesn't kill people, who cares if it works if I make money off it as a pharma? Unfortunately, value-based care can only be pushed top-down. Patients are not in a position of power against pharma companies on this matter.

Reply on news.ycombinator.com