I may be alone on this but I think the Imperial system has advantages for working with middle sized numbers. For very large (for example measuring space) or very small (measuring a molecule), base ten units make computing easy since we use a base ten numbers system. BUT for some things a base 16 system is really nice. For example, I often need to find the center points of various measurements. When a fraction is involved it is often easier to divide a base 16 number than a base 10. Half of one is 1/2, half of 1/2 is 1/4 half of 1/4 is 1/8 half of 1/8 is 1/16. With a base ten system you quickly run into undesirably lone numbers, .5, .025, .0125, .0625. Maybe the best would be to adopt a base 16 numbers system first. Then the "decimal" would align with the "fractional".