diff --git a/README.md b/README.md
index 8d2736c..b728d7e 100755
--- a/README.md
+++ b/README.md
@@ -31,14 +31,20 @@ Web-based, professional and scalable velocimetry analysis for operational river
 # What is LiveOpenRiverCam
 
 LiveOpenRiverCam allows you to run operational measurement stations that estimate river discharge from videos.
-It is meant for National or HydroMeteorological Societies (NHMS), hydropower authorities, waterboards, or service 
+It is meant for e.g. National HydroMeteorological Societies (NHMS), hydropower authorities, waterboards, or service 
 providers of such entities, who wish to establish their own services for such users. 
-It is meant to provide fully automated data streams from operational camera/water level feeds that record videos at set 
-time intervals. The platform can either receive raw videos which are then orchestrated for processing on cloud nodes 
-or receive already processed videos, which are processed on-site ("edge processing"). The processing is performed by 
-[NodeOpenRiverCam](https://github.com/localdevice/nodeorc), the node processing tool around OpenRiverCam.
-The processing methods read videos, selects frames, enhances features, orthorectifies and estimates surface velocity 
-and discharge using state-of-the-art velocimetry methods, all fully automated.
+LiveORC will provide you with the following functionalities:
+* A administration-style front end for managing sites, configuration, videos and time series.
+* Visualization of time series and video analyses.
+* Fully automated data streaming from operational camera/water level feeds in the field with "edge processing".
+* Per-video processing through "cloud processing". 
+* A very fast and easy start to this all, through convenient virtualization of services and a very easy to use set up 
+  script.
+
+Processing on sites ("edge processing") or on cloud nodes ("cloud processing") is performed by
+[NodeOpenRiverCam](https://github.com/localdevice/nodeorc), the node processing tool around OpenRiverCam. The 
+processing methods read videos, select frames, enhance features, orthorectify and estimates surface velocity 
+and discharge using state-of-the-art velocimetry methods.
 
 # Acknowledgements
 
@@ -385,11 +391,13 @@ to report on LiveORC. NodeORC makes ample use of the REST API for callbacks of r
 The REST API also allows you to develop your own applications on top of LiveORC. For instance, if you wish
 to build your own web interface around OpenRiverCam for a specific user or with a specific application in 
 mind this is in principle possible! The API documentation is also disclosed automatically when you start
-LiveORC. If you are on `localhost:8000`, you can find it by browsing to `http://localhost:8000/api/docs`.
-The api calls are available on `http://localhost:8000/api`.
+LiveORC. If you are on `localhost:8000`, you can find it by browsing to `http://localhost:8000/api/docs`. This gives 
+the documentation per API end point in . If you prefer a swagger layout, you can also browse to 
+`http://localhost:8000/api/docs_swagger`. The api calls are available on `http://localhost:8000/api`. Replace your 
+hostname by the one you have configured if necessary.
 
 # Recipes
-b
+
 Recipes describe from top to bottom how a video is treated. They describe in order:
 * the video - what frames to use, what the actual water level during the recording was and if frames must be stabilized
 * frame preprocessing and orthorectification - several preprocessing methods can be applied to enhance 
diff --git a/api/models/base.py b/api/models/base.py
index f4edc01..3029307 100644
--- a/api/models/base.py
+++ b/api/models/base.py
@@ -20,6 +20,7 @@ class Meta:
     def institute(self):
         return self.site.institute
 
+
 class BaseModelNoInstitute(models.Model):
 
     creator = models.ForeignKey(
@@ -32,7 +33,6 @@ class Meta:
         abstract = True
 
 
-
 class BaseInstituteModel(models.Model):
     """
     BaseModel with an additional stored field (instead of property) for retrieving the institute
diff --git a/api/models/time_series.py b/api/models/time_series.py
index 96b51fb..22a1921 100644
--- a/api/models/time_series.py
+++ b/api/models/time_series.py
@@ -60,7 +60,6 @@ def save(self, *args, **kwargs):
                 # video_closest.status = VideoStatus.QUEUE
                 video_closest.save()
 
-
     class Meta:
         indexes = [models.Index(fields=['site', 'timestamp'])]
         verbose_name_plural = "time series"
diff --git a/api/views/camera_config.py b/api/views/camera_config.py
index 0e2c666..6910764 100644
--- a/api/views/camera_config.py
+++ b/api/views/camera_config.py
@@ -128,8 +128,6 @@ def create_task(self, request, *args, **kwargs):
                 status=status.HTTP_400_BAD_REQUEST,
                 content_type="application/json"
             )
-
-
         task_form = get_task_form(instance, query_callbacks)
         record = TaskForm(
             id=task_form["id"],
diff --git a/api/views/device.py b/api/views/device.py
index f01cf85..2511dc7 100644
--- a/api/views/device.py
+++ b/api/views/device.py
@@ -18,6 +18,7 @@ def get_client_ip_address(request):
         ip_addr = req_headers.get('REMOTE_ADDR')
     return ip_addr
 
+
 class DeviceViewSet(BaseModelViewSet):
     """
     API endpoints that allows devices to be viewed or edited.
@@ -27,7 +28,6 @@ class DeviceViewSet(BaseModelViewSet):
     permission_classes = [permissions.IsAuthenticated]
     http_method_names = ["get", "post", "delete", "patch"]
     filter_backends = [DjangoFilterBackend]
-    filterset_fields = ['institute']
 
     def get_queryset(self):
         """