diff --git a/src/jsoniqmagic/magic.py b/src/jsoniqmagic/magic.py
index 46cd75e..777c444 100644
--- a/src/jsoniqmagic/magic.py
+++ b/src/jsoniqmagic/magic.py
@@ -73,8 +73,7 @@ def _execute(self, rumble, language, data, args, start):
         if (args.json or (not args.pandas_data_frame and not args.pyspark_data_frame)):
             capplusone = response.take(rumble.getRumbleConf().getInt("runtime.resultsSizeCap") + 1)
             if len(capplusone) > rumble.getRumbleConf().getInt("runtime.resultsSizeCap"):
-                count = response.count()
-                print("The query output %s items, which is too many to display. Displaying the first %s items:" % (count, rumble.getRumbleConf().getInt("runtime.resultsSizeCap")))
+                print("The query output too many items to display. Displaying the first %s items:" % (rumble.getRumbleConf().getInt("runtime.resultsSizeCap")))
             for e in capplusone[:rumble.getRumbleConf().getInt("runtime.resultsSizeCap")]:
                 print(json.dumps(json.loads(e.serializeAsJSON()), indent=2))
 
diff --git a/tests/test_notebook_errors.py b/tests/test_notebook_errors.py
index 049fb77..e28402b 100644
--- a/tests/test_notebook_errors.py
+++ b/tests/test_notebook_errors.py
@@ -76,7 +76,7 @@ def test_non_java_errors_are_unchanged():
 @pytest.mark.parametrize("stage,option", [
     ("query", "-j"), ("serialize", "-s"), ("df", "-df"),
     ("show", "-df"), ("pdf", "-pdf"), ("availableOutputs", "-u"),
-    ("applyPUL", "-u"), ("take", "-j"), ("count", "-j"),
+    ("applyPUL", "-u"), ("take", "-j"),
     ("serializeAsJSON", "-j"),
 ])
 def test_magic_uses_shared_error_policy(language, debug, stage, option):