What about the more common naming convention of using slash-separated branch names, which Git inherently understands? Git stores branches in .git/refs so by using slash-separators you capitalize on Git's behavior to keep branches in directories together. You can read more at https://git-scm.com/docs/user-manual#how-git-stores-referenc... and "Git - Branches in a Nutshell" is one of the must-read references: https://git-scm.com/book/en/v2/Git-Branching-Branches-in-a-N...

I have yet to see the benefit of worrying about how your branches or organized behind the scenes.

As a user a/b/c is the same as a-b-c, with the exception that with the folder structure I can’t have a branch called a/b also, for technical reasons that aren’t clear to most users.

But what would any of that matter if all branch names are prepended with a unique ID? You’d never have 2 branches in the same sub folder.

Yep. The article seems to muddle up and conflate a bunch of poor practices with "microservices." All the "problems" given are really just poor decomposition of the system into non-discrete components. The "audit" solution is a bit kludgy too.

> poor decomposition of the system into non-discrete components.

You can tar an awful lot of microservices with that brush.

in the first place, the components should not be decomposition like this, But you can't control everyone , But if It happen and went to production, the only thing you can do it is make it better, that what I have done in many products (using auditor or arbitrator )