In our PVS-Studio team, we are very fond of measuring things. Besides money we earn and salaries and taxes we pay, we also like to measure the number of PVS-Studio downloads, installations, percent distribution across IDEs, and so on. Sometimes these measurements suggest very important conclusions. But at other times they just embarrass us. In this article, I'd like to tell you about one case of this kind and ask for help.
Among other parameters, we measure the following:
Now the statistics we've got for a certain reporting period (its exact length doesn't matter) are as follows:
These figures look very strange. Out of the total of 300 downloads, only 20% (32+37) of the users installed the tool, and almost all the users who installed it anew (36, which makes 10%) have used the tool long enough to see the trial renewal dialog.
The question is: what about the rest 80% of the downloads? Who downloads but never installs the tool? Internet bots? Humans downloading the analyzer "just in case"? Linux users?
What if we are having a VERY BIG PROBLEM right now we are not even aware of? Please let us know if you or someone you know has experienced any troubles with PVS-Studio installation. How would you explain such a huge gap between the figures in our statistics?
So once again: 80% of those who downloaded the PVS-Studio distribution haven't installed it. Or we haven't got the installation report messages for some reason. Why? If these mysterious strangers can read this, please respond! Write us what you think about it.