diff --git a/include/gsc.hrl b/include/gsc.hrl
index 65b27df..d644a70 100644
--- a/include/gsc.hrl
+++ b/include/gsc.hrl
@@ -23,14 +23,14 @@
      | qid        % Foo.Bar.baz
      | qcon       % Foo.Bar.Baz
      | tvar       % 'foo, 'foo_bar, '_'foo'_'bar'''
-     % kwds ops and punct are all collapsed by
+     % kwds ops and sep are all collapsed by
      % so_scan:scan down to eg {'contract', {420, 69}}
      % where {420, 69} is the source location
      % these are three different parsers
      | kwd        % contract, interface, payable, etc
      | op         % "=!<>+-*/:&|?~@^"
-     | punct      % ".." | oneof(",.;()[]{}")
-     % kwds and punct are kind of the same thing
+     | sep      % ".." | oneof(",.;()[]{}")
+     % kwds and sep are kind of the same thing
      % but i'll keep them separate now for my own sanity. ok
      % i guess op or symbol or whatever is fine.
      %
diff --git a/scratch/ast-gulp.erl b/scratch/ast-gulp.erl
index 6f66530..b05b583 100644
--- a/scratch/ast-gulp.erl
+++ b/scratch/ast-gulp.erl
@@ -131,7 +131,7 @@ slurp_ct_impls([#gsc_token{string = ":", type = op},
 slurp_ct_impls(_) ->
     reject.
 
-slurp_ct_impls2([#gsc_token{string = ",", type = punct},
+slurp_ct_impls2([#gsc_token{string = ",", type = sep},
                  #gsc_token{string = Con1, type = con}
                 | Rest],
                 Acc) ->
diff --git a/src/gsc_strmatch.erl b/src/gsc_strmatch.erl
index 02992a8..bfaee40 100644
--- a/src/gsc_strmatch.erl
+++ b/src/gsc_strmatch.erl
@@ -88,7 +88,7 @@
 -export([
     smr_sf_ws/0,
     smr_sf_op/0,
-    smr_sf_punct/0,
+    smr_sf_sep/0,
     smr_sf_id/0,
     smr_sf_con/0,
     smr_sf_qid/0,
@@ -175,7 +175,7 @@ match(Matcher, Source) ->
 % -export([
 %    smr_sf_ws/0,
 %    smr_sf_op/0,
-%    smr_sf_punct/0,
+%    smr_sf_sep/0,
 %    smr_sf_id/0,
 %    smr_sf_con/0,
 %    smr_sf_qid/0,
@@ -224,7 +224,7 @@ smr_sf_op() ->
 
 
 
--spec smr_sf_punct() -> string_matcher().
+-spec smr_sf_sep() -> string_matcher().
 % @doc
 % String matcher for parens/braces
 %
@@ -233,7 +233,7 @@ smr_sf_op() ->
 %       , {"\\.\\.|[,.;()\\[\\]{}]", symbol()}
 % @end
 
-smr_sf_punct() ->
+smr_sf_sep() ->
     M_DotDotOp = smr_string(".."),
     M_PunctChars = smr_oneofchars(",.;()[]{}"),
     smr_union([M_DotDotOp, M_PunctChars]).
diff --git a/src/gsc_tokens.erl b/src/gsc_tokens.erl
index 2c6ec56..d20a819 100644
--- a/src/gsc_tokens.erl
+++ b/src/gsc_tokens.erl
@@ -254,7 +254,7 @@ token_shapes_parse_order() ->
     lists:flatten([
         % comments and whitespace
         lcom, bcom, ws,
-        punct,
+        sep,
         % literals
         char, string, int16, int10, bytes,
         ak, ct, sg,
@@ -264,7 +264,7 @@ token_shapes_parse_order() ->
         % keywords need to be parsed ahead of ids
         kwd, id,
         con,
-        % ops [=, =>, >>], punctuation (parens/braces)
+        % ops [=, =>, >>], sepuation (parens/braces)
         op
     ]).
 
@@ -597,7 +597,7 @@ slurp_token_of_shape(ws, Pos, SrcStr) ->
                                    str = WS},
             {tokmatch, Token, Rest}
     end;
-% KEYWORDS, OPERATORS, PUNCTUATION: kwd, op, punct
+% KEYWORDS, OPERATORS, PUNCTUATION: kwd, op, sep
 %
 % all the kwds are valid ids, so we match as an id and then check if it's a
 % kwd
@@ -629,10 +629,10 @@ slurp_token_of_shape(op, Pos, SrcStr) ->
         no_strmatch ->
             no_tokmatch
     end;
-slurp_token_of_shape(punct, Pos, SrcStr) ->
-    case gsc_strmatch:match(gsc_strmatch:smr_sf_punct(), SrcStr) of
+slurp_token_of_shape(sep, Pos, SrcStr) ->
+    case gsc_strmatch:match(gsc_strmatch:smr_sf_sep(), SrcStr) of
         {strmatch, Str, Rest} ->
-            Token = #tk{shape = punct, pos = Pos, str = Str},
+            Token = #tk{shape = sep, pos = Pos, str = Str},
             {tokmatch, Token, Rest};
         no_strmatch ->
             no_tokmatch
diff --git a/src/gso_scan.erl b/src/gso_scan.erl
index 374c191..d414e78 100644
--- a/src/gso_scan.erl
+++ b/src/gso_scan.erl
@@ -299,7 +299,7 @@ pass_types() ->
     [lcom,      % ak_AB//           breaks out of id
      bcom,      % ak_AB/*           breaks out of id
      ws,        % ak_AB\t           breaks out of id
-     punct,     % ak_AB{            breaks out of id
+     sep,     % ak_AB{            breaks out of id
      string,    % ak_AB"            breaks out of id
      bytes,     % ak_AB#            breaks out of id
      ak,ct,sg,  % ak_ABak           [akctsg] all in base58 alphabet
@@ -335,13 +335,13 @@ to_so_token(#tk{shape = SfTokenType,
         %
         % {contract, {420, 69}}
         %-----------------------
-        % kwds ops and punct are all collapsed by
+        % kwds ops and sep are all collapsed by
         % so_scan:scan down to eg {'contract', {420, 69}}
         % where {420, 69} is the source location
         % these are three different parsers
         Sym when Sym =:= kwd;
                  Sym =:= op;
-                 Sym =:= punct ->
+                 Sym =:= sep ->
             Symbol = list_to_atom(SfTokenStr),
             {true, {Symbol, Pos}};
         %------------------------------------