3 ms·
Show HN: usefulscience.org – one sentence summaries of journal articles
- jaan 13y agoLooking for feedback, and we're happy to answer any questions! RSS: http://usefulscience.org/rss.xml http://usefulscience.org/rss.xml
- bensonperry 13y agogreat design! how are the headlines/summaries written?
- jaan 13y agoThanks! We're a team of grad students from various disciplines – we write the summaries ourselves for now. Hopefully some text mining for automating this in the future!
- illdave 13y agoReally really love the design - great work. Could be useful to add a search function (it'd be interesting to headlines that mention "coffee", for example).
- jaan 13y agoThanks Dave, appreciate it (design was by Cameron Spencer @ lab43.com). And search is on our feature list – hopefully it'll be up soon!
- bhudman 13y agoThis is fantastic. It is annoying to come across papers that are incredibly hard to read, but convey very little.
- jezclaremurugan 13y agoAny place where I could sign up to get a digest will be great. Good job on the website.
- jaan 13y agoThank you! You can sign up for the email digest at http://usefulscience.org/about http://usefulscience.org/about
- 11001 13y agoI really don't want to diss your work, I respect your intentions to bring scientific discoveries to the masses, but... This is part of a huge problem in science, the "one sentence summaries", or simple interpretations. It might be tempting to think that knowing a one sentence summary is better than not hearing about the research at all, but it's not so obvious to me. Most (social science, nutrition, environmental science, economics, psychology etc.) research is deeply flawed. That includes study design, statistical analysis and interpretations. Politicians love to through money at scientists that look for simple straightforward causal relationships between such incredibly complex systems as human behaviour, brain chemistry, nutrition, genetics, etc. And that's why there is so little reproducible research. You won't get a right or wrong answer about whether, say, cannabis is harmful without taking into account, a person's genetics, daily stress, the use of coffee, alcohol and other substances, their diet, their level of physical activity, social life, childhood trauma etc. But that would be too hard to interpret. Instead we get headlines like "Teenagers smoking pot are (arbitrary number) more likely to develop schizophrenia" and "Cannabis has been shown to kill cancer cells", which leads absolutely nowhere, because they just delude people to thinking the scientists are answering the original question ("is cannabis harmful?"), which isn't well defined in the first place. The necessary part of improving the current state of social "sciences" is acknowledging that simple rules rarely govern complex systems, and therefore the public should be incredibly skeptical to any headlines that report that "the scientists proved" some straightforward "law". Source: I do data analysis for multiple research groups that include psychology research (hormones, behaviour, depression), nursing research (breastfeeding, childbirth complication) and psycho-oncology research (quality of life of cancer patients and their caregivers).
- jaan 13y agoThanks – I agree with your worries about misinterpretation, especially in regard to flawed study design. We're not trying to be the be-all end-all, but hopefully a decent starting point for further research into the subtleties of a specific topic. That's in addition to trying to make the science more accessible by having it appeal to a wider audience. We're still narrowing down our content guidelines, so would love your input! Feel free to ping me.
- perrincita 13y agoGreat idea! This is definitely going to be useful in class! Glad to find a place where I can easily find interesting research papers. Keep up the good work!
- lcary 13y agoA recent study showed that daily doses of usefulscience.org will enlarge the frontal cortex by 15%.