diff --git a/docs/deviceSpec.md b/docs/deviceSpec.md
index 8ad58d9..e22e10c 100644
--- a/docs/deviceSpec.md
+++ b/docs/deviceSpec.md
@@ -14,6 +14,8 @@ basically what conventional ML/LLM/ANN developers call an ROI ("Region of
 Interest") extraction algorithm. An Implex algorithm is used to scan a frame of
 input sensor data and detect objects and patterns within it.
 
+## Sensor device spec:
+
 The general format of a device-spec for a sensor is:
 ```
 implexor|api|server(server-params)|deviceselector
@@ -25,7 +27,9 @@ implexor|api|server(server-params)|deviceselector
   harikoff to read. Harikoff will run the `implexor` algorithm on the data from
   this `api`.
 * `server` may be a userspace daemon or an OS kernel that provides perceptual
-  data via the `api`.
+  data via the `api`. The `server-params` in parentheses may be omitted, in
+  which case the parenthesis will be empty, but the parentheses must always be
+  written out.
 * `device selector` is the idiosyncratic label/name used by the `server` to
   identify the specific device you want to access via that `server`.
 
@@ -76,7 +80,7 @@ to display `:0`. Use harikoff's `visual-implexor` to implex from display
 
 ### To attach a camera device by connecting directly to its Linux driver:
 ```
-visual-implexor|linux|v4l|/dev/video0
+visual-implexor|v4l|linux()|/dev/video0
 ```
 We specify that we want to use the `linux` kernel's loaded driver to connect
 to communicate with `/dev/video0`, via the `Video4Linux` API. We want harikoff