A good place to start isn't by solving someone else's problem, its solving your own. For example, I've noticed over the years that if a crowd of people is goin

A good place to start isn't by solving someone else's problem, its solving your own. For example, I've noticed over the years that if a crowd of people is going in or out of an entrance that has a set of doors, for some reason most people all crowd towards one door. To me, this is insane and a clear example of herd mentality. I solve my own problem by being cognizant of this fact and always look for other unused doors to go through. In another example, early this year I was working on optimizing the css for one of the sites at my company. I wanted a tool that would scan my site and give me a report of unused css selectors that were not being used anywhere. I look around for a while and the only tool I found that came close to what I wanted was a Firefox extension called Dust-me Selectors. Except that I found it didn't work well for shit. Also, it only ran in Firefox. Not very useful to me. So I spent a couple of days and wrote a pure javascript-based tool called Helium to do this. https://github.com/geuis/helium-css https://github.com/geuis/helium-css I can load this in my dev deployment of my site, pass it a list of urls I want to test, and let it run. Additionally, I can run this in multiple browsers. I solved my own problem, and in the process created something that other people can use to solve the same problem.