> Theory 1: Millions of academics across the world are united in a global conspiracy to do bad science You don't need a conspiracy to do bad science. Just as y

> Theory 1: Millions of academics across the world are united in a global conspiracy to do bad science You don't need a conspiracy to do bad science. Just as you don't need a conspiracy to do bad code - and we all on HN know how much of bad code is around. All you need is to write code - with the best of intentions - and to neglect to follow best practices or cut corners from time to time, or just lack experience or foresight to predict potential problems. Sometimes you don't even know it's bad code until you come back to it a couple of years later and stare at it in horror - how could anybody do this? With code, though, there's an evolutionary pressure acting on it - it is actually meant to be run. If the code does not perform, if it is buggy, if it's insecure or unmaintainable - it creates incentive to be replaced, so some of it dies off and hopefully is replaced with better one (or one bad in different way :). With scientific papers, bad paper can linger around for years until either it obviously contradicts reality in a way making its claims unsustainable, or is refuted by other paper. But for many papers just finding out it is wrong requires doing the same work as the original author did, sometimes more - and the incentives to do that are usually much less prominent. > The scientific process underpins every aspect of our modern era. True. But the fact that this process allowed us to achieve much does not mean every part of it is perfect - just as the fact that Facebook has billions does not mean every piece of code running Facebook is great. Some of them is surely terrible.