QUESTION:
Given the advantages of Galileo's dual-spin design, how come it hasn't 
been used by newer missions (e.g. Cassini, MGS, MPF)?

ANSWER from Nancy Vandermey on April 8, 1997:
Unlike previous planetary spacecraft, Galileo features an innovative
"dual spin" design: part of the orbiter rotates constantly at three
revolutions per minute, and part of the spacecraft remains fixed in
inertial space. This means that the orbiter can easily accomodate
magnetospheric experiments (which need to take measurements while
rapidly sweeping about) while also providing stability and a fixed
orientation for cameras and other sensors. MGS is a planet mapping
mission; the cameras and instruments in general take data of the
surface of Mars, so there is no need for a spinning section of the
spacecraft. "Fields and particles", including magnetospheres, are the
the investigations that benefit from a spinning spacecraft.  So MGS
and Pathfinder (which is a Mars lander/rover mission) would not benefit
from a dual-spin design.

Cassini, on the other hand, is a planetary mission very similar to
Galileo. I asked Bob Gounley, former Galileo systems engineer now on
Cassini, for some information, his reply is below:
  
Galileo was sold as a relatively straightforward design.  Comsats had
used dual-spin design for years, so why not a dual-spin scientific
spacecraft?  This went against a long JPL tradition of building
three-axis stabilized spacecraft -- a technology that JPL pioneered.

It quickly became evident that a dual-spin scientific spacecraft was
not so easy as the proposal folks presumed.  Comsats send only power
through their rotating joints; Galileo had to send power _and_
high-rate data with a minimum of distortion.  A comsat's spin axis
could afford to be off a degree or so; with a working high-gain
antenna, Galileo's requirments were a tenth of that.  On and on the
complexities mounted.  They were overcome (obviously), but
dual-spinners were regarded with suspicions after that.  From the
beginning, Mariner Mark II program (the predecessor to Cassini)
baselined three-axis stabilization.

The pity is, in this engineers opinion, we have proven that
dual-spinners are a viable design.  One 1982 proposal for a
Cassini-like mission essentially had us build a second Galileo with
spare parts.  This would have been much cheaper, but the proposal
didn't go anywhere.  A new design with a reliable three-axis
stabilization and _two_ scan platforms was chosen.

In the end, cost-overruns have pared down Cassini's complexity.  Now
there are no scan platforms and science must make do with moving the
entire spacecraft.