Why the axis order matters in practice

Microscopes and cameras order their axes according to the hardware, not according to what a model expects.

• In biotech: a drug screen run on transposed images is a wrong decision about whether a compound works.
• In tech: a model trained in TensorFlow (NHWC) deployed into PyTorch (NCHW) with no transpose.

Neither case crashes. Both return confident output, and the output is meaningless.