diff --git a/docs/guide/purity.rst b/docs/guide/purity.rst
index e01ff38..bd20d4e 100644
--- a/docs/guide/purity.rst
+++ b/docs/guide/purity.rst
@@ -8,7 +8,7 @@ The major problem we have is that one message (underlying implementation: :class
 through more than one component, and at the same time. If you wanna be safe, you tend to :func:`copy.copy()` everything
 between two calls to two different components, but that's very expensive.
 
-Instead of that, we chosed the oposite: copies are never made, and you should not modify in place the inputs of your
+Instead, we chose the opposite: copies are never made, and you should not modify in place the inputs of your
 component before yielding them, and that mostly means that you want to recreate dicts and lists before yielding (or
 returning) them. Numeric values, strings and tuples being immutable in python, modifying a variable of one of those
 type will already return a different instance.
@@ -130,10 +130,10 @@ Now let's see how to do it correctly:
 
 I hear you think «Yeah, but if I create like millions of dicts ...».
 
-Let's say we chosed the oposite way and copy the dict outside the transformation (in fact, `it's what we did in bonobo's
+Let's say we chose the opposite way and copied the dict outside the transformation (in fact, `it's what we did in bonobo's
 ancestor <https://github.com/rdcli/rdc.etl/blob/dev/rdc/etl/io/__init__.py#L187>`_). This means you will also create the
-same number of dicts, the difference is that you won't even notice it. Also, it means that if you want to yield 1 million
-times the same dict, going "pure" makes it efficient (you'll just yield the same object 1 million times) while going "copy
+same number of dicts, the difference is that you won't even notice it. Also, it means that if you want to yield the same
+dict 1 million times , going "pure" makes it efficient (you'll just yield the same object 1 million times) while going "copy
 crazy" will create 1 million objects.
 
 Using dicts like this will create a lot of dicts, but also free them as soon as all the future components that take this dict
diff --git a/docs/reference/index.rst b/docs/reference/index.rst
index f255e22..cc3a36b 100644
--- a/docs/reference/index.rst
+++ b/docs/reference/index.rst
@@ -1,10 +1,14 @@
 References
 ==========
 
+Reference documents of all stable APIs and modules. If something is not here, please be careful about using it as it
+means that the api is not yet 1.0-proof.
+
 .. toctree::
     :maxdepth: 4
 
     api
     api_config
     commands
+    settings
     examples
diff --git a/docs/reference/settings.rst b/docs/reference/settings.rst
new file mode 100644
index 0000000..1b2ffea
--- /dev/null
+++ b/docs/reference/settings.rst
@@ -0,0 +1,56 @@
+Settings & Environment
+======================
+
+.. module:: bonobo.settings
+
+All settings that you can find in the :module:`bonobo.settings` module.
+
+Debug
+:::::
+
+:Purpose: Sets the debug mode, which is more verbose. Loglevel will be lowered to DEBUG instead of INFO.
+:Environment: `DEBUG`
+:Setting: `bonobo.settings.DEBUG`
+:Default: `False`
+
+Profile
+:::::::
+
+:Purpose: Sets profiling, which adds memory/cpu usage output. Not yet fully implemented. It is expected that setting
+          this to true will have a non-neglictible performance impact.
+:Environment: `PROFILE`
+:Setting: `bonobo.settings.PROFILE`
+:Default: `False`
+
+Quiet
+:::::
+
+:Purpose: Sets the quiet mode, which ask any output to be computer parsable. Formating will be removed, but it will
+          allow to use unix pipes, etc. Not yet fully implemented, few transformations already use it. Probably, it
+          should be the default on non-interactive terminals.
+:Environment: `QUIET`
+:Setting: `bonobo.settings.QUIET`
+:Default: `False`
+
+Logging Level
+:::::::::::::
+
+:Purpose: Sets the python minimum logging level.
+:Environment: `LOGGING_LEVEL`
+:Setting: `bonobo.settings.LOGGING_LEVEL`
+:Default: `DEBUG` if DEBUG is False, otherwise `INFO`
+:Values: `CRITICAL`, `FATAL`, `ERROR`, `WARNING`, `INFO`, `DEBUG`, `NOTSET`
+
+I/O Format
+::::::::::
+
+:Purpose: Sets default input/output format for builtin transformations. It can be overriden on each node. The `kwargs`
+          value means that each node will try to read its input from keywords arguments (and write similar formated
+          output), while `arg0` means it will try to read its input from the first positional argument (and write
+          similar formated output).
+:Environment: `IOFORMAT`
+:Setting: `bonobo.settings.IOFORMAT`
+:Default: `kwargs`
+:Values: `kwargs`, `arg0`
+
+