Breaking Dev: Mobile Apps Must Die

by September 14, 2011

In his Mobile Apps Must Die presentation at Breaking Development in Nashville TN, Scott Jenson made the case fro moving beyond mobile applications and illustrated what a future with apps could be. Here’s my notes from his talk:

  • The history of mobile phones has been copying the desktop and then realizing it just doesn’t work right.
  • Apps are a holdover from the desktop.
  • Native vs. Web treadmill. We take for granted that native is better than Web. Once you make a few apps, you realize there’s a lot of porting pain. So you move to the Web where you quickly realize Web apps are not good enough. Then you go back again.

How People Think About Mobile

  • Fanboi: use a single platform and everything will be just fine. But at some point the paradigm changes
  • Uber Capitalist: if you have enough money, you just port and shut up. Even Google has admitted, it cannot port all apps to every platform.
  • Web Pragmatist: just let me get my job done.
  • Web Enthusiast: wants things to get better and works as a change agent.

Problems with Apps

  • Even though there is a nice integration of Web and native applications possible, people will still focus on native apps. Native apps end up sucking up all the oxygen. Think of it as App Myopia.
  • It is pretty clear things will move towards the Web and the APIs coming to the Web will give us more control.
  • Yet the best we can ever do on the Web is always about 2-3 years behind native applications. We shouldn’t aspire to be as good as native apps. We should aim to be better than them.
  • App management: Apps require too much user attention. The fact Apple had to invent the app folder indicates how bad it is. We are sitting around doing garbage collecting of apps and over time developing a high bar for what apps we’ll stick with/download.
  • Are we really going to have one app for every Website?
  • Apps put a lot of pressure on the user. They need to figure out which one they want, install it, and then actually use it (long term use is rare).
  • Once you realize a lot of problems are about information access –the app model doesn’t hold up very well.

Moving Beyond Apps

  • “On demand” apps is a generic idea. When you are in an area– can we have a discovery protocol for finding relevant content? Have a peek at what’s available around you and decide what you need.
  • Objects we want to interact with are ultimately more important than apps. Just in time interaction allows us to get information from things when we need them. Apps aren’t even close to supporting this model.
  • Break the browser ghetto. You have your hands tied behind your back.
  • Window parity: the browser is viewed as one app not as the apps running inside it. WebOS supported a window parity model but now it’s gone.
  • Background processing: Need this to manage any meaningful interactions.
  • Fingerprinting: Understanding more about the device you are running on.
  • Implement the Dam Spec!: Companies like Google & Apple are motivated to drag their feet on making the Web browser more capable.
  • An anti-phone could be based entirely on the mobile Web (created as Chrome or Opera). Think of it as Linux for mobile devices.
  • Discovery Service: how do you extract meaning from the digital information around you? We need ranking in addition to indexing.
  • A discovery service is the next ‘Google’
  • Just in time interaction won’t happen through apps.
  • Smart devices will overwhelm apps.