This is really interesting. Can you elaborate on what they might have been thinking when commenting out those tests? (which seems pretty blatant). Were the tests commented as to what they were doing, and obviously were important, relevant, and should have passed?
My impression is that when you go to a higher (cleaner) level of abstraction, you lose some immediate abilities. For example if you move a global variable without classes to a static variable in the class it really belongs with (after writing that class), you can never (by doing that) enable additional functionality over having it be a global variable. But the variable might no longer be accessible at some point, perhaps if a different base class has a static member whose initialization depends on another one already being initialized - that sort of thing.
which is the whole point. you are applying a new abstraction, and in the new abstraction they're not global variables but properties of a class. Perhaps some tests become irrelevant or need to be rewritten.
So, you could have some test fail as written, because you've moved to a higher level of abstraction. This makes me wonder what the exact tests were in the case you refer to...
This (sw design) is a multidimensional optimization problem with a somewhat vague cost function. But argmin on the # of bugs and argmax on feature completion is far more important than everyone following some random standard.
Some people are convinced that making components isolated and aesthetically clean is more important than making them functional.