Even worse, the Hutter challenge is based on just a 1GB fragment of (one particular snapshot of) English Wikipedia. Further highlighting what has been said else

Even worse, the Hutter challenge is based on just a 1GB fragment of (one particular snapshot of) English Wikipedia. Further highlighting what has been said elsewhere in this thread, that as far as compression goes this is basically just a variation on compression-via-giant-dictionary