Because every upstream change carries a [however slight] nonzero chance that something unrelated breaks. That's why there's rules. It doesn't follow that an ex

Because every upstream change carries a [however slight] nonzero chance that something unrelated breaks. That's why there's rules. It doesn't follow that an experimental feature should jeopardize stable ones, however small the possibility.