And while roles primarily come from the classic fighter/cleric/rogue/wizard classes
There's actually another source, one that very often gets overlooked or ignored.
Soccer. Or Association Football, for those who don't like the nickname derived (in Britain,
by the Brits!) from the word "association."
Defenders, Strikers, Leaders, and Controllers are
literally tasks players can be assigned on the soccer field. And two of the most common approaches to being a Defender in soccer...are to "mark" (that's
literally the word used) specific opponents to track, or to lock down one specific
area of the field around you. Y'know...literally the two major gameplay mechanics of Defenders in 4e: marks that target specific opponents, and auto-marking "stances" which affect anyone that gets close to you.
For all the hullabaloo about 4e being unrealistic...it's heavily inspired by very real and very meaningful tactical choices in a competitive environment. Of course, soccer players don't normally kill their opponents, but soccer was just that, an inspiration, not a carbon-copy adaptation.
I wish people would stop cautioning others for daring to dream... Maybe its just better to stop relying on others, hoping they fulfill them for us. fingers crossed
There's a tug-of-war between, if we phrase them nicely, "idealism" and "practicality". Too much idealism and you get wishful thinking and endless disappointments that can leave a person bitter and distrustful. Too much practicality and you get cynicism and prodigious lack-of-imagination fallacies.
So, in the spirit of being respectful of both idealism and practicality: If someone
does decide to do this, they'd better make sure they have a good solution for the "Marking problem" I cited above. That is, Marking only works when there's a real person's
brain behind it, making real
choices, rather than RNG or threat-tables or scripted behavior. In the absence of a real, decision-making person, you're left with having to figure out how to program a computer to behave strategically. This is very, very difficult. It is not totally impossible (ask me sometime about the genuine
cleverness of the GalCiv 3 AI), but it is quite hard.
Given how important well-executed Marking mechanics are for making this system work properly, this is a hard problem that needs to be solved. That is a reason to pause and reflect, and to confirm that your ideal product is, in fact, actually possible.