3 ms·

Of course. I'm not sure why anyone would think otherwise. Trying to create a competitive and compelling mobile experience today requires an ecosystem of apps, t

Of course. I'm not sure why anyone would think otherwise. Trying to create a competitive and compelling mobile experience today requires an ecosystem of apps, tooling and most importantly developers. You get all that with Android for free. Unless he's come up with some revolutionary technology that eschews the need to rely on the Android ecosystem he doesn't really have much choice if he wants consumers to buy his phone.

untoreh 10y ago
you can always offer a well refined android compatibility layer :p
- cuu508 10y ago
  The more refined apps (which are the ones customers care about) use a variety of Android platform services, not just the barebones stuff -- homescreen widgets, rich notifications, bundled notifications, app-defined quick settings, fingerprint authentication, Vulkan API, Cast support, Material themes, Play services etc. etc. etc. Building a complete compatibility layer means pretty much reimplementing all Android userspace.
  - akjainaj 10y ago
    You don't need to reimplement all Android userspace. You only need to implement 90% of what the top 20 apps use.
    - eli 10y ago
      90% of Android is a whole lotta code
      - akjainaj 10y ago
        Microsoft did something similar with the Astoria project, but they gave up on it, probably because of the Play Services dependency of most apps. And yes, you could enable it in Insider builds, and it mostly worked. Slap microG on it and you'll see it can be done.
erikpukinskis 10y ago
> Trying to create a competitive and compelling mobile experience today requires an ecosystem of apps, tooling and most importantly developers. I agree those things are all important considerations but you're lumping too many things together. Some of those are antithetical. For example, buying into an existing app ecosystem means you don't get your own developer mindshare. And tooling doesn't suffer from network effects the same way app stores and developers do. For example, they could use Android but write a new UI framework. They'd still make use of Android tooling, get partial entrée into the Android developer mindshare, but divest of the app base. Personally I don't think a phone startup can succeed without creating a new UI framework around a new paradigm, whether it's VR, social, AI, semantic Web, or whatever. Facebook had the right idea although they failed. If you are just competing for the same users with the same apps you will lose to Samsung. You need a new app base. Imagine a game console that only had multiplatform games. Who would buy it? A game platform with no third party games but a few compelling (to a niche) first party ones would have a better foundation.

untoreh 10y ago

you can always offer a well refined android compatibility layer :p

cuu508 10y ago
The more refined apps (which are the ones customers care about) use a variety of Android platform services, not just the barebones stuff -- homescreen widgets, rich notifications, bundled notifications, app-defined quick settings, fingerprint authentication, Vulkan API, Cast support, Material themes, Play services etc. etc. etc. Building a complete compatibility layer means pretty much reimplementing all Android userspace.
- akjainaj 10y ago
  You don't need to reimplement all Android userspace. You only need to implement 90% of what the top 20 apps use.
  - eli 10y ago
    90% of Android is a whole lotta code
    - akjainaj 10y ago
      Microsoft did something similar with the Astoria project, but they gave up on it, probably because of the Play Services dependency of most apps. And yes, you could enable it in Insider builds, and it mostly worked. Slap microG on it and you'll see it can be done.

erikpukinskis 10y ago

> Trying to create a competitive and compelling mobile experience today requires an ecosystem of apps, tooling and most importantly developers. I agree those things are all important considerations but you're lumping too many things together. Some of those are antithetical. For example, buying into an existing app ecosystem means you don't get your own developer mindshare. And tooling doesn't suffer from network effects the same way app stores and developers do. For example, they could use Android but write a new UI framework. They'd still make use of Android tooling, get partial entrée into the Android developer mindshare, but divest of the app base. Personally I don't think a phone startup can succeed without creating a new UI framework around a new paradigm, whether it's VR, social, AI, semantic Web, or whatever. Facebook had the right idea although they failed. If you are just competing for the same users with the same apps you will lose to Samsung. You need a new app base. Imagine a game console that only had multiplatform games. Who would buy it? A game platform with no third party games but a few compelling (to a niche) first party ones would have a better foundation.