I've noted that in some places stanbol uses the platform default encoding and in other utf-8. For example in
http://svn.apache.org/viewvc/incubator/stanbol/trunk/enhancer/engines/autotagging/src/test/java/org/apache/stanbol/enhancer/engines/autotagging/impl/TestEntityMentionEnhancementEngine.java?revision=1089016&view=markup default encoding is used on line 105 while line 125 uses UTF-8. If there is not rationale in this I would suggest to use UTF-8 everywhere and ignore the default encoding of the VM. Cheers, Reto
