It’s hard to think about computers without some notion of applications. You use apps whenever you turn on a computer, use your phone, or surf the web.

There’s an app for writing essays and documents, one for sending messages to friends, one for organizing and editing photos, one for watching movies, one for readings news, one for finding plane flights or hotels, one for planning and tracking projects, and many more. For each and every activity you might want to do, or get done, there’s an app for that.

When you think about it, though, whenever you are trying to solve a problem, you’re not really using a single app to do it, but several. Writing an essay includes researching, reading and sharing. Booking your travels includes checking your calendar, looking at maps, comparing schedules and prices, and talking to the people you want to visit. Most activities include plenty of other tools and information.

But if we want to combine apps, we are forced to do it manually. And most apps come with a fixed range of actions, speaking no common language. They can’t be extended, composed, or combined with other applications without considerable effort.

The limitations we impose on applications are really limitations on what we are able to do in the real world. Like a stovepipe, we’re stuck with app-centric thinking.

When we’re working on a project, what we really want to have is a way to get every object and every kind of information you need for it, and then do things with them, and to them. Each project needs to connect to different people, contacts, notes, documents, calendars, to-do lists, and so on. Wish lists, photos, or calendars might be relevant in the context of your vacation. To-do lists, documents, and emails might be relevant in the context of work projects. Most importantly, there’s almost no way to add both relevant content and context to our existing data, such as combining our calendar with our address book or flight schedules.

And if you think about it, most people work on more than one project on any given day. So we’re stuck in a situation where we build our workplace up, stay inside these separate applications that are not able to integrate, and then tear everything down to switch to a different project. Operating systems and phones don’t really have a notion of the state of your work.

We have been conditioned to think this is how computers work — to get a new app for every new thing we want to do. But what truly matters to people is to be able to solve their problems and complete their tasks. It’s not in any way essential that problems and tasks are handled in the context of an application. Things don’t have to be as they are. The rigid boundaries between applications don’t need to be there. Just imagine if every app could be seamlessly extended, composed, or combined with the functions of other apps in any way we wanted.

This isn’t some kind of utopia I’m talking about here — it’s what programmers do all the time with their own systems. But if you look at the usual interaction of people with computers, we don’t see any of that. These days, phones and computers are more like more a shape sorting toy, where only specific actions are possible: The round block will only go in the circular hole.

Programmers don’t get this for free either. They have to navigate technical restrictions, system limitations, and other hocus-pocus that has nothing to do with what they’re working on or trying to express.

I imagine the computer as a canvas you can control. A playground where you can leverage the whole system to help you work on your idea or solve your problem. We would need to think in terms of the kinds of data we need for a project, rather than which individual apps we need to use.

The current approach limits our ability to invent more powerful ways to think about the world.

We need to use our computer as a system, combining multiple tools rather than collecting single-use apps, so we become better at thinking and problem solving. We need to get better at collaborating, so we can solve our problems collectively, by sharing our ideas and creations with each other. And we need to think in terms of systems, so we can stop reinventing the wheel with each new app, and begin to raise our collective intelligence.