Why Reinvent Science
Science is done in too few ways by people who think too much like each other. We plan to publish a series of short provocations suggesting new ways to do science and new ways to think about it.
Science is the act of creating new knowledge and human capabilities. However, a particular way of doing this has been enshrined in institutions with an incentive to resist change. The world is changing. Science must change too. This is an opportunity to do better than we ever have before.
So much ink has been spilled about the problems with science, but there is shockingly little writing about how we could do science better. Even very sophisticated conversations about science tend to be critiques of existing institutions like peer review or replication, not the act of creating new knowledge.
Like cold butter, it is hard for new ways of doing science to stick or spread. Science is a craft transmitted from master to apprentice. Practitioners are too busy scaling volcanoes, breeding fruit flies, and arguing at chalkboards to experiment with new ways of doing science. Institutions lock in ways of doing things and there are only a few types of institutions doing science – academia, corporations, national labs, and startups to some extent.
It’s a good time to try new things. The research ecosystem built after WWII is no longer fit for purpose: it’s now a hamster wheel in which hypercompetitive primary investigators spend hundreds of thousands of hours chasing decreasing funds doled out by risk-averse committees. The 80-year-old alignment between academia and the US government is fraying. It’s not yet clear how, but AI will change how we do things. The world is becoming a more chaotic place. Chaos creates many problems, but it also creates space for trying new things.
Our goal is to suggest some of those new things: small, concrete provocations that hint at how the actual practice of science could be done differently and possibly better.
Like real science, many of our ideas will be wrong! There are many wrong ways to do science, but there are also many good ways to do science. Just as we can identify good science in the past, posterity will know how well we did. Let’s unleash a cambrian explosion of approaches!