Github user zjffdu commented on the issue: https://github.com/apache/zeppelin/pull/2709 @Leemoonsoo binary package is the same as before, spark/scala-2.11 and spark/scala-2.10 will be packaged together into `spark-interpreter-0.8.0-SNAPSHOT.jar` which doesn't contain scala library. scala library is in the spark distribution. So SparkInterpreter will load the correct scala version of SparkInterpreter (`SparkScala211Interpreter` or `SparkScala210Interpreter`) based the scala version in your spark distribution (depends on SPARK_HOME you set).
---