diff --git a/README.md b/README.md
index 88e34d2f925b8da6099874219db4ed43e3c0a473..8fbaa15be58360ba8c01b7ef18ceac025500ebcd 100644
--- a/README.md
+++ b/README.md
@@ -26,16 +26,25 @@ in the tutorials.
 
 ## Development Environments
 
-The basic idea is that the developer already has a Linux Ubuntu ROS installation 
-and then either checks out publicly available ROS package repos or
-uses Docker to get access to the software that is not publicly available.
+When you have installed 
 
-When you have a ROS machine, follow the following instructions to set up and test basic functionality:
+software according to the instructions above,
+you can follow the following instructions to set up and test basic 
+functionality.
+
+- [Log in to Docker Registry](https://gitlab.liu.se/lrs/devenv_docker_images) -- Just follow the instructions to log in using a Gitlab user/password or deploy key.  Do *not* download any images using `docker pull`.  If you can't access this page, ++++++++ what to do?  Ask us for access?
 
-- [Log in to Docker Registry](https://gitlab.liu.se/lrs/devenv_docker_images) -- Just follow the instructions to log in.  Do not download the image.
 - [Prepare Existing ROS Computer](doc/prepare_ros_computer.md)
+
 - [Using a Local ROS Installation with LRS Docker Modules](doc/local_ros_installation.md)
 
+
+The basic idea is that the developer already has a Linux Ubuntu ROS installation 
+and then either checks out publicly available ROS package repos or
+uses Docker to get access to the software that is not publicly available.
+
+When you have a ROS machine, 
+
 For more information on how to integrate your code we have split the information in different sub-areas:
 
 - [Development Environment: Gränsö](https://gitlab.liu.se/lrs/lrs_devenv_granso): Sensor simulation in Gränsö doe DJI, MQTT interface also to images
diff --git a/doc/install.md b/doc/install.md
index b53a98116eec626fdd152543f6338cabd51ad211..48af7b84568c7dbbe6b3db4a7a5453daed795b11 100644
--- a/doc/install.md
+++ b/doc/install.md
@@ -171,5 +171,5 @@ roslaunch lrs_wara_gazebo dji_to_gazebo.launch name:=dji0 with_laser:=true x:=5
 
 All files to run these tests should be available in lrs_wara_gazebo
 which is installed in wara_ws by the Playbook above. The first command
-starts Gazebo and the second command add a model of a DJI M100 to the
+starts Gazebo and the second command adds a model of a DJI M100 to the
 Gazebo world.