4 ms·
Forever ago. Science and Nature are in the top 10 rated impact factors for life science journals. Other than NEJM (top rated), the nature derivatives round ou
by mcarlise 10y ago
Forever ago. Science and Nature are in the top 10 rated impact factors for life science journals. Other than NEJM (top rated), the nature derivatives round out most of the top ten as well (nature genetics, nature reviews, etc...).
I think the problem stems from the inability for peer review to adequately test or reproduce papers before they are published. Which gives Nature and Science (which both make a large amount of money from their publishing practices) a lot of leeway to publish 'headline' based science. What bothers me, is that a lot of headline based papers paint a 'either or scenario' - or attempt to proove positives. When in reality, the scientific method can only disprove hypotheses. Only holistic second source papers (review articles) should make large claims; for instance, no single paper prooves the theory of evolution, but rather a large corpus of papers fail to disprove the theory.
An additional problem is accepting anything peer reviewed as valid because it was peer reviewed. Which is the driving force behind the peer review paywall - that peer reviewed artcles are somehow more valid then other forms of communication mediums. This implies that there exist somewhere, a publication format that is infalliable. And since retractions and fraud occur all the time in peer review, there are plenty of counter arguments to disproove this notion.
While this annoys me some, it will never be fixed. The number of 'scientists' that do not employ sound thinking and reason, will always out weigh the number of brilliant minds. There is a reason why almost all science can be traced back to five individuals (Copernicus, Galileo, Kepler, Newton, Einstein). These people reached a level of analytical thought that was unmatched. Surely we might be able to argue that they reached the low hanging fruits - but it was really their ability to not accept popular thought, and dismiss assumptions, and in most cases show a revered defiance towards establishments (Galileo especially) because they refused to not appeals to the fact (nullius in verba). So rather then spend a large amount of time to rid the world of bad publishing form - we should just accept it; and unfortunately most of us will never 'free our minds from the allegory of the cave'. And probably even more so for the groups that spend a large amount of advertising and headlines to convince others that they have reached this level of thought.
Wow - that comment got away from me fast; I really only intended to write the first two sentences, and just kept typing.
- EGreg 10y agoWhat would be better than peer review? Arxiv lets preprints be published. Would be cool if afterwards they'd have reviews from credentialed users.
- mcarlise 10y agoI agree with the reviews from credentialed users. People should put their names on the arguments for or against a paper, and let others evaluate them in return. I like pre-print services, and I wish they would go one step further and allow people to comment/critic the paper. This way, when you find a paper, you could see all associated thoughts and join the conversation. The really powerful idea behind pre-prints is that it would speed up academic publishing, by eliminating the ~6 month publication time. The only problem with pre-prints (currently) is they are not widely accepted for academic tenure. With academic publishing being a multi-billion dollar industry, I'm kind of surprised there are not more tech startups working on this issue.
- EGreg 10y agoSo why not make a service for arxiv papers or any other source of resources and have experts critique them? Maybe there is already a good source of experts on Quora or something. Or start a new site where experts are invited.
- mcarlise 10y ago> So why not make a service for arxiv papers or any other source of resources and have experts critique them? Yeah I think this would be a great idea. If you had the capture ability of sci-hub with the invite ability of quora to get experts to critic papers. Then add on-top of that some simple blog features like top highlight, sharing, subscribe to citation links, and comment/discussion (that matches HN's no counter productive comments/etc). The large obstacle would be paywalls - like sci-hub; any attractive paper would be illegal to have on the service; if you simply pulled from pre-prints (arxiv; plos pre-print, etc..) you will probably lose academic disciplines that do not value these services (biology primarily comes to mind). On the other hand, CS would probably embrace it as a complement to their conference publishing culture.