diff --git a/pyspark/Dockerfile b/pyspark/Dockerfile
index 5827bb7..3ad2140 100644
--- a/pyspark/Dockerfile
+++ b/pyspark/Dockerfile
@@ -1,13 +1,15 @@
 # Don't use the official jupyter/pyspark-notebook image because it's massive!
 FROM nstanger/spark
 
-ARG NB_USER="pyspark"
-ARG NB_UID="1000"
-ARG NB_GID="1000"
+# Ideally we should run as a non-root user, but it's problematic to set
+# up shared files, especially if running Docker within, say, VirtualBox.
+# ARG NB_USER="pyspark"
+# ARG NB_UID="1000"
+# ARG NB_GID="1000"
 
 USER root
 
-RUN adduser -D -u $NB_UID $NB_USER
+# RUN adduser -D -G root $NB_USER
 
 RUN apk add --no-cache --virtual .build-deps \
         build-base \
@@ -41,26 +43,26 @@
 
 # ENV SPARKMONITOR_UI_PORT 8080
 
-USER $NB_UID
+# USER $NB_USER
 
-RUN ipython kernel install --user --name $KERNEL_NAME
+RUN ipython kernel install --name $KERNEL_NAME
 # kernel.json is based on the template from <https://github.com/Anchormen/pyspark-jupyter-kernels>.
 COPY kernel.json pyspark-kernel.sh $KERNELS_DIR_PATH/$KERNEL_NAME/
 
-RUN jupyter nbextension install sparkmonitor --py --user --symlink \
-    && jupyter nbextension enable sparkmonitor --py --user \
-    && jupyter serverextension enable --py --user sparkmonitor \
+RUN jupyter nbextension install sparkmonitor --py --symlink \
+    && jupyter nbextension enable sparkmonitor --py \
+    && jupyter serverextension enable --py sparkmonitor \
     && ipython profile create \
     && echo "c.InteractiveShellApp.extensions.append('sparkmonitor.kernelextension')" >> $(ipython profile locate default)/ipython_kernel_config.py
 
-RUN mkdir -p /home/$NB_USER/work
-WORKDIR /home/$NB_USER/work
+# RUN mkdir -p /home/$NB_USER/work
+# WORKDIR /home/$NB_USER/work
 
 ENTRYPOINT ["/sbin/tini", "--"]
 
 # pyspark-kernel.sh will automatically set up the PySpark context when
 # the kernel is loaded.
-CMD ["jupyter", "notebook", "--ip=0.0.0.0", "--port=8888"]
+CMD ["jupyter", "notebook", "--ip=0.0.0.0", "--port=8888", "--allow-root"]
 
 # debugging
 # CMD ["bash"]