Tangentially related, for the chart related to american gdp embedded in this article, even though it claims to be on a logarithmic scale, that doesn't mean it d

Tangentially related, for the chart related to american gdp embedded in this article, even though it claims to be on a logarithmic scale, that doesn't mean it doesn't have some sort of "banking to 45" applied. http://vis.stanford.edu/papers/banking http://vis.stanford.edu/papers/banking Paradoxically (to me at times, perhaps to you) banking to 45 is supposed to make the chart more accurately read. It does so by more apparent to the viewer where the relevant inflection points are (in this case, the great depression really sticks out visually). Theoretically, the great depression might not stick out as well if banking to 45 were not done (as I suspect may have been done here). Whether or not it turns out to be the case here, I suppose lots of the other curious questions in the article have similar answers. When dealing with qualitative measures using qualitative measures, I think even ethical people can forge a system where emergent fudging arises. I think we (the tech folk) need to sort of up our game in casual analysis from tools like excel (which in actuality relatively good) to something like www.anylogic.com. With systems dynamics https://en.wikipedia.org/wiki/System_dynamics https://en.wikipedia.org/wiki/System_dynamics software I think we have a tool that can better elucidate this curious phenomenon where we notice questions whose data "feels a little funny" but may turn out to actually have a demonstrable excuse/justifiable explanation for looking so.