The article discusses different metrological problems connected with t
he use of a hot-wire technique in the measurement of turbulence energy
dissipation. In particular a short review of the experimental evidenc
e for and against the local isotropy hypothesis is presented and its c
onsequence for the design of hot-wire systems and data processing is d
iscussed. Finally, the finite spatial resolution of a hot-wire techniq
ue is analysed with special emphasis on measurement of the dissipation
rate and the correction of experimental results.