Breaking Development: Beyond Mobile, Beyond Web

by September 24, 2012

In his Beyond Mobile, Beyond Web talk at Breaking Development in Dallas, TX Scott Jenson talked about native apps, the Web, the future of mobile, and the role of just in time interactions in the Internet of things. Here's my notes from his talk:

  • Things will be getting faster, smaller, and cheaper. That’s easy to predict. The bigger picture changes are harder to determine.
  • Big ideas are usually belittled as insignificant but they are already here. We just don’t recognize them as such until they explode.
  • Default thinking: throughout the history of technology, people have applied what they were doing yesterday to the new thing of today. The history of mobile phones has been copying the desktop.
  • Mobile apps are not the only tool in the toolkit for developers and designers. Apps are a holdover from the desktop. There are a number of trends that highlight how we can move beyond apps on mobile.

Beyond Apps

  • App Glut: are we going to have an app for every store, product, or Web site we use? It’s not sustainable. What happens when you have too many apps? The user is becoming the bottleneck for apps. We now have to “garden our phones” by deleting and managing our apps.
  • The core rule of all user experience design: value must be greater than pain. As designers we think a lot about pain. A good designer is an insurance policy. If you get pain down to zero, value can go down to zero too –why not?
  • Size and cost reduction: the zombie apocalypse of smart devices is coming. How does an Internet of things impact us? We aren’t just going to end up with smart phones. We’re going to end with all kinds of devices.
  • Cheap devices will leverage other devices to for display and control. They are hopelessly dependent on other devices. Size and cost reductions are giving us to a lot more devices that require more apps.
  • App glut + size & cost reduction + reliance on other devices for screens = an issue for apps.
  • Yahoo! was a hierarchical list of functions on the Web, then Google introduced a indexing and ranking algorithm that brought relevant things to you as you needed them.
  • We need a new discovery protocol to enable us to locate and interact with things that are relevant to us in the real world. This service needs to be on demand.
  • Just in time interaction: come up to something use, and lose it. Apps are a bad solution for this: they are more pain than value in a lot of real-world situations.
  • Kuhn cycle: we start with normal science, research comes up and things don’t fit, then there’s a crisis because only part of a model doesn’t work, from there a new model is developed, and the loop repeats.

Beyond Web

  • What is the model crisis for us? Software. We buy it, install it, and then feel we need to reuse it. Instead we need to discover something, use it, and then forget.
  • The Web is designed for this but it is stuck in a browser ghetto. You have to open an app to run an app. To lower pain, we need to get rid of the browser altogether.
  • Native apps are awesome and will always be better than the Web. We need native apps to work together with the Web. It’s not native vs. Web.
  • A local discovery service needs to gather up things near me and put them into a cloud-based sorting mechanism to highlight what is valuable.
  • Micro-functionality: start with the obvious uses then open an conduit to a networked device.
  • Once you network something, you can use it across multiple devices and screens.
  • Native apps tie interactivity to one particular set of hardware. The Web’s value is to work everywhere.
  • The Web and the Internet are not the same thing. The internet is DNS, http, etc. it’s a system of technologies not business models.
  • We have to think in layers. Everything won’t talk to everything right away. Start with layers: discover something, interact with it, then later focus on communicate and organize things.
  • It takes on average 20 years for a technology to hit maturity.

Device Types

  • Bears: smart devices that can do a lot. They don’t like to play nice with each other and want to own everything themselves. All these want to play nice with the Internet of things but they all have their own proprietary systems.
  • Bats: cheap smart devices with some computation in them that they need to be interacted with.
  • Bees: worker bees are classic machine-to-machine interactions -thousands of sensors streaming information to a system. Production bees are digitally identified and can generate a history based on its interaction with ID readers over time. Interactive bees are digital sensors that you can engage with: walk up to one of them and get information or take action.
  • How can a just in time interaction system be built today? QR codes and NFC are the same basic flow but NFC lowers pain quite a bit. Bluetooth and WiFi can be used to collect a name or URL but then you can’t connect to the Network so maybe you need multiple devices.
  • WiFi concurrent direct: ability to stay connected and broadcast out an identifier as well. Discovery: a series of wireless protocols. Interact through HTML. Communicate through a RESTful protocol.
  • Apps prevent just-in-time interactions. We need a model shift to move from Software to Experience. A discovery service needs to break out of the “browser ghetto”. Everything is lying at our feet –we just need to act.