This is the written form of this article.

He basically starts with a false definition of what OOP is and then proceeds to make disjointed arguments that have very little or nothing to do with his premise and have little to do with actual OOP is. What he has written seems to be targeted more towards Java and its particular coding style (which I, personally, would agree is terrible) than OOP in general.

If you don't actually know what object-oriented, procedural, or functional programming is I would advise you not to listen to random people on the Internet telling you that OOP results in "bad abstractions" and that procedural programming is "superior".

After years of using dynamically typed languages I'm a strong static-typing convert. Sure, you have easy up-front flexibility with dynamically typed languages. Sure, this may lend itself to rapid prototyping. But I've personally found that trying to write a program of any significant size with a dynamically-typed becomes a giant cluster-fuck. I might hit the ground running on a project, but I often find myself bogged down and having to backtrack and re-write significant amounts code if I don't enforce a stringent type-system in my program to begin with. Using a dynamic language and manually enforcing a type system doesn't make much sense to me when I can get the language to assist me with it.

In my opinion, dynamic languages are great for scripting or writing small chunks of code of maybe a few hundred lines. Little tasks where it's easier to mess around until you hit a solution and re-write as much code as needed (or even the whole program for that matter).