diff --git a/README.md b/README.md
index f53422bf1ebc81fdb2be5502f6c5a15557340cad..fe075f12dbef8fdc9f002c0a2c0dd4473badf298 100644
--- a/README.md
+++ b/README.md
@@ -36,7 +36,7 @@ sudo docker run -it \
 This will take a few seconds, in which the container starts, xle parses the sentences and the results are written to a file. Afterwards, the container terminates.
 
 <details>
-<summary>Advanced tips</summary>
+<summary>Advanced information</summary>
 
 Note that the `xle` specified in the `docker run` command is the name of the tag you've given the image when building, _not_ the command to be executed (although under the hood the container will call a Linux command also called `xle`).
 </details>
@@ -72,6 +72,15 @@ This script is run inside the Docker container. It first creates an `xlerc` file
 
 It then starts the headless X server, waits to ensure it has properly started and then executes the `xle` command. `xle` will load `xlerc`, create a parser, parse the sentence and display windows with the parse results.
 
+<details>
+<summary>Advanced information</summary>
+
+There actually is a nice wrapper that starts an arbitrary command within a virtual X server environment taking care of everything for us. It is called [`xvfb-run`](https://manpages.ubuntu.com/manpages/trusty/man1/xvfb-run.1.html). We tried to use the wrapper in earlier versions of the image, but didn't manage to get it to work on the first try.
+
+Still, it might be interesting to look into that more in the future.
+</details>
+
+
 As last step, we take a screenshot of the displayed windows, which is the output of the container.
 
 :books: [How do I take a screen shot of my Xvfb buffer?][screenshot]