Replies

I haven’t looked at the details of Verlinde’s work but note that his formulation was able to better map the rates by which outer fringe stars orbit around the center of the galaxy.

But here are some things I would look for if I was reviewing:

1) First, any equation or any of its candidate forms can be ‘tuned’ to any curve shape by fitting coefficients with basic graph shapes (exponential, hyperbola, inverse, etc.).

2) Second, noting the first fact in 1) above, I would have the concern that Verlinde merely ‘fit a curve’ and called the coefficients out to represent something new when in fact, ....

... he could be replacing one fudge factor with another, calling it a breakthrough.

Remark: I am always leery when science gets in a rut it starts to lower the bar in a search for new ‘heroes’ to extol, because in doing so, a tabloid of attractiveness is created. For example, in the field of social media, look at the phenomenon of Zuckerberg.

3) Third, I would look at the ‘new’ gravitational composition that Verlinde posits and ask if it is fundamentally measurable and not simply calculated from something else that is directly measurable.

Remark: The positing of ‘dark matter’ eerily mimicks the positing of ‘ether’ in the 19th-century investigations of electro-magnetism. Verlinde’s new formulation may be just an exercise in changing the labels or adding a new variable or a composite variable to make an equation fit the data.

All of the above is also subject to overparameterization (over tuning) where a model fits data so well that as soon as new data appear, the model breaks down.

The key to understanding is to observe what are the directly measurable quantities and how they can be arranged in a model to predict where future data should be observed, and then testing the model against new data. If Verlinde is deleting the ‘dark matter’ variable and forming a composite ‘story’ variable borrowing gravitational effects here and mass effects there, then his new variable might tell a good story couched in a good data fit, but could and likely would be nothing more than a magic trick.

If we have only one fundamental breakthrough every century, we are doing great.
If we have only one fundamental breakthrough every 500 years, we are doing well.
If we have only one fundamental breakthrough every millennium, we can and must try to do better.

Science is hard, damned hard.

With the push to publish and the abundance of journals to publish in, and the constant neverending quest for funding, reviewers need to be careful to navigate between reporting and tabloid publicity.