3 ms·

Plagiarism and fraud may happen in many schools but one would certainly expect more honest work out of a school costing 5-10x more than the competition. What ar

Plagiarism and fraud may happen in many schools but one would certainly expect more honest work out of a school costing 5-10x more than the competition. What are people paying for, if not a great reputation?

aprilthird2021 3y ago
Right, I'm saying perhaps there is a great deal of surprisingly innovative honest work coming out of such schools also. We only have a list of the negatives from OP. In every large organization there are both good and bad actors. If the good ones are 5x-10x better than other institutions' good ones, that changes our discussion here.
- t_mann 3y ago
  The thing is, results need to be genuine to constitute progress (fake results are arguably worse than none). That's why this undermining of trust in results is fatal. You shouldn't 'offset' fraudulent research with "they won a Nobel" (which Harvard actually did last year, if you count economics), because even the work of Nobel winners can be called into question later [0, also has a Harvard degree btw]. Every instance of academic fraud is problematic, and multiple independent cases in a year, across a wide spectrum of disciplines and departments at the same institution, is definitely not common, and we shouldn't start normalizing it I think. [0] https://forbetterscience.com/2020/10/07/gregg-semenza-real-nobel-prize-and-unreal-research-data/ https://forbetterscience.com/2020/10/07/gregg-semenza-real-n...
  - aprilthird2021 3y ago
    > That's why this undermining of trust in results is fatal. It's not though. Every one of the top schools in the US has some amount of cheating done by its students. The students still tend to have an outsized degree of success and innovation in the world. The US is a world leader in innovation, business, technology, etc. and it's also a country plagued by gang violence, homelessness, loneliness, isolation. All human systems are not like scientific theories that once you falsify one aspect of them, the whole entire thing is spoiled.
    - t_mann 3y ago
      > All human systems are not like scientific theories that once you falsify one aspect of them, the whole entire thing is spoiled. Virtually everything that counts as a research accomplishment is predicated on studies that are vetted through peer review. Imperfect as this process is, it is not capable of flagging research misconduct reliably. So most things we could list under accomplishments will likely require some amount of trust in the researchers. If you don't see how multiple independent examples of this trust being abused constitute a problem in this regard, then I suggest we leave it there. To be clear: the examples that I listed don't just relate to plagiarism, which would be misappropriation of results that are (hopefully) still true. Read the articles, there are many cases of suspected outright faking of results (often unlikely to be honest mistakes). I hope you never have a relative who has to rely on a cancer treatment based on fake research. I'm also not saying that Harvard is necessarily an outlier in this regard (unfortunately). But when it happens, I think we should call it what it is: bad, without buts.
- wakawaka28 3y ago
  How much other research is sabotaged by fraud from top universities? They are the "leaders" supposedly, and convinced everyone to be snobs about choice of school. So they disproportionately screw up everyone with fraud, potentially more than whatever discoveries they make. It can take decades and billions of dollars to undo bad research done at such a high level.
  - aprilthird2021 3y ago
    > How much other research is sabotaged by fraud from top universities? Academic fraud like the kind we are discussing occurred in many many universities across many countries. It was undetectable until recently when AIs were trained on almost the entire Internet and became very capable of showing similarities between papers from different places. > convinced everyone to be snobs about choice of school They did not do this. Many of these schools were not designed to be elite institutions. They reached this place in large part because of the success of the students who went through them. > So they disproportionately screw up everyone with fraud, potentially more than whatever discoveries they make My first point was, we don't know what way the scale tips if we only see the bad things and none of the good things. > It can take decades and billions of dollars to undo bad research done at such a high level. We already spend billions of dollars reproducing research for exactly this reason. Remember the unreproducible superconductivity research that had everyone excited here on HN a while ago? That wasn't able to be reproduced by many scientists. We spent definitely millions of dollars to debunk that one study. Given the scale of research in our world, we all as humans spend easily billions reproducing research already

aprilthird2021 3y ago

Right, I'm saying perhaps there is a great deal of surprisingly innovative honest work coming out of such schools also. We only have a list of the negatives from OP. In every large organization there are both good and bad actors. If the good ones are 5x-10x better than other institutions' good ones, that changes our discussion here.

t_mann 3y ago
The thing is, results need to be genuine to constitute progress (fake results are arguably worse than none). That's why this undermining of trust in results is fatal. You shouldn't 'offset' fraudulent research with "they won a Nobel" (which Harvard actually did last year, if you count economics), because even the work of Nobel winners can be called into question later [0, also has a Harvard degree btw]. Every instance of academic fraud is problematic, and multiple independent cases in a year, across a wide spectrum of disciplines and departments at the same institution, is definitely not common, and we shouldn't start normalizing it I think. [0] https://forbetterscience.com/2020/10/07/gregg-semenza-real-nobel-prize-and-unreal-research-data/ https://forbetterscience.com/2020/10/07/gregg-semenza-real-n...
- aprilthird2021 3y ago
  > That's why this undermining of trust in results is fatal. It's not though. Every one of the top schools in the US has some amount of cheating done by its students. The students still tend to have an outsized degree of success and innovation in the world. The US is a world leader in innovation, business, technology, etc. and it's also a country plagued by gang violence, homelessness, loneliness, isolation. All human systems are not like scientific theories that once you falsify one aspect of them, the whole entire thing is spoiled.
  - t_mann 3y ago
    > All human systems are not like scientific theories that once you falsify one aspect of them, the whole entire thing is spoiled. Virtually everything that counts as a research accomplishment is predicated on studies that are vetted through peer review. Imperfect as this process is, it is not capable of flagging research misconduct reliably. So most things we could list under accomplishments will likely require some amount of trust in the researchers. If you don't see how multiple independent examples of this trust being abused constitute a problem in this regard, then I suggest we leave it there. To be clear: the examples that I listed don't just relate to plagiarism, which would be misappropriation of results that are (hopefully) still true. Read the articles, there are many cases of suspected outright faking of results (often unlikely to be honest mistakes). I hope you never have a relative who has to rely on a cancer treatment based on fake research. I'm also not saying that Harvard is necessarily an outlier in this regard (unfortunately). But when it happens, I think we should call it what it is: bad, without buts.
wakawaka28 3y ago
How much other research is sabotaged by fraud from top universities? They are the "leaders" supposedly, and convinced everyone to be snobs about choice of school. So they disproportionately screw up everyone with fraud, potentially more than whatever discoveries they make. It can take decades and billions of dollars to undo bad research done at such a high level.
- aprilthird2021 3y ago
  > How much other research is sabotaged by fraud from top universities? Academic fraud like the kind we are discussing occurred in many many universities across many countries. It was undetectable until recently when AIs were trained on almost the entire Internet and became very capable of showing similarities between papers from different places. > convinced everyone to be snobs about choice of school They did not do this. Many of these schools were not designed to be elite institutions. They reached this place in large part because of the success of the students who went through them. > So they disproportionately screw up everyone with fraud, potentially more than whatever discoveries they make My first point was, we don't know what way the scale tips if we only see the bad things and none of the good things. > It can take decades and billions of dollars to undo bad research done at such a high level. We already spend billions of dollars reproducing research for exactly this reason. Remember the unreproducible superconductivity research that had everyone excited here on HN a while ago? That wasn't able to be reproduced by many scientists. We spent definitely millions of dollars to debunk that one study. Given the scale of research in our world, we all as humans spend easily billions reproducing research already