High level aspects of adding randomized content to your assignments
There are two main routes one could take to implement randomization, LaTeX and Sage.
LaTeX is a Turing-Complete language, which means it certainly can provide randomization. Indeed, I have actually written some useful randomization packages for LaTeX myself.
Unfortunately, although one could implement this randomization into the LaTeX code for Xronos, the problem comes in how Xronos is currently published .
Currently, Xronos is published using the Xake tool, which essentially translates the TeX files into html and javascript (among other things), and then pushes those files to the server. But this means that all the LaTeX content is run and translated into the webpage when the page is published. In other words, the only time LaTeX is run (and thus the only time any randomization at the LaTeX level that occurs) happens when an update is published to the web.
So, in order to get any kind of LaTeX-level randomization, you’d need to republish the assignment each time you wanted to randomize it. Moreover, this would only randomize the content served to the server, not to the students - in other words, every student would necessarily get the same randomized content, it wouldn’t be different for each student in any way.
The other option is to use the sagetex package for LaTeX and sage code to implement randomization. This has the benefit of having a much more powerful programming syntax and tools to enable much more complex mathematical manipulation. In order to use sage however, you would likely want to get a local installation of sage and sagetex in order to be able to debug content locally before publishing it to the web. You can read more about getting a local installation of sage and sagetex here - the details are beyond the scope of this document.
In the case of sage, the seed for all randomization is set behind the scenes at the server level. This is important for three reasons.
