> No. It is very much possible when the measure is undefined, as was the condition setup before the bet was made.
I promise you, it's not, because that's not how medians work. This is the Wikipedia definition of a median:
> In statistics and probability theory, the median is the value separating the higher half from the lower half of a data sample, a population, or a probability distribution.
There have to be half on each side or it's not a median. The mean doesn't have to be distributed like that, but the median does. It doesn't matter if the measure is undefined; the median divides the upper half from the lower half. There are always 50% on each side (and one on the median, if there are an odd number of samples).
> If we take it to the logical end where your chosen measure is simply the inverse of my chosen measure, then those above the median by my measure fall below the median by your measure and vice versa, leaving the entire set of those outside of the median to be seen below the median (and also above the median).
Sure, but that's not a median. That's the intersection of people above 2 separate medians of 2 separate measures. You can't combine the median of separate measures and have it still be median; you can combine the measures into a single measure and that has a median where half are above and half are below.
> As it pertains to the larger discussion, how the quality of a developer is measured has not been established, which means everyone is likely to measure differently and often in conflicting ways, and so most of us are bound to be below the median according to someone.
Totally agree there, I'm just beating a pedantic, dead horse that that's not a measure (it's a bunch), and it doesn't have "a" median (it has a bunch). You're talking about the intersection of people above the medians of a bunch of measurements, which doesn't share any of the properties of a median or being above/below it.