“Your article answered some of my questions, but I’m still
not sure why the sensor cannot be the same distance from
the flange (by which I assume you mean lens mount?) as on a
35mm camera?

Is it a design choice that everyone has gone for, or a
necessity? ”

A basic problem with digital cameras is the cost of the sensor chip gets MUCH higher as the area increases. It’s not just a linear factor that a chip of 4x the size costs 4x as much–it’s actually worse than that. There are a certain number of unavoidable defects scattered randomly across a silicon wafer; and if you are only getting 6 chips out of the wafer rather than 60, the odds of any particular one being spoiled and unusable shoot way up. So that’s where the economic pressure comes from for cameramakers to use the smallest sensor they can get away with.

The rule of thumb is that a normal lens is one whose focal length roughly equals the diagonal of the image format; so the smaller the image sensor, the shorter the focal length needed to yield the “normal” angle of view.

Canon, Nikon, and Minolta wanted to keep their lens mounts compatible with their earlier film cameras; but even Olympus who were starting with a blank slate with the Four-Thirds system kept a longer flange-to-sensor distance than the “normal” lens focal length for the format. One reason is that the early generations of digital sensors didn’t perform very well when light struck them at oblique angles. So it was preferable to hold the rear element of the lens a bit further away from the sensor, so light would reach it at more of a perpendicular angle.

Otherwise it ought to have been possible for the camera to be extremely compact–as Olympus themselves proved with their “Pen” series of SLRs which shot a half-frame 35mm format.