According to a news release today from JPL, studies based on infrared observations from the Spitzer space telescope indicate that many, and perhaps most nearby Sun-like stars may form rocky planets (estimates range from 20% to 60%). This is based on analysis of multiple infrared wavelengths associated with rings of dust around the stars (neither Spitzer nor Hubble can directly detect small Earth-like planets even for nearby stars).
In about a year (February 2009), NASA will launch Kepler, a mission specifically aimed at detecting Earth-like planets in the habitable zone around nearby stars. Kepler won't be able to directly image these planets, but instead will use the the transit method to detect the slight dimming caused by a small planet passing in front of its much larger sun. It will literally stare at a single region of the sky for over 3 years to collect photometric data on around 100,000 stars in a patch of sky between Deneb and Vega (see graphic). Since the transit method depends on the planet passing the star across the line of sight to Earth, only a very small percentage of stars with such planets will be "aimed the right way." This is why they will need to monitor 100,000 stars for over 3 years (because like Earth, these small planets would need to be moderately far from their star to be "habitable" and therefore would have a period of something like an Earth year).
The dust-based Spitzer studies (being presented this week at the AAAS conference in Boston) are encouraging for Kepler's planet-finding prospects. If the optimistic end is right and 60% of sun-like stars have rocky planets, then Kepler could detect a pretty decent harvest of well-aimed rocky planets.