I was reading about "error analysis", in particular about rounding errors in the floating point operations representing the real operations such as multiplication or division.
There was the following:
$$x \mathbin{\hat{\circ}} y = (x \circ y)(1 + \epsilon)$$
for an $\epsilon = \epsilon(x, y)$ with $|\epsilon| \leq eps$.
Is there some kind of conventional way for mathematicians to pronounce $x \mathbin{\hat{\circ}} y$? I would say "x circle hat y" or "x approximated floating point operation y", but as you may have noticed I'm not a native English speaker.
Second and maybe more important question. What does the notation $\epsilon = \epsilon(x, y)$ actually mean?