A Charlottesville nonprofit laser-focused on an ideal and armed with a free, open-source online resource for scientists is growing fast.
The main floor at the Center for Open Science’s long, narrow office in the Omni Business Center is lined with communal desks and standing work stations occupied by T-shirt-clad millennials. The office’s east end contains a large, well-stocked kitchen.
Founded in 2013, the organization — which added 14 employees in the first six months of 2016 — considers itself a tech startup, albeit with an unusual business model.
Rows of computer programmers, researchers and sales and outreach people are all working on the Open Science Framework, a web-based program that will give scientists tools to track their experiments step-by-step.
The problem COS is trying to solve, said manager Andrew Sallans, is simply that scientists across disciplines have a hard time showing their work.
“Fundamentally, we are trying to make science better and improve the quality of how science is done and the results from science,” he said.
That means increasing openness, transparency and reproducibility of experiments — notions that often are lost as scientists distill the findings of their research into journal articles that are limited in length.
Because findings don’t include a road map of how researchers drew their conclusions, anyone who wants to test the findings of a research paper for themselves has to work backward from the findings and more or less guess how the original researcher did their experiments.
“That is a lot of guessing, that is not how it is supposed to be,” Sallans said. “It is supposed to be prescriptive, like a recipe.”
In its first three years, the center created the Open Science Framework and completed the Reproducibility Project: Psychology. The study reopened 100 psychology research projects and asked new research teams to achieve the same results as the original studies.
Without road maps for the original studies and having to rely on guesswork, the project showed that only about 30 percent of the projects’ findings could be repeated.
The findings drew some backlash, but were enough to convince the center’s founders and funders of the validity of their work.
“We made all of the data from the repetitions open, so that anybody could come along and say, ‘Hey, this was wrong,’” Sallans said. “And a couple of people did, but the whole point was that they could not have done that if the steps were not available, and so it proved the point anyway.”
The point, Sallans was quick to add, was not to prove any researcher wrong.
“Our message and what we are trying to communicate in all of this is not that anybody was wrong, it was simply to prove that the detail they gave us was not enough,” he said.
With the problem defined, the next step is getting the science community to buy in, Sallans said.
Enter the Electrochemical Society, a group that supports research in electrochemical and solid-state science, which includes batteries. To promote the expansion of research in these fields and speed advancement, the society recently began an initiative called Free the Science.
The goals are similar to those of the Center for Open Science, and the Electrochemical Society hopes to use a modified version of the center’s Open Science Framework to power Free the Science, said Executive Director Roque Calvo.
“One of the things that makes them such an exciting opportunity for us is they represent something that hasn’t existed, which is a strong advocacy for open science,” Calvo said. “In our case, that is important because that is such an alignment with our mission and what we are trying to accomplish.”
Calvo said he hopes the Open Science Framework will be the platform the Electrochemical Society will use to spread information about its research and discoveries.
“That is the lynchpin, or the enabling piece, for us,” he said. “We are trying to find a platform where we can essentially create complete open access, complete dissemination of content. In the current environment, that platform costs us a lot of money.”
The center opened after Brian Nosek, a psychology professor at the University of Virginia, and then-Ph.D. student Jeffrey Spies published a paper outlining problems with transparency in science. The two studied inherent bias at the university.
The paper caught the eye of John Arnold, a hedge fund manager with a nascent foundation.
Arnold, whose foundation focuses in part on scientific integrity, offered $5.25 million to start the Center for Open Science and begin to tackle the problem Nosek and Spies had identified.
The COS employs about 80 people, including interns, Sallans said, and has an operating budget of nearly $5 million a year. The center’s 2016 funding includes more than $6.1 million from the Laura and John Arnold Foundation and more than $660,000 from the Alfred P. Sloan Foundation and the Institute of Museum and Library Services, the center’s website shows.
The center was nominated for the Charlottesville Business Innovation Council’s Top Job Creator award in 2015 and 2016, missing out on the award in 2015 to Apex Clean Energy and this year to WillowTree Apps.
