Skip to content

Appendix-i-Principles.md

Latest commit

 

History

History
52 lines (38 loc) · 2.56 KB

Appendix-i-Principles.md

File metadata and controls

52 lines (38 loc) · 2.56 KB

Appendix I - Principles for the Simple Server

This server will need to function both now and in the future. Spinning it up once is half the work; the other half is keeping it up and getting it back up when it goes down. If you’re like me—someone who doesn’t work with computers everyday—there’s a good chance you’ll forget the tedious details if not some big ideas. By designing a server that works for the future you, it is more likely to be kept up for the long term. This is not to say don't experiment with your server. You should. But do it with the principles below in mind.

The primary goals of the RockPro64 Simple Server are:

  1. Straight-forward setup.
  2. Minimized complexity.
  3. Maintanable by your future self, assuming that the skills you learned during initial setup will have waned.

The KISS principle is also applicable:

The KISS principle is best exemplified by the story of Johnson handing a team of design engineers a handful of tools, with the challenge that the jet aircraft they were designing must be repairable by an average mechanic in the field under combat conditions with only these tools. Hence, the "stupid" refers to the relationship between the way things break and the sophistication available to repair them.

Also, following the Rule of Least Power: use the smallest software you need to get the job done.

Finally, the Simple Server is also based on the design principles for the Clock of the Long Now, which have been modified below:

Longevity: With occasional maintenance, the server should reasonably be expected to function for 10 years.

Maintainability: The server should be maintainable with minimal coding know-how.

Transparency: It should be possible to determine operational principles of the server and its code by close inspection.

Evolvability: It should be possible to improve the server with time.

Scalability: It should be possible to increase the storage capacity of the server. However, increasing I/O (Input/Output) capacity would likely necessitate a new design.

Longevity:

  • Go slow
  • Avoid heat
  • Avoid excessive fan speed and energy use
  • Stay clean
  • Stay dry
  • Expect power outages
  • Expect hardware failure
  • Expect attacks
  • Don't tempt hackers

Maintainability and transparency:

  • Use familiar hardware and code
  • Allow inspection
  • Rehearse backups and restarts
  • Make it easy to buy spare parts
  • Expect restarts
  • Include the manual

Scalability and Evolvabilty:

  • Make all parts similarly
  • Separate functions
  • Provide simple interfaces