> For those of you familiar with the binary format of floating-point numbers, we do not think of them as being “gamma 2.0” — they are still just linear values, we just bear in mind that their precision characteristics are similar to those of a gamma 2.0 curve.
Actually floating point numbers are distributed roughly like log₂x, not like √x. https://commons.wikimedia.org/wiki/File:Log_by_aliasing_to_i...
> Monitors were (and still are) gamma-space display devices
This is misleading. LCD displays are “gamma-space display devices” not because of anything inherent to the medium, but because they have built-in lookup tables which interpret incoming data that way and set output to look correct based on the particular characteristics of the display. Cathode Ray Tube displays do have a nonlinear relation between voltage and light output which is fairly similar to an inverse gamma curve (of gamma ~2.5), but that’s more a convenient coincidence than a reason for gamma encoding digital images, which is primarily about using a limited number of bits in an efficient way.
* * * * *
Anyone who needs to deal with gamma encoding issues should really read Charles Poynton’s FAQ and FQA: