In August the White House announced that the results of all federally funded research should be freely accessible by the end of 2025. This will be a big change for scientists in many fields but ultimately a good move for the democratization of research.
Under this new guidance, many peer-reviewed papers would be free for the world to read immediately upon publication rather than stuck behind expensive paywalls, and the data that underlay these papers would be fully available and properly archived for anyone who wanted to analyze them. As an astronomer, I’m pleased that our profession has been ahead of the curve on this, and most of the White House’s recommendations are already standard in our field.
NASA, as a federal agency that funds and conducts research, is onboard with the idea of freely accessible data. But it has a plan that goes much further than the White House’s and that is highly problematic. The agency currently gives a proprietary period to some scientists who use particular facilities, such as a 12-month period for the powerful James Webb Space Telescope (JWST), so that those scientists can gather and analyze data carefully without fear of their work being poached. NASA is looking to end this policy in its effort to make science more open-access.
Losing this exclusivity would be really bad for astronomy and planetary science. Without a proprietary period, an astronomer with a brilliant insight might spend years developing it, months crafting a successful proposal to execute it, and precious hours of highly competitive JWST time to actually perform the observations—only to have someone else scoop up the data from a public archive and publish the result without giving fair credit. This is a reasonable concern.
Without a proprietary period during which the astronomers who proposed given observations have exclusive access to the data, those researchers will have to work very quickly in order to avoid being scooped. Receiving credit for discoveries is especially important for early-career astronomers looking to establish their credentials as they search for a permanent job. Under such time pressure, researchers will need to cut corners, such as skipping checks and tests that define careful work. Such a sloppy approach will lead to hasty results and incorrect conclusions to the detriment of the entire field.
It also can lead to the erosion of work-life boundaries, with astronomers working long hours, sacrificing their health and family time so their result gets out before the competition’s. This is bad for the culture of science and disproportionately affects those with children or other time-consuming personal circumstances (such as being a student, a caretaker or a full-time college instructor while also performing research). Allowing researchers to properly benefit from their work is critical for making astronomy as fair and equitable as possible.
The leadership of most observatories realizes all this and correspondingly enforces a proprietary period—usually between six and 18 months—within which the architects of an observation get to work on their data without competition. The period is at least six months because good science takes time: cutting-edge observations, for instance, often require devising novel data-analysis techniques to interpret signals of low statistical significance. Such periods rarely exceed 18 months, as a compromise to guard against researchers indefinitely sitting on taxpayer-funded data that really should go public eventually. This ensures the practice doesn’t grant researchers complete exclusivity—just a reasonable and well-deserved head start. As a result, astronomers keep producing robust results at a good pace.
There are, of course, situations in which proprietary periods are undesirable. For instance, one is allowed to bypass the usual proposal machinery to use the Hubble Space Telescope for emergent, especially timely observations that cannot wait for the completion of a many-months-long standard proposal cycle. The trade-off for taking this less rigorous route to winning time is that the entire community then gets to work on the problem. There are also large surveys and other foundational projects with broad community support and corresponding community benefits. Data from these programs are typically free for all to use immediately. This “open-science” approach has produced blockbuster results and amplified the output of the observatories that pursue it.
But most observatories also acknowledge the value of proposals from small groups or even lone individuals hoping to execute an idea on their own. It would be a shame if, in pursuit of open science, JWST closed this historically fruitful avenue of discovery.
Some in favor of abolishing proprietary periods have argued that doing so will safeguard equity in astronomy by allowing underresourced scientists the same access to data as everyone else. But by eliminating proprietary periods, the only data sets such scientists would gain access to would be those for which other researchers are already hard at work. Such a change would thus only allow them to scoop other (potentially better-resourced) scientists while at the same time ceding control over any data they themselves might produce, enabling those other scientists to scoop them right back. On balance, the better-resourced scientists would win out, all while creating an unhealthy and unnecessary culture of haste and competition. This would be a bad trade.
NASA’s deputy associate administrator for research Michael New has argued that if eliminating proprietary periods will disadvantage underresourced astronomers, the solution is to provide them with more resources. But time to work with the data is the most precious commodity here, and it is also the hardest thing for any amount of extra funding to buy. Providing extra funds to help free up a researcher’s time—with a new lab assistant or a nanny, for instance—is an inevitably piecemeal fix. It’s much cheaper and simpler to use proprietary periods, which are a narrowly tailored and specific solution to a real problem.
Without proprietary periods, astronomy would need to find new ways to ensure that credit goes to those who gathered the data when other scientists publish it. Without such periods, there is only a loose culture of shaming to prevent this: astronomers who scoop others, especially students, with their own data may be stigmatized within the community. This is not a universal attitude, however, and is not a very effective or desirable way to solve the problem. Such stigmas also work against the whole premise behind making data sets public, which is that everyone should be encouraged to use them.
One potential alternative is to create a professional requirement that those who proposed an observation but have not published from it should be offered co-authorship on any paper that uses the data. This is not currently the cultural norm in astronomy—in part because inviting “strangers” to be co-authors on one’s papers also comes with a whole host of complications—but it still merits exploration. Another option is to change the standard for how credit is assigned for any observational work. Astronomers could, for example, demand that any paper citing a result also cite the proposal that generated the enabling data. In this way, the proposal team could still accrue credit for its work, even if it wasn’t the first to publish.
In the end, though, such adjustments are secondary to the heart of the matter, which is that NASA’s plan to eliminate the proprietary period for JWST data is bad for astronomy.
The Space Telescope Science Institute, which manages and operates JWST, has started polling astronomers on the topic. I hope that once NASA hears our positions, it reconsiders this stance and maintains a healthy, reasonable proprietary period on appropriate classes of JWST data. Astronomy and astronomers will be better for it.
This is an opinion and analysis article, and the views expressed by the author or authors are not necessarily those of Scientific American.