Reinventing Discovery_ The New Era of Networked Science - Michael Nielsen [96]
Indeed, even the possibilities that are being explored are not thriving as they should. While undertakings such as the Sloan Digital Sky Survey and the Human Genome Project are opening up their data to other scientists, the data from the great majority of scientific experiments remains closed. Scientists typically have little incentive to disclose their data, and so instead they hoard it. In the words of medical researchers Elizabeth Pisani and Carla AbouZahr, in science it’s “publish [papers] or perish,” not “publish [data] or perish.” And as long as that remains true, much of the world’s scientific knowledge will remain locked up, preventing the scientific data web from reaching its full potential.
Equally concerning are the disincentives for scientists to develop new online tools. While I was writing this book, a well-known physicist told me that Paul Ginsparg, the physicist who created the arXiv, had “wasted his talent” for physics by creating the arXiv, and that what Ginsparg was doing was like “garbage collecting”: it was good that someone was doing it, but beneath someone of Ginsparg’s abilities. Keep in mind that this astonishing narrow-mindedness was coming from a person who uses the arXiv every day. Ginsparg has perhaps done more for physics (not to mention the rest of humanity) than any other physicist of his generation. Yet sentiments such as these are often voiced privately by scientists. People who build tools such as the arXiv are dismissed as “mere” tool builders, as though it is somehow unworthy to be building tools that speed up the whole process of doing science. This lack of regard extends to the institutional level, where there is often little support for building new tools. Projects such as Galaxy Zoo and the arXiv often begin with little or no funding, in part because their first stage involves creating a tool, not writing a paper. How can ideas such as citizen science and the data web reach their potential in an environment where building new tools is held in such low regard?
The overall pattern, then, is that networked science is being strongly inhibited by a closed scientific culture that chiefly values contributions in the form of scientific papers. Knowledge shared in nonstandard media isn’t valued by scientists, regardless of its intrinsic scientific value, and so scientists are reluctant to work in such media. The potential of networked science—ideas such as the data web, citizen science, and collaboration markets—is thus remaining unrealized. To reach its full potential, networked science must be open science.
The irony in all this is that the value of openly sharing scientific information was deeply understood by the founders of modern science centuries ago. It was this understanding that led to the modern journal system, a system that is perhaps the most open system for the transmission of knowledge that could be built with seventeenth-century media. The adoption of that system was achieved by subsidizing scientists who published their discoveries in journals. But that same subsidy now inhibits the adoption of more effective technologies, because it continues to incentize scientists to share their work in conventional journals, while there is little or no incentive for them to use or develop modern tools. Indeed, when the scientists of today resist sharing their data and ideas, they are unconsciously echoing the behavior of Galileo, Newton, and company, with their secrecy and their anagrams. It may be a practical response to immediate personal concerns, but over the long run it’s the wrong way to do science.
To take full