Thanks for the article. One quick question though: isn’t the example in “Covariance of Return Types” violating the “Countervariance of Method Arguments”? The subtype can only receive strings, but the super type can also receive numbers.