diff --git a/README.md b/README.md
index db4649dc7..6106a2552 100644
--- a/README.md
+++ b/README.md
@@ -9,7 +9,7 @@ With Miller you get to use named fields without needing to count positional indi
 % mlr --csv filter '$status != "down"' mydata.csv
 ```
 
-This is something the Unix toolkit always could have done, and arguably always should have done.  It operates on key-value-pair data while the familiar Unix tools operate on integer-indexed fields: if the natural data structure for the latter is the array, then Miller's natural data structure is the insertion-ordered hash map.  This encompasses a **variety of data formats**, including but not limited to the familiar CSV.  (Miller can handle positionally-indexed data as a special case.) 
+This is something the Unix toolkit always could have done, and arguably always should have done.  It operates on **key-value-pair data** while the familiar Unix tools operate on integer-indexed fields: if the natural data structure for the latter is the array, then Miller's natural data structure is the insertion-ordered hash map.  This encompasses a **variety of data formats**, including but not limited to the familiar **CSV**.  (Miller can handle positionally-indexed data as a special case.) 
 
 Features:
 
diff --git a/c/todo.txt b/c/todo.txt
index 84398ca54..adaf2faf2 100644
--- a/c/todo.txt
+++ b/c/todo.txt
@@ -6,8 +6,10 @@ TOP OF LIST
 
 * make a -D for hash-collision stats ...
 
-* go through remaining functions to decide when null-through is ok.
-* also document thoroughly. emphasize this is crucial for heterogeneous data.
+* document nullability thoroughly: emphasize this is crucial for heterogeneous data.
+  - UTs cases for all
+  - note null-loses logic for min/max.
+  - separate doc section?
 
 * doc w/ very specific examples of sed/grep/etc preprocessing to structurize semi-structured data (e.g. logs)
 
@@ -15,7 +17,6 @@ TOP OF LIST
 
 !! booleans into put-dsl: should be able to do: mlr put '$ok = $x < 10'
 
-* min/max functions: f_ff. need to impl null-loses logic.
 
 * regularization mapper: when same names, reorder the same as 1st occurrence.
   e.g. one record w/ a,b,c & subsequent with a,c,b.