You can make a case for including 0,0 as a data point in some data sets, assuming you can measure it!
Density is defined as mass divided by volume. It is plausible that something with a mass of zero would have a density of zero; while it is difficult to imagine an object with a mass equal to zero, a photon would be an example that fits the bill. Therefore you could have a 0,0 in your data if you were using a list of densities that included photons. However, if you are only charting things that you actually measured, 0,0 probably is not going to appear on the list.