I can't help but notice that all of those methods take a single command line string as a parameter. That's arguably fine on Windows, but not on Linux. So: how is that string broken down into the array of strings for the ultimate execve() call, in Ruby? Does Ruby parse it on its own (using some dialect of shell for quoting), or does it invoke the shell to do the heavy lifting, or?..
There is also a variant of 'system' which takes multiple agruments. It invokes the program in the first argument and passes the other arguments directly without going through shell so no globs etc. can take place. But this is not mentioned in the article but it is in Ruby docs
