3 ms·
It honestly makes me really nervous for the future. There is a LOT of distrust in experts that I see across the internet (and here in HN). While there are bad
by cogman10 5y ago
It honestly makes me really nervous for the future. There is a LOT of distrust in experts that I see across the internet (and here in HN).
While there are bad actors, consensus is generally going to guide you closer to the truth than fringe experts. What does the community of experts who have studied this say about it? Is it something generally accepted by the experts or is it something that most experts disagree with?
It is nearly impossible that an entire field of study will succumb to bias. Climate change is the poster child here. Where experts disagree isn't that it is happening and caused by man, they disagree about the degree that it will be a problem (With all of them agreeing it's a major issue). Yet, I've often seen it claimed that those experts are all biased and this fringe expert funded by Koch research is really the one to be trusted (which, shocker, finds we should be burning MORE fossil fuels!)
- ErikVandeWater 5y agoTrust in scientists is too high when they keep pumping out studies that don't replicate, and don't publish the original data and methods used to reach conclusions. I wholly trust science that independently replicates and the fact that it's missing from the process means much of what we see is less than wholly trustworthy.
- cogman10 5y agoNot all fields of science are created equal. Whether or not studies are replicated or data is available will depend entirely on the field of study. Certainly, a lot of the social sciences and "soft" sciences have a fair bit of replication/data problems. That said, the biomedical, environmental, geologic, and chemical sciences by and large have really solid data and repeats of that data. The areas of sciences we have a harder time controlling for are those where it'd be unethical to get better data and examples (nutritional science, for example). There are far too many fields of study to lump all science into the same bucket and call it all untrustworthy. A bigger problem is that scientific journalism treats every study as being equal. When, in fact, some studies are far better than others and carry a much larger weight of truth.
- fallingknife 5y ago> Not all fields of science are created equal. But FB, and other platforms, treat them as if they are when making their decisions on what to censor. FB even trusts so called "independent fact checkers," who don't even claim to be scientists and write mostly on political topics, to act as arbiters of truth. > A bigger problem is that scientific journalism treats every study as being equal. And it's these journalists that FB is relying on for its determination of fact, not the actual scientists.
- AlgorithmicTime 5y ago
- fallingknife 5y agoThen stop censoring. Look at what happened in this post. Experts said X, the author said Y, and Y was censored because it disagreed with the experts. Later the experts admitted that X was incorrect. Of course, the experts, who ended up being wrong faced no consequences, while the author who ended up being right was censored. The experts are essentially a privileged class whose opinions are considered to be true by default, and when they get things wrong, it is assumed that this is just a good faith mistake, and forgiven. Everybody else is assumed to be wrong by default, and spreading politically motivated misinformation. When they are right it is assumed to be a fluke and swept under the rug. In this environment the only logical response is to mistrust the appointed "experts." If they are what they say they are let them defend their point against all challengers.
- nescioquid 5y ago> Then stop censoring. And stop dissembling. For instance, Fauci's early statements on mask-wearing were calculated to preserve the supply of N95 masks for medical personnel. However laudable the intent, it betrays what I imagine is an endemic world-view that the masses must be manipulated, rather than dealt with honestly. This morning, an NPR report about the new CDC quarantine guidelines acknowledged criticisms/allegations of corporate lobbying for the policy change and internal fears of staffing for critical services, which was followed-up by a "follow the science" platitude.
- sutuplesu 5y agoFacebook's actions keep being cast as if there is a completely open, level, neutral field that the censorious fact checkers are now distorting. But that's not the starting point. Facebook already has its thumb -- if not both hands and an elbow! -- on the scale, as far as what content is dispersed. The view is distorted to begin with, by Facebook choosing what is promoted and what is not. They may not be doing a particularly good job un-distorting it, but it's not a choice between pure and free exchange of information and censorship. It is between whether Facebook is going to manipulate in both directions -- promoting and taking down -- or just one. (Note that I don't think either of these choices are particularly good. I would prefer that Facebook did neither.)
