That may be survivor's bias, though. How many software products end up not making it, at all? If you look at all software projects, everywhere, combined, and sp

That may be survivor's bias, though. How many software products end up not making it, at all? If you look at all software projects, everywhere, combined, and split them up: functional vs non-functional. The latter group would several orders of magnitude larger than the former. Now what percentage of the non-functional ones ends up becoming successful enough for you to hear about them and know what they were built with? Apply that percentage to the functional projects, you might just be left with naught. I have a feeling that if there's one thing more rare than functional programs, it's highly successful ones.