I wrote
a = sc.parallelize([1,2,3])
in spark-shell and got error
error: illegal start of simple expression
a = sc.parallelize([1,2,3])
^
but when I wrote this in PySpark, it worked.
What's the difference between the two?