User:Domin/Harmonic Entropy

Honestly, I think it's time to call harmonic entropy what it is. A failure.

First, it's a model that assumes a certain mechanism of how our music perception works. There is no evidence for it being related to workings of our brains, and what's even more problematic it is not at all clear how we could show such relation. And that part is important, because if the model is not a reflection of some perceptual mechanism, the model becomes an instance of fitting the data with a function that is needlessly complex. I'm sure that given the background Dr Erlich is was well aware of that fact.

So far, the explanatory power is not there, as the model is entirely speculative without any apparent path forward. And that's the charitable view as Dr Erlich's background fully justifies the expectation that he was fully aware how problematic is showing that model assumptions reflect the workings of out perception. It's quite convenient to put forward that work to a community of people where a large part of them lacks the knowledge to even notice the issue. If we take the charitable view, the model is in a limbo state. More balaced view is that it's a non starter. Other implications are left as an exercise for the reader.

Second, practical implications. Sintel has shown that harmonic entropy loses on the correlation with empirical data to simpler models, in particular Wilson height, which is in the realm of mental math. What harmonic entropy does for us is inferior fit for substantially costlier computation. In simple terms, there is no practical value.

Third, it's not the end of trouble for harmonic entropy as there are fundamental problems that face any instance of computational consonance measure of intervals. In partucular any such measure that does not take actual, absolute frequency of the pitches making the interval, it mises how that the effect of width of interval on consonance is dependent on actual frequecy. In general, given how contextual consonance is, the mathematical consonance measures can either provide explanatory power or are just rules of thumb. Rules of thumb require simplitcity, by definition.

Harmonic entropy is a failure on both accounts. So far, on the former, utter on the latter.

I'm not claiming that he is wrong or not valid. It's a project that so far has failed to yield any important results and that's fine. Progress is made one failure at a time. The issue here is tacit misrepresentation. People behave as if it was established scientific theory, while it's only an unproven hyphotesis. In practical terms that amounts to misinformation, as a lot of community members lack the knowledge to see that context, and the wrong idea of the status of this theory gets propagated.