diff --git a/.editorconfig b/.editorconfig
index fdafd76c3cf..1993143df5c 100644
--- a/.editorconfig
+++ b/.editorconfig
@@ -153,10 +153,8 @@ end_of_line = unset
 charset = unset
 insert_final_newline = unset
 
-# These are valid UTF-8 text but contain a few stray control characters that
-# editorconfig-checker's binary detection flags as non-utf-8.
-# configure is autoconf-generated and uses BEL (\007) as an awk field
-# separator; lib/external/ccmath is vendored third-party code; the two
-# r.drain READMEs are historical, pre-existing content.
-[{configure,raster/r.drain/README,scripts/r.drain/README,lib/external/ccmath/C01-matrix}]
+# configure is autoconf-generated and uses BEL (\007) as an awk
+# field separator, which editorconfig-checker's binary detection flags as
+# non-utf-8.
+[configure]
 charset = unset
diff --git a/lib/external/ccmath/C01-matrix b/lib/external/ccmath/C01-matrix
index 4cc90fa843b..65cd771c205 100644
--- a/lib/external/ccmath/C01-matrix
+++ b/lib/external/ccmath/C01-matrix
@@ -877,8 +877,8 @@ qrbd1
             updated by the computation
        v = pointer to store of n by n orthogonal matrix V updated
            by the computation
-       m = number of rows in the U1 matrix
-       n = size of the d and e arrays, number of columns in the U1
+       m = number of rows in the U1 matrix
+       n = size of the d and e arrays, number of columns in the U1
            matrix, and the number of rows and columns in the V matrix.
       return value: N = number of QR iterations required
 
diff --git a/raster/r.drain/README b/raster/r.drain/README
index f6d18654856..93c72dfc10c 100644
--- a/raster/r.drain/README
+++ b/raster/r.drain/README
@@ -21,7 +21,7 @@ r.d2 reproduces most of the behavior of r.drain.  It does not use the
 all of the flags defined for r.drain.  My first draft of the model
 didn't correctly handle the ew,ns and diagonal resolutions (as
 r.fill.dir does not).  I believe that I fixed that problem.  There are
-also some other minor behaviors that differ a little from r.drain.
+also some other minor behaviors that differ a little from r.drain.
 
 r.d2 runs pretty quickly on a small, clean dem, but it can take quite a
 while  to run on a large cell map - particularly one with very many
diff --git a/scripts/r.drain/README b/scripts/r.drain/README
index f6d18654856..93c72dfc10c 100644
--- a/scripts/r.drain/README
+++ b/scripts/r.drain/README
@@ -21,7 +21,7 @@ r.d2 reproduces most of the behavior of r.drain.  It does not use the
 all of the flags defined for r.drain.  My first draft of the model
 didn't correctly handle the ew,ns and diagonal resolutions (as
 r.fill.dir does not).  I believe that I fixed that problem.  There are
-also some other minor behaviors that differ a little from r.drain.
+also some other minor behaviors that differ a little from r.drain.
 
 r.d2 runs pretty quickly on a small, clean dem, but it can take quite a
 while  to run on a large cell map - particularly one with very many