Docs encourage futile usage of cache to share data between different jobs of the same pipeline
Quoting:
"cache is used to specify a list of files and directories which should be cached between jobs. You can only use paths that are within the project workspace."
"By default caching is enabled and shared between pipelines and jobs, starting from GitLab 9.0"
I am told the cache is local to a particular runner, and never shared between runners. With more than one runner this makes is largely useless for sharing data which is only valid within one pipeline. I.e. keyed on key: "$CI_COMMIT_SHA"
. In retrospect, I dare say this was not the intended use case, but the docs make it seem like it is. Even encourage that use case via "...directories which should be cached between jobs".
Artifacts support that use case, but to the uninitiated, they seem inappropriate because they are also "published" externally to the pipeline. Therefore a new reader may embark on a futile attempt to use cache:
instead.