The Kepler telescope, a one tonne satellite launched in 2009 to scan for planets outside of our solar system, is dead.
Over the four years of its mission, the spacecraft has found over 3,000 stars which might have planets orbiting them, and the earth-based analysis has confirmed 134 planets orbiting 76 of them.
In April, the mission even announced the discovery of two potentially habitable planets. The two were roughly earth sized, and roughly earth temperature as well. Each of them might just lie in the "habitable zone", where liquid water can potentially exist.
Sadly, shortly after that discovery, Kepler ran into problems. The telescope manoeuvres through space with four reaction wheels, which keep it pointing in the right direction using a gyroscope effect. In July 2012, one of the wheels failed, but the mission was designed to only require three to accurately aim. In May this year, however, a second wheel failed. Yesterday, Nasa announced that they had given up trying to fix the wheels. Kepler can no longer be targeted accurately.
But, good news! Rumours of Kepler's death have been greatly exaggerated (by me, in paragraph one. Sorry.), because while the telescope can't be aimed any more, it's still useful. The imaging functionality works fine, and with two remaining reaction wheels and a limited amount of thruster fuel left it's even got a bit of manoeuvrability. NASA has opened it up to the community to work out the best way to make use of what's left working; one proposal, for instance, involves heavily post-processing the images to remove drift caused by the lack of a third wheel.
There's hope for the mission yet, which is great, because as Ars Technica's John Timmer writes, there's a lot more to learn:
A longer mission would identify planets further from their host stars. To identify a candidate, Kepler needs to see it pass between Earth and the star the exoplanet orbits three times. The further out a planet is, the longer one orbit takes, so the longer it will need to do three passes. As such, the existing data is heavily biased toward planets that orbit very close to their host stars; this also means that most of the potentially habitable planets we've spotted are orbiting dwarf stars, which are dim enough that water can remain liquid close in.
In other words, we simply haven't looked long enough to detect planets in a habitable zone around energetic stars. We have a much better picture of the diversity of exoplanets, but it's far from a complete one.