If ETRS89 (EPSG:4258) is tectonically fixed and WGS84 (EPSG:4326) isn't, the Helmert coefficients shouldn't be zero, right? In fact these parameters should diverge more and more with every new release of QGIS
I understand they're based on the same ellipsoid, but I think their central meridians are diverging, no? Why is the transformation (0,0,0,0,0,0,0)??