Just know the boundary between "socializing" and "fraternizing" and you should be fine. I've seen a few examples where crossing this boundary was rather detrimental to the PhD advising process (though nobody got fired or expelled; it just got very difficult to force the student to meet her obligations and to carry out the required assignments because the student got an idea that she had a right to argue with and to question everything she was told to do). However, I don't think that maintaining some "class hierarchy" makes much sense at the PhD student level or even during extracurricular undergraduate activities like Putnam training, etc. though I'm all for it in the low level undergraduate classes ever since the time I was being kept on teaching nothing higher than engineering calculus for four years in a row, and the idea of introducing some kind of "faculty uniform" like in the military is not altogether alien to my mind.
One thing to remember however is that no matter how friendly you are with your students outside the classroom, when you are lecturing, you are the boss and they are subordinates. The joint soccer game or beer jug yesterday should not become an excuse for not turning in the homework today. If this principle is understood and followed by both parties, I guess that's all "hierarchy at the graduate level" we need.
As to the journal article in question, it was written by an administrator, albeit a clever one. Most advices he gave are excellent but at some places you certainly get the feeling I had some twenty years ago when I was pulled over for speeding by a policeman, asked him what would be the maximal speed they would allow in that state (North Dakota), and got the reply "Sorry son, but all I can tell is that if you go under the posted limit, we won't bother you").