Alex Arslan

Requires is not a good solution to conditional dependencies. Since this package is used by a number of others packages, we shouldn't propagate a Requires dependency.

@galenlynch You should now have received an invitation to join the organization.

Good luck on your defense!

@andreasnoack We could try something like `allunique(x) ? Inf : zscore!(...)`, where `Inf` could also be `NaN` or whatever. Dunno if that's the best course of action, just an idea.

From a purely mathematical standpoint, a z-score of 0 wouldn't make sense for a constant vector since you're dividing by a standard deviation of 0. Btw for what it's worth...

In terms of implementation would it make more sense to catch a constant vector using `allunique` or something like `isapprox(std(x), 0)`?

Seems like adjusting the `mean` computation would be a good candidate for an enhancement in Base rather than having a competing `mean` implementation here.

Regarding speed, I was looking at [this paper](http://arxiv.org/pdf/1505.05571v1.pdf) and it indeed seems that precision-preserving summation is much slower.

Huh, weird edge case. Perhaps worth noting that 2e-314 is [subnormal](https://en.wikipedia.org/wiki/Subnormal_number), i.e. it's less than `floatmin()`, and using `floatmin()` here works as expected. This isn't my area of expertise; perhaps...

This looks good but the prior uses of `Float64` make me worried that the code may have issues around under/overflow when using other types.