If you "deploy" to a server that's not running, then the module will be started next time the server starts. If you "distribute" to a server that's not running, then the module will not be started next time the server starts.
You mean, you hope that it starts - you would have to look at the server to see if it actually did. So this has the same "copy-and-pray" utility as dumping the module in a directory.
I think either you would be checking as you watched the server start (the --add option) or you would test by pinging the server once it had started. The latter option here is exactly as hard as just calling "start"
My issue is the ambiguity in the outcome of what you propose.
-- Jeremy
