That's the problem with desktop software though -- it's expensive to build and often, there's a solution already out there or will appear in the next release of

That's the problem with desktop software though -- it's expensive to build and often, there's a solution already out there or will appear in the next release of the OS. For instance in Activity Monitor, you can view every process simply by clicking the View menu, then "All Processes". It will even show them hierarchically. Which then leads to users often picking something that works and rarely switching unless an app fails on them somehow. Contrast this with the web, where we routinely "switch apps" because there's so little cost to install them, keep them up to date, commit to them and save our data in their format, etc. AV software is much less useful these days as systems ship more securely by default. Disk cleaners also have less utility as folks learn of free or built-in options, or buy larger drives than they need... About all I can say to sum up is, yes, pick your app carefully, but solve a problem you've researched and haven't found an easy solution to, and then ... see if you can build that into a business? :) It'd also be fun to build a meta-app: e.g. gamify learning about OS X, such as the activity monitor. Maybe it could be social, where you get points for being the first to get something set up amongst friends on the same wifi? That sounds like an incredibly niche market though...