Basically Orbilius is a tool to search for words within a large and ever-expanding corpus of Latin texts of our time.
Like other computational linguistics applications, in the wake of WordNet, Orbilius searches the corpus for the concordances of each word occurrence. Or, to put it in simpler terms, it provides for each word searched, if present in the corpus, the translation and the context in which it is used; which means showing the real use of the language to facilitate learning.

The very choice of using texts written today, rather than those produced in the more than 20 centuries of Latin literature or in Latin, is not one of convenience or whim, but is thoughtfully motivated. As I wrote years ago in the appendix to the volume Dell'utilità di studiare il greco e il latino (id est The usefulness of studying Greek and Latin), I consider Latin not a dead language, but a language that too many people would like to see dead for a number of reasons.
Setting a corpus with only texts produced today is therefore a way to make Latin's vitality tangible, removing it from the mothballs in which the decline of humanities has put it.

In this spirit, as well as for the pleasure of writing some nice PHP code, I have written Orbilius, keep it up-to-date and make it freely available to those who want or need to study Latin.
I wish you a good studying.