diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
index c24ef6b90e2e01cc979b09f4b6f0f391edf92742..0e55d2c8ac55ba7887063052d54c37af8f1c08ff 100644
--- a/.gitlab-ci.yml
+++ b/.gitlab-ci.yml
@@ -1,7 +1,7 @@
 # gitlab-ci file for aedmi
 
 default:
-    image: rboman/waves-py3:2020.3
+    image: rboman/waves-py3:2023.0
 
 .global_tag: &global_tag_def
     tags:
diff --git a/README.md b/README.md
index 8a2690dc877407783b7fa448ccfed2fd89349033..e19004d0e17bb343f309abe42ce765b6df827bb6 100644
--- a/README.md
+++ b/README.md
@@ -1,20 +1,26 @@
 # aedmi
 Aeroelastic flutter solution based on dynamic modes interpolation  
-University of Liège, 2021
+University of Liège, 2021  
+:warning:  
+This code is a partial transcription of the method discribed in [this PhD thesis](https://hdl.handle.net/2268/245578) for calculating flutter using high-fidelity aerodynamics. It is given as a reference and is not intented for general purpose analysis and optimization. Users interested in practical flutter-constrained optimization problems should explore the following codes:
+- [OMFlut](https://gitlab.uliege.be/am-dept/omflut)
+- [SDPM](https://gitlab.uliege.be/am-dept/sdpm)
+- [PyPk](https://gitlab.uliege.be/am-dept/pypk)
+:warning:
 
 ## Features
 aedmi computes the solution of the aeroelastic equation `u/l * M * p^2 + K - 1/2*r*u^2 * Q(k) = 0`, where `M` and `K` are the modal mass and stiffness matrices of the structure, `Q` is the modal aerodynamic loads matrix, `u` and `r` are the freestream fluid velocity and density, and `l` is a reference length. `p` is a non-dimensional parameter defined as `p = (g+1i)*k`, where `g` is the true damping and k is the reduced frequency.  
 The flutter solution is obtained using a kind of reduced order modeling technique called dynamic mode interpolation. This technique was originally developed by Hüseyin Güner during his [doctoral thesis](http://hdl.handle.net/2268/245578). Practically, the mode shapes of the structure are first pre-computed. Then, the aerodynamic loads at some reference Mach numbers and reduced frequencies are pre-computed by imposing the motion for each structural mode shape. Finally, aedmi will compute the modal load matrix `Q` for each Mach number and reduced frequency, and solve the aeroelastic system by interpolating this matrix for any reduced frequency.
 
 ## Requirements
-aedmi needs a python 3 interpreter and its libraries, as well the `numpy`, and `scipy`. The `vtk` packages is also needed to read VTK formatted date, and the `matplotlib` package is optional (needed to save the graphical solution to disk).
+aedmi needs a Python3 interpreter and its libraries, as well the `numpy`, and `scipy`. The `vtk` packages is also needed to read VTK formatted data, and the `matplotlib` package is optional (needed to save the graphical solution to disk).
 
 ### Linux
 If you are using Linux, you can install python and the packages using Aptitude.
 ```bash
 sudo apt-get update
 sudo apt-get install python3-dev
-sudo apt-get install python3-numpy python3-scipy python3-vtk7 python3-matplotlib
+sudo apt-get install python3-numpy python3-scipy python3-vtk9 python3-matplotlib
 ```
 
 ### Windows and MacOS