liveBook · Manning

Native application

As explained in Fast by Default

An application compiled or integrated specifically for the operating system and hardware environment on which it runs.

A native application runs directly on a device operating system, with no abstraction layer between its code and the hardware. Native architecture gives platform-specific control over concurrency, lifecycle behavior, memory allocation, device capabilities, and SDK initialization. This control matters most where responsiveness or launch time affects retention. section 5.2.2

Installed desktop and mobile software carries a high expectation of immediate feedback. When responsiveness falls behind that expectation, users complain through support and reviews, revert versions, consider alternatives, and may force performance work ahead of planned features. Browser-based tools can still compete when rendering and collaboration responsiveness are strong, so native status alone is not a performance guarantee. section 1.2.3

Treat startup as an architecture decision rather than a fixed sequence of SDK work:

  • Inventory launch-time dependencies, including analytics, crash reporting, authentication, push notifications, and mapping services. section 5.2.2
  • Initialize lightweight, essential services synchronously when needed for the initial experience. section 5.2.2
  • Defer nonessential work until after the first screen, such as push registration, and load specialized SDKs only when their feature opens. section 5.2.2
  • Optimize for the devices target users own, not only flagship hardware, because slow startup is especially harmful on budget phones and slower networks. section 5.2.2

Launch-time SDK triage

An application accumulates analytics, crash reporting, authentication, push notifications, and a mapping SDK. Starting all of them at launch can add seconds. A native implementation can keep lightweight analytics and crash reporting synchronous, delay push registration until the first screen appears, and initialize mapping only when a map view is opened. section 5.2.2

Responsiveness as a release blocker

Typing delays of 100–500 ms made a messaging interface feel non-instant, generated complaints, and required performance regression work to take priority over new features. Native applications are judged by this visible feedback loop, not merely by their feature list. section 1.2.3

How Fast by Default covers it

How deep it goes
Explains platform control through startup SDK sequencing, memory and lifecycle access, without detailing lower-level implementation mechanics.
Examples
Uses messaging typing lag, browser rendering competition, and deferred analytics, notifications, and mapping SDK initialization.
What you can do after
Operate/configure native startup by choosing which SDKs initialize immediately, after first render, or only on feature use.
How it connects
Connects native applications with launch time, lazy initialization, memory management, concurrency, rendering performance, and user retention.

Also explained in

Compare how 2 books explain native application →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Fast by Default. The section links point to the parts of the book it draws on.