Initial LonCapa Thoughts

A few years ago, Purdue University transitioned some math classes to a new instance of LonCapa. In the spring of 2017, I was assigned to adapt content for LonCapa as a half-time teaching position.

I have a complicated relationship with the LonCapa project. Philosophically, a free-as-in-free-speech alternative to commercial softwares such as WebAssign, which we still use for several classes, appeals to me. High student fees, corporate-owned and controlled content, and the frequent outages we experienced all made me hope there was a better way.

LonCapa, however, suffers from a number of issues that make it incredibly frustrating to work with. Poor documentation, coupled with a strange, locked-down old version of Perl, can make development cumbersome. (I just learned of the "allow" tag.) Frequent violations of best practice - or even on some occasions basic principles of web safety - make the dated interface seem like the least of our concerns.

The complete lack of any asynchronous functionality means that students and instructors have to frequently reload, which is especially notable on problems which use complicated javascript-based content.

However, I feel like the flexibility of a platform like this is worth the risks. Several other, similar softwares exist - I am particularly excited by edX, but haven't myself done much research - but even within the LonCapa framework, we can link together Geogebra Applets, external resources, and problem systems of our own design.

I am excited to see what I can make using the available technology, and in what ways we can use the new technology to teach students better or more efficiently.