After the first edition of the BOOMERanG data were released and were shown to fit the no-CDM prediction nicely, the most frequent criticism I heard was that no-CDM wasn't a genuine MOND prediction. The implication was that the success of the prediction could therefore be ignored.

Now, after the second edition of the BOOMERanG data release, the only thing I hear is that the third peak is too big.

Which is it?
Should MOND be ignored because no-CDM is just the best first-approximation to its predictions for the CMB, and not a complete extension of General Relativity?
Or must MOND be wrong because the data do not follow every detail of this [previously inadequate] approximation?

It may not surprise the reader to learn that these intrinsically contradictory attitudes have been expressed by the same people.