Well, it's not that bad - the side-effect from learning Prolog is knowing a bit of Prolog, which is definitely much more useful than being hurt in the foot...
No, seriously: Erlang shares some features with Prolog and learning Prolog lets you learn those features in isolation, without unnecessary distractions (like, say, processes and sending messages, or gen_server/gen_fsm/supervisors or rebar, applications, releases...). For example "=" not being an assignment, and pattern matching in general, starts making a lot of sense. Looping via tail-recursion (only), single assignment variables, and even some of Erlang syntax stops seeming intimidating quickly.
I think the best metaphor for this would be learning to swim via jumping into a deep water, nearly drowning but learning the most important parts (like don't try to breathe underwater) and then going to shallow water to leisurely practice swimming there. It's certainly not for everyone, but for me it sped up the Erlang learning process by many weeks, if not months. (And also, I learned a bit of Prolog along the way.)
No, seriously: Erlang shares some features with Prolog and learning Prolog lets you learn those features in isolation, without unnecessary distractions (like, say, processes and sending messages, or gen_server/gen_fsm/supervisors or rebar, applications, releases...). For example "=" not being an assignment, and pattern matching in general, starts making a lot of sense. Looping via tail-recursion (only), single assignment variables, and even some of Erlang syntax stops seeming intimidating quickly.
I think the best metaphor for this would be learning to swim via jumping into a deep water, nearly drowning but learning the most important parts (like don't try to breathe underwater) and then going to shallow water to leisurely practice swimming there. It's certainly not for everyone, but for me it sped up the Erlang learning process by many weeks, if not months. (And also, I learned a bit of Prolog along the way.)