ku is the macro-language code for a number of mutually unintelligible dialects ( http://lmp.ucla.edu/Profile.aspx?LangID=61&menu=004 ). There are two problems associated with its dialects. It can be written with latin or arabic characters depending on the dialect.
The "ku" value currently reads "Kurdî" and the appearance of it on interwiki links is problematic as it should be with both scripts.
Also "ckb" کوردی which reads "Kurdi" in Arabic script should be something like کوردی (سۆرانی) which would be "Kurdi (Sorani)" or just "سۆرانی" which would be "Sorani" http://en.wikipedia.org/wiki/Sorani
"diq" split from ku.wikipedia on 5 January 2007 and "ckb" spit from ku.wikipedia on 14 November 2010.