Open Labs

The renewed excitement about improving science offers plenty of "fixes" for problems, but it's worth approaching it from the opposite perspective: Can we make the good parts of science even better?

Graduate school experiences vary widely. I've heard nightmarish stories that derailed careers, and many others that cherished the time. At its best, the relationship between graduate students and advisors can be a positive and generative source of intellectual exchange. Those discussions can turn into career-making ideas and experiments. This isn't always the case, of course, but when it's good, it's a wonderful thing. 

I’d like to test an idea that amplifies the best aspects of scientific mentorship: Open Labs.

Open Labs would be a place for experienced scientists to post their open questions and curiosities — a list of the ideas they don't have time to pursue personally but wish someone would. Instead of these questions being confined to the walls of their lab and personal inboxes, they are open to the world. In the test I'm proposing, we'll host the page on Experiment so anyone can submit a project or experiment to address any one of the questions. The scientist can choose to fund the projects at their discretion and can follow along with advice and input as needed via the Lab Notes section of the Experiment page. It’s an implementation of the “loose-play, early-stage” funding that Chiara Franzoni and her co-authors describe in their “Funding Risky Research” working paper. 

[Note: This is structurally and philosophically similar to the "Science Angel" program we already run, just a different way to frame the idea.]

An Open Lab is not as intimate and engaging as the traditional academic lab, but it's much more scalable. It's different. I imagine the Open Lab as useful for two specific types of advising scientists. The first is the dropouts: the researchers who have left academia for any number of reasons, like joining (or starting) a startup or Focused Research Organizations, but still want to keep moonlighting as a scientific advisor and enabler. 

The second is for the open scientist: researchers with academic appointments who also want to leave the door open for serendipitous involvement from outsiders. Talented amateurs or students from around the world would have an easier on-ramp to working with more experienced scientists. The Open Lab doesn't replace the existing lab, it augments it. 

It's an even better idea for graduate students. Mentorship is a critical part of the scientific lifecycle and the mentor/mentee relationship is a strong predictor of academic success. But it’s not a one-for-one match. Research has shown that mentees that have separate graduate and postdoc mentors with different specialties were found to have greater success. Open Labs could help make that type of diverse mentorship easier and more common. 

The Open Lab could also help ease the problem of "graduate-student illiquidity" as Milan Cvitkovic has described: 

"It’s extremely hard for graduate students to change advisors or work without an advisor. This labor illiquidity is due to the advisor’s monopoly over graduate students’ funding and future career prospects. Graduate students are funded by their advisor, except for a small minority of students who receive personal fellowships like the NSF’s GRFP. And a graduate student’s (or postdoc’s) advisor reference letter is by far the most important factor in future academic employability.

The negative effects of this illiquidity include advisors taking excess credit for their students’ work, advisors forcing students to work on projects they don’t care about or do jobs that benefit the advisor but not the student, and students being trapped into working under abusive advisors."

A clearinghouse of Open Lab opportunities could be the lifeboat that keeps a talented scientist afloat through a bad situation. 

The easy critique of Open Labs (and the reason it's not currently being done) is the logistical and paperwork headaches. We've solved many of those problems with the work we've done at the Experiment Foundation to create an effective model that works with existing institutional realities to get projects funded quickly. It's technically and legally possible to do this tomorrow. The harder part is breaking the operational habits of researchers and funders who are already overloaded and stretched thin. But we should try. 

Another common critique of all new funding techniques is the difficulty in defining success. Measuring the returns to investing in science is fraught with problems, but we could apply the same rubric we’re using with Science Angels. A good initial test would be 10 Open Labs, each with budgets of $100k. In the near term (1-3 years), we will measure the success of the program by the number and quality of funded projects. We will also measure success by the additional grant funding raised by crowdfunding, other philanthropic funders, or federal grants (similar to how venture capitalists measure progress in the near and short term). In the long term (3+ years), we will measure the success of the program by the number of published papers, new patents, or intellectual property created. If additional funding allows, we can run a program evaluation using a similar methodology that was applied to the Wagner and Alexander assessment of the National Science Foundation's SGER program.

The highest vision for open science has always been more than just access to scientific literature. It must also involve access to tools and mentorship. Open Labs is an easy "science better" experiment that could yield progress on the latter.