Earlier quoted context omitted.
> That seems false as applied to most of my examples. If the Big Bang theory turns out to be true, then that doesn't mean that steady state theories were "incomplete", it just means that they were wrong. They still seem incomplete (and not "wrong" to me). They served their purpose and were good approximations for calculations we did and observations we had all those years. > Similarly, if it turns out that eating lot…
>They served their purpose and were good approximations for calculations we did and observations we had all those years. In other words they were empirically more-or-less correct. That's what I was talking about in my first paragraph. >Only if "previous dietary advice" was only all about fat, and only if it consisted of just saying "fat makes you fat". I don't see the point of picking over the details of the example.…
That's the whole point of Asimov's article - the relativity of wrong. The example he gives for this is not at all different from yours (earth being flat and earth being non-flat are also opposite propositions). So it's not like the "plate tectonics" examples changes anything in his argument.
The argument is that P and ¬P can both be useful if the first could explain a phenomenon to the best degree available at a time, and the second to an even better degree available at a later time. In that sense P and ¬P are degrees of wrongness and truthness.
His is the argument that "wrong" depends on utility, as there's usually no absolute match with the objective status universe (ding an sich), so we just have approximations with different degrees of wrongness.