diff --git a/doc/np_diff.pdf b/doc/np_diff.pdf
new file mode 100644
index 0000000..1cf512b
Binary files /dev/null and b/doc/np_diff.pdf differ
diff --git a/src/clj/clj_diff/core.clj b/src/clj/clj_diff/core.clj
index 65612e9..b876b49 100644
--- a/src/clj/clj_diff/core.clj
+++ b/src/clj/clj_diff/core.clj
@@ -45,7 +45,7 @@
 
 (defmulti ^{:arglists '([s edit-script])} patch
   "Use the instructions in the edit script to transform the sequence s into
-  a new sequence. If the edit script was created by using diff on a and b then
+  a new sequence. If the edit script was created by using diff on a and b, then
   patch will use the edit script to transform a into b.
 
   (diff a b) -> x, (patch a x) -> b."
@@ -84,14 +84,14 @@
   needed to transform one string into the other, with the allowable edit
   operations being insertion, deletion and substitution of a single character.
 
-  This function works not only with strings but with any Clojure sequence.
+  This function works, not only with strings, but with any Clojure sequence.
 
-  Warning! Technically this function is estimating the Levenshtein distance
+  Warning! Technically, this function is estimating the Levenshtein distance
   from a computed diff. Most of the time, it is the same as the real Levenshtein
-  distance but in same cases it may be larger. The reason for this is that
-  there may be multiple paths through an edit graph with the same edit
-  distance but with differing Levenshtein distance. A future improvement to
-  the diff algorithm whould be to find all paths and prefer the one with the
+  distance, but in same cases it may be larger. The reason for this is that
+  there may be multiple paths through an edit graph, with the same edit
+  distance, but with differing Levenshtein distance. A future improvement to
+  the diff algorithm would be to find all paths, and prefer the one with the
   minimum Levenshtein distance."
   ([a b]
      (levenshtein-distance (diff a b)))