The OP is a bit ambiguous about where they want the project to go. If the whole point is to just have 'fun' while learning something new, then there are many different directions you can go with no accountability except to yourself. Pick any area of personal interest and start prototyping. It can be really fun to create something that does something nothing else does or does some common thing much faster or better than other solutions. Even if no one else uses it, it can be rewarding in its own right (i.e. a hobby).
If, on the other hand, you want to actually build something that you hope other people will try, adopt, give feedback, and (heaven forbid) actually pay you something for your trouble...then that is hard. You have to worry about things like what operating system or browser it runs on. If you want the project to also be rewarding to many other people, then all the business processes come into play (finding market fit, raising capital, evangelizing it, etc.) which can often be not fun to a technical person.
I know from experience with my own project. Shameless plug: https://didgets.com/
If, on the other hand, you want to actually build something that you hope other people will try, adopt, give feedback, and (heaven forbid) actually pay you something for your trouble...then that is hard. You have to worry about things like what operating system or browser it runs on. If you want the project to also be rewarding to many other people, then all the business processes come into play (finding market fit, raising capital, evangelizing it, etc.) which can often be not fun to a technical person.
I know from experience with my own project. Shameless plug: https://didgets.com/