- nradov 5y agoExpert consensus has some value, but it's often too subject to inertia and fails to adjust fast enough to new evidence. Up until 1982 the medical experts told us that peptic ulcers were caused by stress and spicy food. Except that was totally wrong. https://www.news-medical.net/health/Peptic-Ulcer-History.aspx https://www.news-medical.net/health/Peptic-Ulcer-History.asp... What else could they be wrong about today? Often when you really dig into consensus statements you find that they're based on a handful of old studies which don't really meet modern evidence based medicine criteria.
- cogman10 5y agoThe problem I have with statements like these is they focus on the misses while ignoring hits. Expert consensus says that washing hands prevents the spread of disease. It's been like that since the 1950s. Has that consensus been wrong? What about the expert consensus that azithromycin is an effective treatment against (most strains of) streptococcus? Is that consensus wrong? You are correct, expert consensus will be wrong at times and it will take a fairly large push and evidence to change course. However, for every example of the experts "getting it wrong" there are 100s if not 1000s where you are better off following the consensus. Further, as time goes on, the process of developing consensus improves itself. It looks for faults that lead to bad conclusions and corrects for those faults in future studies. It is a truth that experts and their consensuses will be wrong. However, if I follow and accept what a strong expert consensus says about something in that field of study 99% or even 99.99% of the time, they will be right. Or said another way, if you make the claim that "This expert consensus is wrong" You'll be right 0.01% of the time and wrong 99.99% of the time. Unless you are an expert in that field of study and you have a bunch of very strong research backing the claim, you've got no footing to really go against the consensus.
- tiahura 5y ago“Expert consensus says that washing hands prevents the spread of disease. It's been like that since the 1950s. Has that consensus been wrong?” WRT Covid, yes that advice was wrong and has been abandoned as Covid isn’t spread by surface transmission.
- 5y ago
- autokad 5y agoI worked first hand in a research group. the 'experts' are extremely political, and anything that doesn't fit their political agenda is ignored, and if the study/experiment doesn't fit their political agenda they keep trying or fudging the data until it does. I'll give you an example, I noticed a map of obesity overlaps very well with a map of diabetes. high alcohol consumption mapped very well with the areas the prior map didnt correlate. in other words, in areas where obesity was not correlated with diabetes, there was high alcohol consumption. paper worthy or worth looking into? no. I lied to a different section of the group and said these were areas of increased correlation, and it wasn't alcohol but soda consumption. They immediately wanted to seek grants to write papers. You can see how this leads to junk science. if you ignore things that dont fit your political agenda and only research things that fit your agenda, that's junk science. it gets worse because when the data doesnt suggest the political agenda pushed, they try to 'account' for other factors or re-run the study in a different way until it says what they wanted it to say. worse case, it doesnt get published. The experts are not to be trusted.
- colejohnson66 5y agoIs that related to the “reproduction crisis” science is going through right now?
- autokad 5y agoyes because results are jelly beaned. If you keep changing what to measure for success criteria until you get what you want, you are more likely reject the null hypothesis even though you should not have OR if you dont publish results that doesnt fit your political bias, the 1/20 ones that do get published and also incorrectly rejected the null hypothesis
- umvi 5y ago> It is nearly impossible that an entire field of study will succumb to bias It's happened before, especially when a select few command the majority of the field's grants/publications. If you are interested in how this could happen, here's an example of where basically the entirety of the Alzheimer's disease research field of study succumbed to bias: https://www.statnews.com/2019/06/25/alzheimers-cabal-thwarted-progress-toward-cure/ https://www.statnews.com/2019/06/25/alzheimers-cabal-thwarte... as well as the HN discussion on it: https://news.ycombinator.com/item?id=21911225 https://news.ycombinator.com/item?id=21911225
- quinnjh 5y agoExperts have spent more time tuning their biases. If someone claims to be free of bias then they dont understand training sets and inheritance.