So, rather than promoting "the most inclusive definition of fairness....even if that renders the definition pretty much useless to any DM that's not a psychopath" I am promoting a reasonably inclusive definition of fairness, that recognizes that there is a difference between intrinsic fairness and whether or not something is fair within a particular subset of context.
"Reasonably inclusive" seems subjective to me. "Reasonableness" is not objective. If it were, we wouldn't need a jury of people to vote their opinion on what constituted reasonable, but that's exactly what they do. I would be very skeptical of someone who claims to be able to define reasonable in a way that excludes cultural and personal idiosyncracies and still resembles what people's working definition of the word means. "Reasonable" is just another subjective word, and so would be of little help.
What the law establishes is a set of procedures and definitions that are used to judge a situation. It attempts to remove layers of subjectivity in terms of definitions, but not results. So if the difference of degrees in murder is clearly defined as differences in premeditation, it helps to clarify that, even if in the end it comes down to the subjective opinions of the jury as to what happened.
That premeditation
is an important factor in how a murder should be punished is
not objective. It is entirely derived from custom and habit. It's what "feels" right to the majority and it was decided based on consensus.
That a trap should allow a chance of survival is equally subjective and based on custom and habit.
In the world of DnD and judging traps, we're in a very primitive situation (and also in a situation that you would be hard pressed to say is a matter of "the common good"). Nobody (AFAIK) has even agreed on what the % fatality of a trap ought to be, much less what a given traps % fatality is. IMO people are never going to agree on the former (differences in old-school and new-school, for example) and there's no Council to vote on the latter.
In the case of traps:
1. is the rogue responsible for maxing out his ranks in search, and does he deserve to die if he takes fewer ranks (or puts a sub-optimal score in int, etc.)
2. is it unfair if one person dies and everyone else lives? (cf. Survivor)
3. is it unfair if one person lives and everyone else dies? (cf. Monopoly)
4. is it unfair if everyone eventually dies (cf. Pac Man, real life)
5. is the chance of survival to include the use of augury?
6. is the chance of survival to include not having discovered the trap (ie. not gone down that passage, not pulled the lever) to begin with?
That's off the top of my head. Those are
all subjective issues AFAICT and I really don't think much is to be gained by assuming the answers (as people often do) and then arguing at cross-purposes with someone else who has a different set of answers.
If two people agree on the basics (which can be an involved process to identify/define to begin with) then maybe there's some use in ferreting out inconsistencies. For example, if I believe that a "fair" game is one where the PCs survive all of the time, then pointing out that a trap might kill my PCs would be useful.
One thing that could help would be defining a set of criteria. The "Raven Seal of Trap Approval" for example (sounds grim). Then you could say, according to that defined standard, how a given trap rates (which reduces the level of subjectivity by one). Calling such a standard "reasonable and universal" (and by implication, those who do not agree are marginalized and unreasonable) is somewhat risky (to put it nicely). It amounts to telling people how to play RPGs and I don't think the industry (consensus) really has the stomach for it, for what that's worth. So why should we?