TAQ:
This means that out of the 203,000 ERV's reported in the human genome only 82 are not shared with chimps at the same base.
Are you sure this is a valid method of calculating Orthologous ERV's? I tried to find an independent source and the best I could find was
Orthologous endogenous retroviruses exhibit directional selection since the chimp-human split which says "Overall, we identified 336 chimp-human pairs of sequence from a variety of genomic locations".