diff --git a/README.md b/README.md
index ac0bcdfde74a55c512d464ddb3dc4d4d1cb7aacc..5014d9f159f919bd690327292afbb9a955bd188d 100644
--- a/README.md
+++ b/README.md
@@ -75,7 +75,7 @@ Examples of input and outputs are provided in `./data/examples/`
 
 Ziped folders of "unlized" XML files of the corpus are (or will be) available in the ./data folder.
 
-To convert a file use a command like the following (it takes a moment because the notebook is executed before saving) :
+To convert a file use a command like the following :
 ```
 python unlizeToNotebook.py input.xml output.ipynb
 ```
@@ -84,9 +84,12 @@ python unlizeToNotebook.py input.xml output.ipynb
 
 By default, the ipynb notebook uses a [unlTools](https://gitlab.tetras-libre.fr/unl/unlTools) jar executable to convert UNL texts to SVG graphs. 
 You can download the jar in the [release section of unlTools](https://gitlab.tetras-libre.fr/unl/unlTools/-/releases), it must be placed in the same directory as the notebook.
+This is the recommended way to execute an entire notebook.
 
 As an alternative, you can modify the second cell of the notebook so it uses the unl2rdf Webservice instead of a local jar.
 Modify `svg = unl2dot(unl, "unl2rdf-app-1.0-SNAPSHOT-jar-with-dependencies.jar")`
 in `svg = unl2dotWeb(unl)`
 as explained in the comments.
 
+Using the webservice is slow so executing all the notebook is not recommended.
+