On Sun, Sep 13, 2015 at 12:14 PM, David Roberson <[email protected]> wrote:
But, when I detect curve fitting instead of basic understanding of the > underlying processes I get annoyed. > I think by "curve fitting" you might mean the ad hoc creation of an explanation together with a dogmatic insistence that it's final and correct. I see curve fitting as unavoidable. You have a bunch of messy data points, and you have to find some way to connect the dots. You do so in the best way you can, and you include some arbitrary constants and some fudge factors to keep things together. I see all explanations as provisional in this manner. An explanation is good for a certain scope and domain of applicability, and then when you zoom in to some corner of the domain or zoom further out, it ceases to be helpful anymore. The question for me is whether an explanation provides insights that illumine something more complex that is going on below the surface, or whether it gets in the way of further exploration and understanding. For this reason I don't have a problem with curve fitting as such; just with dogmatic assertions about the validity of an explanation along with the shunning of people who try to look at things from a different angle. Eric
