diff --git a/README.markdown b/README.markdown
index d291ba8..e350f47 100644
--- a/README.markdown
+++ b/README.markdown
@@ -41,22 +41,38 @@ Documentation
 TODO
 ----
 
-`byte buffer` and `pointer` types.  Basically, a `buffer` is a table that can
+### `byte buffer` and `pointer` types
+
+Basically, a `buffer` is a table that can
 be longer than 256 bytes, and a `pointer` is an address within a buffer.
 A `pointer` is implemented as a zero-page memory location, and accessing the
-buffer pointed to is implemented with indirect addressing.  We will likely
-need a new instruction for this, or at least a mode, and it will likely
-trash the `x` register, and it will likely be unchecked, at least to start.
-Basically, this is to allow us to write to the `byte buffer[2048]` known as
-"the screen".
+buffer pointed to is implemented with "indirect indexed" addressing, as in
 
-`word table` and `vector table` types.
+    LDA ($02), Y
 
-`low` and `high` address operators (turn `word` type into `byte`.)  Possibly.
+We will likely have a new mode of `copy` for this, like
 
-Save registers on stack or in memory (this preserves them = not trashed).
+    copy 100, p + y
 
-And at some point...
+where `p` is a user-defined storage location of `pointer` type, and `+ y`
+is mandatory (and you can/should set it to zero yourself if you want.)
+
+This instruction will likely be unchecked, at least to start.  Basically,
+this is to allow us to write to the `byte buffer[2048]` known as "the screen",
+(and doing that is valuable enough that we can sacrifice checking, for now.)
+
+### `word table` and `vector table` types
+
+### `low` and `high` address operators
+
+To turn `word` type into `byte`.
+
+### save registers on stack
+
+This preserves them, so semantically, they can be used even though they
+are trashed inside the block.
+
+### And at some point...
 
 *   initialized `byte table` memory locations
 *   always analyze before executing or compiling, unless told not to