• Choose a category
• All
• Classic-Fiction

These problems are serious but not insurmountable. A system can be open without requiring that all participants receive equal attention. And you can share your knowledge openly, without having to pay attention to everyone (or, indeed, anyone) else. In general, for open collaborative systems to work most effectively, participants must have powerful ways of filtering information, so they can concentrate on the information of most interest to them, and ignore the rest. In the MathWorks competition, for example, recall how the score helps participants filter out unhelpful ideas, and focus on the best ideas from other users. And if low-quality contributions become more of a problem in the Polymath Project, it too could be filtered. Ideally, science is open-but-strongly-filtered. This is a natural consequence of the fact that while our attention doesn’t scale, sharing knowledge does. In an open-but-filtered world there is no problem with people such as Grothendieck pursuing their own solitary program.

Won’t open science sometimes be used for ends that many scientists find distasteful? In November of 2009, hackers broke into a computer system in one of the world’s leading centers for climate research, the Climate Research Unit at the University of East Anglia, in the UK. The hackers downloaded more than 1,000 email messages sent between climate scientists. They then leaked the emails (and many other documents) to bloggers and journalists. The incident received worldwide media attention, as many climate change skeptics seized upon the emails, claiming that they contained evidence to prove that the notion of human-caused climate change was a conspiracy among climate scientists. One of the examples used to support this claim was an email from Kevin Trenberth, a well-known climate scientist from the National Center for Atmosphere Research in Boulder, Colorado. In his email, Trenberth says; “The fact is that we can’t account for the lack of warming omputer syment and it is a travesty that we can’t.” In fact, the sentence was being quoted badly out of context. In the email, Trenberth was discussing a paper he’d recently published, which was looking at the causes of the year-to-year variation in the Earth’s surface temperature—why we have hotter and colder years—and how that variation relates to the long-term overall increase in temperature. The year-to-year variations are presumably due to changes in the way surface heat is redistributed into the ocean, into melting ice, and so on. Trenberth’s email and paper were pointing out that we don’t fully understand all the processes causing these variations, and so we can’t necessarily explain why any given year is hotter or colder. Although the email expressed some frustration at this state of affairs, it didn’t in any way contradict his belief in the long-run rise in temperature, which swamps the short-term variations. Note that the issue here is not about whether you agree with Trenberth about climate change. The issue is that a careful and honest skeptic of climate change could not possibly interpret Trenberth’s email as expressing any doubt on his part that humans are causing climate change. Nevertheless, many skeptics chose to quote the sentence out of context, either maliciously, to further their own ends, or carelessly, from genuine ignorance of the original intent.

This kind of incident illustrates a major risk facing climate scientists who are considering working more openly. On the one hand, open sharing of ideas and data has the potential to speed up discovery. On the other hand, every piece of information shared by climate scientists, no matter how innocent, stands a chance of being attacked by groups who want to bring climate science into disrepute by exaggerating minor problems, or by reporting remarks like Trenberth’s out of context. Given this, how openly should work on climate science be done? This is not an easy question to answer. If the issues were solely scientific, then the climate scientists should move quickly to work