From time to time, you hear people talking about what the next "killer app" will be. In one of my books, I made several arguments about why the notion of a killer app is flawed. Here's the argument in a nutshell.

As computer professionals, it's easy to forget that we are building tools that help people do their jobs. Consider the telephone. The telephone is a tool. People don't have jobs that focus on the telephone. Although people may spend hours on the phone at their job, their jobs aren't the telephone. Their jobs may be telemarketing or tech support, which use the telephone, but the phone isn't the job in itself. It's just a tool.

I use the computer constantly, but my job isn't the computer. My job is writing and editing and blogging about computers. In my previous position, I worked as a software engineer, building software for telecom companies. Our clients didn't exist for the computer; our client's business was the telecom business, and the computer was a necessary evil. (That's why they contracted other companies to build their software for them; their business wasn't software, it was telecom.)

As the decades pass, the computer will become more and more invisible. We're already headed that way. When we meet new people and ask them for their email address, we don't get into a big discussion about how cool computers are and say, "Oh wow! You have email too!" The thrill of email itself is gone, and now we simply use it, just as the thrill of the telephone is long gone and we don't sit around talking about telephones and join telephone user groups, any more than we join refrigerator user groups or digital clock user groups.

In the 90's, people were trying to get rich by building the next killer app. Such applications were new, cutting-edge products that took advantage of every aspect of the latest technology. They were flashy and impressive, but didn't have much use for the majority of the world. (And even if they did, the majority of the people weren't about to drop what they were doing and adopt the technology. You have to first convince the people they need the technology so it'll go beyond the early adopters. Most people don't buy products due to the cool technology; they buy products they need.)

We saw tons of "killer apps" back in the dot-com boom of the 90's. How many of these are left? Almost none. The truth is, most of these killer apps were incredibly narrow in scope and targeted a niche market while making use of every latest "cool" technology. And the majority of them are gone due to lack of sales, which resulted in the software companies that created them going bankrupt and shutting down. Or they're gone because they got purchased by a bigger company and were later dropped. Sure, a few survived, but only a few.

As software developers, we should not try to build a killer app. Instead, we should strive towards tools that are so useful, simple, and invisible that the users don't notice them and can focus on their jobs, yet when asked, will say how much they like the application because it's easy to use. As I type this blog, where is my focus? My focus is on the topic of killer apps, not on the application I'm using (the web browser). I'm not paying attention to the menus and the buttons at the top of the window, which is a good thing; the browser isn't in-my-face with its GUI; the browser is invisible to me right now and not taking my attention away from my job. When I get something out of the refrigerator, I don't marvel over how cool the fridge is and what an amazing contraption the door's hinge is. Instead, I don't notice the fridge, as my focus is on the food, even if the fridge has some amazing, cool, high-tech device like an ice maker. (There was a time when ice makers were new and amazing and cool, after all.)

The refrigerator is invisible to us, and our focus is on what's inside. Similarly, software should be the same way. Such invisibility is the sign of a great application, not some silly "killer app."