We provide a concise introduction to CSS and CSS-T codes from the perspective of classical coding theory. We demonstrate that pairs of linear codes that yield a CSS code with good correction capability can be easily produced using a randomized construction when the cardinality of the base field is sufficiently large. Next, we prove that CSS-T codes exhibit the opposite behavior, showing that, under very natural assumptions, their rate and relative distance cannot be simultaneously large. We conclude with a simple construction of CSS-T codes derived from Hermitian curves. This is joint work with Elena Berardini and Alberto Ravagnani.