Xuxen, the spelling checker for Basque, is a very effective tool to help people more confidence when they are writing in Basque. In fact, this program is one of the most powerful tools in the ongoing standardization of Basque. Today, Xuxen is publicly available from http://www.euskara.euskadi.net, where there have been more than 20,000 downloads. There are versions for Office, OpenOffice, Mozilla, PC, Mac, and also an online web service (http://www.xuxen.com).
The spelling checker is more complex than equivalent software for other languages, because most of those are based on recognizing each word in a list of possible words in the language. However, because of the rich morphology of Basque, it is difficult to define such a list, and consequently, possible morphological analyzes must be included.
Such a complex program was really difficult to implement computationally when Ixa group tried to face this problem twenty years ago; it would have been almost impossible if Kimmo Koskeinmi had not defined the Two Level Morphology some years before. Koskenniemi defined this technology to face the complex morphology of Finnish and other languages. We decided to reprogram it. After Karttunen and Beesley (www.fsmbook.com/) and Mans Hulden (http://code.google.com/p/foma/) provided new tools for improving our applications.
Next week (July 23-25, 2012) the 10th edition of the International Workshop on Finite State Methods and Natural Language Processing 2012 – FSMNLP 2012 will be held in Donostia – San Sebastian. Kimmo Koskeniemi will be with us and we will take advantage of its presence in Donostia to thank him personally for his contribution to the world of computational morphology, a contribution that has been so useful for computational development of Basque.