Good list, I disagree with the Ansible commentary, primarily in regards to a comparison to chef and eventual convergence etc. Not only is Ansible much easier to work with in this regard but it doesn't require anything other than ssh, so client failures are a thing of the past. Using ansible-pull one can pull from repos/branches etc and if you are writing your playbooks such that they fail gracefully or rollback using handlers you can be sure that the system is in whatever you state you intended based on behavior.
Hi Christopher, thanks for the comment.
Ansible is my tool of choice and it is just as capable (and nicer to use than chef) but my commentary comes from my experience of how these tools tend to be used in the field.
Very few use ansible-pull and of those that do I've seen no one runs it on a schedule like chef-client daemon.
If you're using chef-server & chef-client then you are guaranteed frequent converges.