If you assume every caller is false and do assign resources, why would you not assign them based on the same triage system?
Because if you assume every caller is false then you don't have enough (or maybe any) information to implement a triage system.
If you
do enact a triage system based on information the calls are giving you, then you're
not assuming the calls are false. Get what I'm saying? Your still-making-a-triage-from-the-calls emergency center is more a facility that dislikes its callers than one which has a policy of believing its callers false.
If you've assigned all your resources there's nothing saying you are any more or less likely to have your next call (which you can't handle) be trivial or a 3,000-person building explosion (this is true whether all units are distributed to true calls or if some are distributed to false). There's also nothing saying the false negative is more or less likely to be trivial or tragic.
If the emergency call is "trivial" (i.e. doesn't actually call for emergency intervention) then it's not a "false negative," now is it?
While I appreciate your counterexample, I don't think it's applicable.
First, because our hypothetical presumes-callers-false center has no reason to allocate resources to that emergency.
Second, because the presumes-callers-true center, while naive, has no special reason to waste resources or massively over-allocate to a call. In fact, I've been assuming that the call centers make optimally intelligent allocations based upon what they consider true about their information (the calls for service).
Also, an emergency
response center, by its nature, can't predict future emergencies. What emergency response policy ignores known emergencies to instead be ready for an unknown future emergency? If all of the center's resources are allocated it doesn't make logical sense to blame false positives over true positives or to malign an aggressive response strategy.
There's all sorts of problems with the assumptions, and the extremes at the end don't really help prove anything.
I'm assuming that we have: call centers with finite resources; that can optimally allocate to reported calls; receive only information about the readiness of their resources and the incoming calls requesting service; and that their assumptions about incoming calls actually effects policy.
If the presumption of good or bad faith as part of policy doesn't otherwise effect policy or outcomes then how can we tell the difference?
Anyway, I think I mentioned that I might have been using presume and assume interchangeably.
The justice system (and most civil machinery, come to think of it) needs to assume presume that its users come in good faith
Presume in good faith until proven in bad faith.
Anyway-
I suspect the correct and actual policy is similar to the legal one: "Assume all parties are in good faith at all times, but make the system difficult or impossible to use without a showing of it."
This is essentially what I think occurs.
If you're going to presume people come to you in bad faith, by what machinery could a request for service prove that it is in good faith?
We should probably also decide what we mean by "good faith" vs. "bad faith" at some point. It feels pretty freaking vague right now. I know what I
want it to mean, but I'd rather make sure we have a sensible consensus about it.
EDIT:
Fuck, I hit post instead of preview.
Giving it the once-over.
EDIT EDIT:
It probably deserved a few more minutes on the drawing board, but going back and editing it now is probably unfair if I caught someone in the middle of reading it.