When building Docker images, what is a recommended action for size optimization?

Building images from scratch when possible is recommended for size optimization because it allows for greater control over the contents of the image. By starting from a minimal base, you can include only the necessary files and dependencies needed for your application, reducing unnecessary bloat and optimizing the overall image size.

This approach is particularly effective because many base images include a variety of tools and libraries that may not be required for your specific application, which can lead to larger image sizes. Creating a custom image tailored specifically to your application's needs helps ensure that you include only what is necessary, thereby reducing the amount of storage space required and improving deployment times.

While complex commands and avoiding the examination of image layers can lead to inefficiencies, they do not directly contribute to size optimization. Engaging in continuous integration is beneficial for automating builds and tests but is not specifically aimed at reducing the size of Docker images.