To launch a platform, you send a message to a data table. In return, you receive a reference to the analysis layer object. You can save this reference for later use. Let's take a look at an example. The message name identifies the platform. So in this example, the message name indicates the Bivariate platform. The arguments to the message cast columns into analysis roles and can specify platform options. So, the example casts the Strength column into the Y role and the Humidity column into the X role. These analysis roles are named arguments, so they can appear in any order. Also, the two examples shown here are equivalent. In both examples, the bivariate platform is launched by sending the message to the data table reference stored in the variable dt, and the object reference for the analysis layer of the Bivariate platform is stored in biv. In the first example, the Fit Line command is sent as a message to biv after the platform has already launched. But in the second example, the Fit Line command is an argument to the Bivariate message, so in this case it's included as part of the platform launch. But these two examples accomplish the same thing. In some scripts it might make more sense to do it one way, and in some scripts it might make more sense to do it the other way.
