Sry to revive a dying thread but I like the analogy of DR to a ruler. I always envisioned increasing dr as increasing the divisions between white and black. I do not see how increasing dr in the captured image 'makes the ruler longer'.
You don't see it because you fail to understand why several captures are needed. It is not a matter of number of divisions, it is a matter of
noise.
In a single capture divisions run from 0 to the end of the ruler, but all those levels close to 0 (which correspond to deep shadows in the real world scene) are ruined because of noise. Making extra shots with higher exposure levels will blow highlights but will shift those dark areas to noisefree areas of the sensor. Finally your favourite HDR program blends all together, taking each part of the scene from the appropiate captured RAW files (highlights from the lowest exposure, dark shadows from the most exposed files).
Number of levels is not the reason for multiexposure in the HDR process, they are just a consequence. Noise is the reason. Once DR of sensors reach high enough performance, multiexposure will be unnecesary and HDR photographers will shoot just once. But the tone mapping process and techniques will still be necessary to accomodate in a natural looking way captured information (now from a single input file) into the limited DR output devices (print, screen, projector).