3 ms·

A few responses here highlight a very common problem in science: namely that people don't appreciate negative results as much as positive ones. We've generally

by hnhg 8y ago

A few responses here highlight a very common problem in science: namely that people don't appreciate negative results as much as positive ones. We've generally bemoaned (forgive me for not bothering to find examples but they're there) how difficult it is to get negative results published in scientific journals but we're exhibiting exactly the same attitude that leads to this in many comments here.

robertAngst 8y ago
This is entirely why I find engineering far more trustworthy than 'science' in 2018. Academia has a professor who wants to prove their idea is true. They rig up an experiement to prove their idea is true. They teach a kid how to collect the data. Then wow, they proved it true! Discovery? No. Engineering- prove your idea works and dont stop until the problem goes away.
- andybak 8y ago
  > Then wow, they proved it true! I don't know of any reputable scientists that would claim to "prove" anything. Popper rather put an end to that.
  - um_ya 8y ago
    It would be improper to say it doesn't revolve around certain outcomes though. I know a few guys in college that would get free money and travel to some remote part of the world to "research" climate change. It was pretty much a paid vacation for them. If you were studying how the climate followed cycles and it may have less to do with what humans are doing, you wouldn't get the funding.
    - andybak 8y ago
      I think there's a lot of bullshit in academia but it's not so much about the engineering vs science distinction being made here as it is about any situation where people can play the game. I think any large organisation will have it's fair share of bullshit and people playing the system. It's hard to align incentives with outcomes.
      - robertAngst 8y ago
        Producing unverified researching is different than releasing a product. One gets checked by customers, the other gets... reported?
KSteffensen 8y ago
The problem with negative results is that you as the publication reader don't know if the thing just plain didn't work or if the people trying to do it were incompetent.

by hnhg 8y ago

robertAngst 8y ago

This is entirely why I find engineering far more trustworthy than 'science' in 2018. Academia has a professor who wants to prove their idea is true. They rig up an experiement to prove their idea is true. They teach a kid how to collect the data. Then wow, they proved it true! Discovery? No. Engineering- prove your idea works and dont stop until the problem goes away.

andybak 8y ago
> Then wow, they proved it true! I don't know of any reputable scientists that would claim to "prove" anything. Popper rather put an end to that.
- um_ya 8y ago
  It would be improper to say it doesn't revolve around certain outcomes though. I know a few guys in college that would get free money and travel to some remote part of the world to "research" climate change. It was pretty much a paid vacation for them. If you were studying how the climate followed cycles and it may have less to do with what humans are doing, you wouldn't get the funding.
  - andybak 8y ago
    I think there's a lot of bullshit in academia but it's not so much about the engineering vs science distinction being made here as it is about any situation where people can play the game. I think any large organisation will have it's fair share of bullshit and people playing the system. It's hard to align incentives with outcomes.
    - robertAngst 8y ago
      Producing unverified researching is different than releasing a product. One gets checked by customers, the other gets... reported?

KSteffensen 8y ago

The problem with negative results is that you as the publication reader don't know if the thing just plain didn't work or if the people trying to do it were incompetent.