Kepler is NOT an improvement on Hubble. It is a single purpose PHOTOMETER, not an imager. It covers a wide field recording for five years one particular star field with part-per-million photometry. It is looking for transists by Earth sized planets. While Hubble has recorded occultations, it can't search blindly for them, and can't find planets as small as Earth. Likewise, Kepler can't and won't take great pictures, or spectra. Its images will be lower resolution than amature ground based images, and will be monochromatic (black and white) to boot.<br /><br />The James Webb Space Telescope, while much bigger than Hubble, will not be a direct upgrade. One of Hubble's strengths is visible and ultraviolet images. JWST is primarily for longer than two microns in the infrared.<br /><br />I believe that the HST servicing mission is scheduled for April 08. That is based on all the preceding missions going up on schedule. (Who here thinks NASA can launch one shuttle in July and another in August?) According to some online sources, $200M is being spent to stay prepared for this mission, including continuing work of the two instrument COS and WFC3.<br /><br />While Hubble is an absolute triumph, it is joined at the hip to the Shuttle. We won't be able to finish up the last 18 Shuttle missions for even $27B, (to use easily divisible numbers for 2003-2010) so each is an incredibly expensive proposition of around $1.5B. (Note that most of this money is being spent whether or not the Shuttle flies.)<br /><br />Hubble 2 has been proposed. If done without servicing, it has been claimed to be cheaper than a Shuttle flight, but that is easier said than done. There is even a second Primary mirror in storage, one polished by Kodak that would not suffer from spherical aberration. However, I don't think anyone expects that idea to go anywhere.