JetTredmont said:
LagunaSol said:
The graph is relevant not because it shows total numbers, but because it shows rate of growth.
No, it shows absolute numbers, not rates of growth. However, one could take the same data and plot rates of growth on it.
Um, no. Are you looking at the first image? It shows numbers of subscribers (Y) as they increase over time (X). That's a growth curve (per Wikipedia, "A growth curve is an empirical model of the evolution of a quantity over time.").
Yes, that is a growth curve, and the slope of a growth curve is the rate of growth. However, you are NOT seeing a comparison of
rates of growth, or even a comparison of growth, but rather a comparison of absolute quantities (current size). The difference is that between the value of a line at a point - f(x) - and the slope of that line at a point relative to its value at that point - f'(x)/x.
If you want a Wikipedia reference, see
http://en.wikipedia.org/wiki/Population_growth_rate and skip half a page down; you'll see the f'(x)/x formula in all its glory.
In terms of rate of growth, adding 100 customers when you have 1000 is a much smaller rate of growth than adding 50 customers when you had 100.
Again, if you were comparing rate of growth you'd have things like x% growth from one month to the next. Apple's
rate of growth here is not overly impressive compared to the other folks (primarily because those folks have some legendary growth curves).
Another way of expressing growth (and IMHO a far more meaningful way) is as a
% of the overall market, which these graphs do not convey at all. I don't know how the numbers stack up there, but having lived through AOL and Netscape growth periods (less directly AOL) I have a pretty good idea that these guys gained a near monopoly on their respective markets very early and grew those markets incredibly fast. They grew their own subscriber base, in the first several months, by about the size of the entire market. Doesn't seem to be the case in Apple's case - they have grown the consumer smartphone market, but not like the WWW in 1994/5.
I think you're completely missing the point of the graph - in this case being a comparative tool.
Yes, but if you are comparing multiple lines in a single graph you need a comparable scale for the different plots. The simple fact is that a single subscriber in 1995 and in 1990 had significantly more value - both in terms of what that meant for market dominance and in strictly financial terms - than a single subscriber today.
Example: if I plot the number of people who stepped into the first Macy's store in 1929 to the number of people who stepped into the first Microsoft stores in 2009, the Microsoft store's numbers would look surprisingly like the iPhone numbers in that chart and the Macy's numbers would look surprisingly similar to the AOL/Netscape numbers. Yet, I think a rational observer would understand that Macy's growth rate is significantly more meaningful than Microsoft's.
I'm not saying Apple's position is not good here (and I do expect the iPhone to be around a lot longer than the Microsoft Retail experiment). I'm just saying that these charts are not showing it. They are simply comparing apples and oranges, and that's no way to present a case.
