I kept running into the same issue working with discrete probability distributions: representing a potentially infinite set of outcomes, and in a way that holds up under transformation.

On paper, this is straightforward. You write something like \[\{1, 2, 3, \ldots\}\] and move on.

It wasn’t supposed to be a “-verse”. I just needed to be able to manipulate some probability distributions. But then three clearly different tasks needed special handling.

• I needed to define the distribution object in the first place. Out came distionary.
• Manipulation of these distributions could then be handled by distplyr.
• The need to fit and tune distributions to data quickly followed, and so did famish.

And, to tie them all together, probaverse. If you’re familiar with how the tidyverse package loads the core tidyverse packages, probaverse is similar: it’s a meta-package whose purpose is simply to load… well, the probaverse packages.