Reality is that most people use Windows for the dive logging software, so writing it for Windows would benefit most people (short of an OS-independent solution).
However, if you want to do something that people will want, I suggest a different poll. One asking what features people would like in logging SW. Have you checked out the ones that are currently available? If you Google "dive logging software" you will see that there are dozens of them.
It sounds like a great personal project.
I'm going to consider writing it for multiple operating systems. If I didn't I'd write it just for the Mac. It has to benefit me first.
I like to take things a step at a time. First, get a poll going on which operating systems to support. Second, start looking at what features exist in current dive software. Next find out who likes what or if anyone has new features they'd like to see. After that start thinking about what technology to use to create the software with (based on the operating systems I'm going to support). There is no rush. No timeline. I'm not trying to get this thing to market. Budget is zero.