Software-Defined Delivery (SDD) Manifesto

Today I’m sharing the “Software-Defined Delivery Manifesto,” which was issued over a week ago and is still collecting signatures.

Today I’m sharing the “Software-Defined Delivery Manifesto,” which was issued over a week ago. The manifesto’s official website (https://sdd-manifesto.org/ - no longer maintained) is still collecting signatures.

Cloud-native, through immutable infrastructure and declarative configuration, serves as the foundation for software-defined delivery. Combined with continuous delivery tools, it can greatly improve software delivery efficiency. The manifesto’s drafters include many staunch advocates of cloud-native concepts like Kenny Bastani, Matt Stine, and others.

The manifesto was drafted through GitHub collaboration (https://github.com/sdd-manifesto/manifesto) and hasn’t reached version 1.0 yet. Below is the Chinese version of the “Software-Defined Delivery Manifesto.”

Software-Defined Delivery Manifesto

We recognize from daily production and practice that software shapes our world. We recognize that code is the best way to specify precise operations. We recognize that code is only useful when delivered.

Developed software can be delivered - this is the purpose of software existence. Now it’s time to apply our core skills to actual work. It’s time to design delivery. We will distinguish the roles humans and computers play in delivery work: humans make decisions, computers complete automated tasks.

Every delivery job is essentially unique. Applications, organizations, deployment environments, and team combinations vary widely. We recognize that each team needs to understand the uniqueness of this delivery and automate delivery. We recognize that while continuous delivery is crucial to meeting business needs, automating all repetitive tasks is equally important.

We use approaches similar to accelerating application development to accelerate software delivery: using modern architectures and programming languages, frameworks, libraries, and services for common functionality.

Delivery infrastructure is now programmable, and we will program it.

Software-Defined Delivery Means

Core: Delivery is every software team and organization’s foundational and strategic capability.

  • Priority: Delivery code is production code.
  • Strategic: Decide strategies at team and organizational levels; implement precise control in code.
  • Continuous Evolution: Continuously improve delivery.

Engineering Design: Powerful, testable code. Scripting languages from the 70s aren’t enough.

  • Modern Software Architecture: Event-driven and scalable.
  • Modern Programming Languages: Logic is best specified in code, not images or GUIs. Scripts don’t scale well.
  • Model-Based: Supported by models in the software domain, with understanding of code.
  • Testable: Enable small-scale application before production to discover errors.

Collaboration:

  • From the masses: Everyone can express their expertise in code, which benefits everyone.
  • Into software: Use the best tools, but combining them creates something unique.
  • Between humans and software: Collaborative automation can enhance our perception and help us make decisions. Putting information into action enables us to perceive software’s automated behavior. Distinguishing teams’ shared delivery goal sets from their implementation through code.

Acceleration:

  • Through automation: Automate repetitive tasks, speeding up work and avoiding errors.
  • Through reuse: Share common functionality among developers, teams, and organizations.

Observability: Usually used to observe and troubleshoot what happens during delivery as a production system.

  • Tracing: Observe activity in the system and track relationships between operations.
  • Debugging: Inspect and interact with the delivery process.
  • Metrics: Obtain metrics throughout the delivery process activities.

Authors (listed alphabetically by surname): This manifesto was drafted by Kenny Bastani, Marc Holmes, Rod Johnson, Jessica Kerr, Mik Kersten, Russ Miles, Erin Schnabel, Matt Stine, and other community members.

©2018, the above authors. This statement may be freely copied in any form in its entirety.

Jimmy Song

Jimmy Song

Focusing on research and open source practices in AI-Native Infrastructure and cloud native application architecture.

Post Navigation