web/project-website/content/usage/manager-configuration/_index.md

@@ -38,18 +38,17 @@ variables:
   blender:
     values:
       - platform: linux
-      value: /home/sybren/Downloads/blenders/blender-3.2.2-release/blender
+        value: blender
       - platform: windows
-      value: B:\Downloads\blenders\blender-3.2.2-release\blender.exe
+        value: blender
       - platform: darwin
-      value: /home/sybren/Downloads/blenders/blender-3.2.2-release/blender
+        value: blender
   blenderArgs:
     values:
       - platform: all
         value: -b -y
 ```
 
-## Creating a Configuration File
 The usual way to create a configuration file is simply by starting Flamenco
 Manager. If there is no config file yet, it will start the setup assistant to
 create one. If for any reasons the setup assistant is not usable for you, you

web/project-website/content/usage/variables/_index.md

@@ -1,55 +1,74 @@
 ---
-title: Manager Configuration
-weight: 3
+title: Variables
+weight: 2
 ---
 
-Flamenco Manager reads its configuration from `flamenco-manager.yaml`, located
-next to the `flamenco-manager` executable. The previous chapters
-([Shared Storage][storage] and [Variables][variables]) also described parts of
-that configuration file.
+For managing default parameters for Blender and FFmpeg, as well as for
+mixed-platform render farms, Flamenco uses *variables*.
 
-[storage]: {{< ref "shared-storage" >}}
-[variables]: {{< ref "usage/variables/multi-platform" >}}
+Each variable consists of:
 
-## Example Configuration
+- Its **name**: just a sequence of letters, like `blender` or `whateveryouwant`.
+- Its **values**: per *platform* and *audience*, described below.
 
-This is an example `flamenco-manager.yaml` file:
+The variables are configured in `flamenco-manager.yaml`.
+
+Here is an example `blender` variable:
 
 ```yaml
-_meta:
-  version: 3
-manager_name: Flamenco Manager
-database: flamenco-manager.sqlite
-listen: :8080
-autodiscoverable: true
-local_manager_storage_path: ./flamenco-manager-storage
-shared_storage_path: /path/to/storage
-shaman:
-  enabled: true
-  garbageCollect:
-    period: 24h0m0s
-    maxAge: 744h0m0s
-    extraCheckoutPaths: []
-task_timeout: 10m0s
-worker_timeout: 1m0s
-blocklist_threshold: 3
-task_fail_after_softfail_count: 3
 variables:
   blender:
     values:
     - platform: linux
-        value: blender
+      value: /home/sybren/Downloads/blenders/blender-3.2.2-release/blender
     - platform: windows
-        value: blender
+      value: B:\Downloads\blenders\blender-3.2.2-release\blender.exe
     - platform: darwin
-        value: blender
-  blenderArgs:
-    values:
-      - platform: all
-        value: -b -y
+      value: /home/sybren/Downloads/blenders/blender-3.2.2-release/blender
 ```
 
-The usual way to create a configuration file is simply by starting Flamenco
-Manager. If there is no config file yet, it will start the setup assistant to
-create one. If for any reasons the setup assistant is not usable for you, you
-can use the above example to create `flamenco-manager.yaml` yourself.
+Whenever a Worker gets a task that contains `{blender}`, that'll be replaced by
+the appropriate value for that worker.
+
+## Platform
+
+**The goal of the variables system is to cater for different platforms.**
+Blender will very likely be installed in different locations on Windows and
+Linux. It might even require some different parameters for your farm, depending
+on the platform. The variables system allows you to configure this.
+
+The platform can be `windows`, `linux`, or `darwin` for macOS. Other platforms
+are also allowed, if you happen to use them in your farm.
+
+## Two-way Variables for Mixed-platform Farms
+
+Variables are not just used to point to specific things, like `blender` above.
+They can also tell Flamenco that the path `/media/shared/flamenco` on Linux is
+the same as `S:\flamenco` on Windows, or `/Volumes/shared/flamenco` on macOS.
+This is documented further in [Two-way Variables for Multi-Platform Support][two-way].
+
+## Custom Job Types
+
+This documentation section focuses on pre-existing variables, `blender` and
+`blenderArgs`. There is nothing special about these. Apart from being part of
+Flamenco's default configuration, that is. When you go the more advanced route
+of creating your own [custom job types][jobtypes] you're free to create your own
+set of variables to suit your needs.
+
+[jobtypes]: {{< ref "usage/job-types" >}}
+[two-way]: {{< ref "usage/variables/multi-platform" >}}
+
+## Advanced: Audience
+
+The audience of a value is who that value is for: `workers`, `users`, or `all`
+if there is no difference in value for workers and users.
+
+This is an advanced feature, and was introduced for in-the-cloud render farms.
+In such situations, the location where the workers store the rendered frames
+might be different from where users go to pick them up.
+
+- `all`: values that are used for all audiences. This is the default, and is
+  what's used in the above example (because there is no `audience` mentioned).
+- `users`: values are used when submitting jobs from Blender and showing them in
+  the web interface.
+- `workers`: values that are used when sending tasks to workers.