Big science means big opportunities — huge centres, multimillion-dollar contracts and, often, hundreds of funded positions for investigators, postdocs and students. But choosing to take part in such enterprises, at any level, also comes with big risks — or at least some pretty serious questions about job security.

Take the announcement earlier this year that the US government plans to abandon the Hubble Space Telescope (see Nature 427, 574; 200410.1038/427574b). Hundreds of scientists are involved in ground support at multiple centres in the United States and Europe. When Hubble stops taking data, perhaps as soon as 2007, will the support staff stay on with the same institutions? Or will they slowly be phased out by attrition of contracts?

The answers remain unclear. But Steve Beckwith, director of the Space Telescope Science Center in Baltimore, Maryland, which provides ground support for Hubble, says that as things stand, the end of the project could mean layoffs for about half the centre's workforce over the next two to three years. The lack of clarity comes from the possibility that the institute's staff may still be of service — there are proposals to maintain Hubble using robotics. But such an endeavour has never been successfully done, so it is unclear whether it will be funded, planned and staffed in time.

Similar questions can be asked of job security at other centres. What will happen to scientists at Fermilab in Batavia, Illinois, when the larger, more powerful Large Hadron Collider makes its equipment effectively obsolete?

The moral of Hubble's recent news is to evaluate the longevity of a big project before you sign on. And try to develop skills that will be relevant to another effort once that big project inevitably comes to an end.