I think applying a fractional template like that would be almost the classical definition of a rule that could be skirted around by those intent on causing trouble. And rather difficult for a moderator to enforce.
Seems to me it's just a judgment call. In the tech forums, the mods have no difficulty differentiating between someone who's just calling names for the sake of it, and others who are generally engaging in a quality discussion that veers off into flame-land. The latter situation is pretty easily corrected with a word to the wise.
But there are much more objective/testable realities and truths to be found in computer hardware than there are in politics.
If A calles B else a name, while correcting him, in hard-ware because B is an AMD fan boy and A knows B is giving bad advice; then there are reasonably accepted objective criteria for this.
On the other hand, if A says "Bush didn't mind bringing economic ruin to the country to benifit his own wealth" and B "corrects" him and similarly throws a barb at A, what with A being a "lib-tard"... well A is making a hyperbolic statement, one well outside main-stream thinking, but one that you could back up with observations of a set of political behaviors; on the other hand, from even a moderate liberal perspective, to argue that a president is anti-american is loony and saying as much about A would be true from B's perspective, and hardly an insult.
So hardware is much less ambiguous than politics; meaning that the butt-hurt when you make a call on either side of the above-style of back-and-forth is exponentially higher.
***
more technically:
In applied psychology we have three kinds of 'justice' that we think of. Interactional (degree to which the people affected by decision are treated by dignity and respect), distributive (is justice meted out/provided for everyone equally?) and procedural (fairness in the processes that resolves disputes and allocates resources).
Looking at the situation through these lenses, we can see that when B insults A, there a judgment made by the mod to either do something or not, regarding the interaction. One could easily see that interactional justice is a primary concern at this point: does the insulted party get the dignity/respect of a moderator that comes from doing something about the insult.
Distributive justice requires that 'class' not be taken into account when distributing justice. While Perk would prefer not mention this, the fact is that some folks here feel like they aren't part of the preferred clique, and thus are not receiving distributive justice. On the other hand, basic moral fairness and concern pulls on a mods conscience to be as fair and equal as possible. The truth is that everyone is subject to subconscious influences that make them think they are thinking/behaving rationality, when there are many other things going on that influence us; no matter how logical/ethical we intend to be.
Procedural justice, the level at which we are talking about in this thread; It's built on suppositions about the aforementioned levels of interaction and can't exist without the previous two. Factually, no justice can prevail because it's illusory, a story made up by people in power to justify power to themselves and others. There's no jury by peers, no trial for banning; no public airing of disagreement: there is bringing a request before the empowered elite and asking if they support their fellow mod-friends or not. Rules, in procedural justice, are not about even handedness, they are about fairness (mitigation of the over-lord's power) in solving a disagreement with a moderator (as is the case when someone disagrees with points/banning).
I don't think we should even be looking for procedural justice because it just doesn't make sense for an online forum. Everyone wants distributive justice, but when it is perceived (or actually) not there, there is often an imbalance between a side that thinks its just perceived and the other that think's its actual. As for distributive justice: we tried allowing people to distribute justice based on clicking the ! icon... that worked to get trolls to bring about 'justice' against those that they didn't like. In the end, then, just rule enforcement is an impossibility.
Instead I suggest we appeal to a kind of ethically-answerable pragmatism. In this senario the mods are not jail keepers, making sure the inmates stay in line; but instead are coaches, helping folks that are going off the rails. We all have moments where we lack clarity, times when our ideas are expressed but they don't load onto the debate, ways of presenting the other person that reflects what they think of themselves in our mind, but does not reflect what they think they think of themselves.
If moderation started with the perspective "how do I help this person work better in the community" and then banned when someone continually didn't even
try to shape up, then you could vacation them so they can cool off and figure out what to do differently. I don't think this would be all that much more work in the long-run; though there is a start-up cost.
I think this will work because personally, Harvey did this with me. With no intent of trolling, I was a religious troll, and by working with me toward being a better member of the community I also became a much better person IRL.
The precision of static rules often falls flat in the fluid, dynamic world of complex people interacting.
In organizational theory we deal with this by adding ambiguity to the rules
