bazel-remote VS bitkeeper

I did PoC experiments with compression, chunking, and IPFS here: https://github.com/SaveTheRbtz/bazel-cache

If you need a mature compression implementation for bazel I would recommend using recent bazel versions w/ gRPC-based bazel-remote: https://github.com/buchgr/bazel-remote

bazel nowadays supports end-to-end compression w/ `--experimental_remote_cache_compression`: https://github.com/bazelbuild/bazel/pull/14041

I would rather go with a standalone cache to utilize bazel inceremental builds. Take a look at bazel buildfarm or remote cache (https://github.com/buchgr/bazel-remote). Once you populate cache builds will be very fast.

I do run bazel-remote at my home server for remote cache, though.

This is not historically accurate, even in the narrow context of Git branch names. Git was initially written as a replacement for an old source control system named BitKeeper. BitKeeper had a concept of a “master repo” (the main remote repo); all other copies of the repo were “slave repos”. Example from the BitKeeper docs. Git inherited that naming convention.

> git repositories don't have "slave" branches, just a "master' as in template or origin

That isn't true - master in git comes from master in BitKeeper which did have slave replicas.

https://github.com/bitkeeper-scm/bitkeeper/blob/master/doc/H...

Git was originally based on Bitkeeper which uses the terminology in that way, for example: https://github.com/bitkeeper-scm/bitkeeper/blob/master/doc/HOWTO.ask

Git was created as a replacement for BitKeeper. BitKeeper does use the terminology master/slave. Thus, Git's use of master is in fact a reference to master/slave, not a master recording.