How we aim to accelerate scientific progress

We believe a more open collaborative model of scientific research can accelerate progress by an order of magnitude. There are just so many ways in which the current academic reward system produces behavior that is not aligned with the best interests of science and society. Here’s how:

The exchange of ideas is accelerated — Fundamentally, science progresses according to the speed at which scientists share ideas which then allows other scientists to learn from, be inspired by, and build upon those ideas. The current journal-based model of sharing ideas is a relic from the 17th century. A much faster exchange of ideas is possible — Open Discovery brings science into the Internet era.

More scientific insight is captured and organized — Within the current system, a year’s worth of work is often boiled down to a single paper — often with important details intentionally withheld. Open Discovery encourages a highly collaborative model where everything is openly shared from the beginning of a project. The resulting discussion is neatly organized and searchable — ready for future researchers to discover and benefit from.

A marketplace of expertise emerges — A typical research project has many small components that would benefit from the attention of experts in different areas. The research team will never know everything themselves, and there are always small questions that come up along the way. Open Discovery intends to create a marketplace whereby scientists external to a project are financially (and reputationally) rewarded for contributing. This can be thought of as a marketplace for scientific attention — attention naturally moves to where it has the most impact.

A meritocratic system opens the door for more participation — An important aspect of the marketplace for expertise is that anyone can participate. It doesn’t matter if you’re in industry but have some unique knowledge, or if you’re a young researcher in a developing country. All that matters is you have things to say that provide value to the scientific community.

Data analysis is verified by experts - Many modern research projects require sophisticated use of statistics in order to correctly interpret the results. This data analysis ought to be performed or verified by external experts to reduce mistakes and unfounded conclusions.

Interdisciplinary conversation produces serendipity & new insights — Academic science is often carried out in artificial silos. Open Discovery is an open platform where people from different fields can easily find themselves in conversation. We believe it’s precisely this interdisciplinary conversation that is likely to lead to the greatest breakthroughs.

Research is not distorted by the need for ‘interesting’ results — Currently, academic scientists are rewarded almost entirely based on their publication record. This creates incredible pressure to get “interesting” results worthy of publication. This pressure distorts the research process in both conscious and unconscious ways and plays out on both an individual and institutional level.

Publication bias is eliminated — At present it’s hard to trust research results because you never know if there were 10 studies that asked the same question, got a null result, and were therefore never published. This problem is eliminated by having research projects published openly from the beginning.

“Risky” research becomes viable — Within the current system, research projects only get started if there’s a high likelihood of being able to publish a paper — or better yet, many papers. This dynamic means high-risk but high-reward projects don’t get proposed and don’t get funded. A system where researchers are rewarded based on how well they conduct their research — not what the results happen to be — solves this problem.

Redundant or fundamentally flawed studies are nipped in the bud — In today’s system, fundamental research flaws are often not discovered until after the research team has conducted the work and goes to publish a paper. An incredible amount of time and resources could be saved if those proposals had been shared publicly and the scientific community had vetted them before work began.

Needless, repetitive work is drastically reduced — Most scientists don’t share their data, software, and methods because they consider it a competitive advantage. A consequence of this is that everyone is constantly reinventing the wheel and duplicating each other’s labor. The solution is to compel or incentivize the adoption of new norms around sharing these key research outputs.

Research becomes more impactful — The impact of a research project can be measured by how effectively future scientists are able to build upon the results. In order to ensure maximum impact, it makes sense to get external scientists (the ones who will be building upon the work later) to validate from the beginning and throughout the project that the research team is doing things in a way that will produce results they can trust and build upon.

Research becomes more reproducible — So much of today’s research is not reproducible. Sharing all the data, software, and detailed methods would go a long way to solving the issue. So would having a trail of documentation from soliciting feedback from external scientists throughout the project. Removal of publication bias and the perverse incentive to get “interesting” results would each also help the issue.

Peer review becomes a productive contribution to science — Today’s peer review system is incredibly slow, inefficient, and flawed. It happens behind closed doors — often creating opportunities to sabotage the work of competitors. Reviewers often don’t have the expertise, the time, or the motivation to get it right. The solution is to make peer review public — to make it a legitimate contribution to science that is rewarded financially and reputationally like any other.