Helioseismology has provided us with what we need
for highly accurate models of the Sun and (one assumes) solar-like
stars. This permits the distance scale for nearby clusters to
be determined to great precision. Future astrometric satellites
will generate accurate stellar parallaxes, but in order to use
these effectively we also need good stellar temperatures and metal
abundances. I will present a discussion of the many issues involved
in using stellar parallaxes to calibrate stellar evolution models,
particularly of metal-poor main-sequence stars (subdwarfs). The
main problem is the inability of spectroscopists to settle on
a uniform temperature scale.