12 ms·
> Using the same limitations you seem to have in mind, we can't get one third exactly either. We can work with exact rationals; we can do arithmetic on them an
by lmm 8y ago
> Using the same limitations you seem to have in mind, we can't get one third exactly either.
We can work with exact rationals; we can do arithmetic on them and normalize them. In particular, we can compute whether two rationals are equal (in finite time), which we can't do with reals.