diff --git a/pokedex/db/load.py b/pokedex/db/load.py
index da3130f..c87304a 100644
--- a/pokedex/db/load.py
+++ b/pokedex/db/load.py
@@ -3,9 +3,11 @@ from __future__ import print_function
 
 import csv
 import fnmatch
+import io
 import os.path
 import sys
 
+import six
 import sqlalchemy.sql.util
 import sqlalchemy.types
 
@@ -196,16 +198,20 @@ def load(session, tables=[], directory=None, drop_tables=False, verbose=False, s
 
         try:
             csvpath = "%s/%s.csv" % (directory, table_name)
-            csvfile = open(csvpath, 'rb')
+            csvfile = open(csvpath, 'r')
         except IOError:
             # File doesn't exist; don't load anything!
             print_done('missing?')
             continue
 
-        csvsize = os.stat(csvpath).st_size
+        # XXX This is wrong for files with multi-line fields, but Python 3
+        # doesn't allow .tell() on a file that's currently being iterated
+        # (because the result is completely bogus).  Oh well.
+        csvsize = sum(1 for line in csvfile)
+        csvfile.seek(0)
 
         reader = csv.reader(csvfile, lineterminator='\n')
-        column_names = [unicode(column) for column in reader.next()]
+        column_names = [six.text_type(column) for column in next(reader)]
 
         if not safe and engine.dialect.name == 'postgresql':
             # Postgres' CSV dialect works with our data, if we mark the not-null
@@ -253,10 +259,12 @@ def load(session, tables=[], directory=None, drop_tables=False, verbose=False, s
             session.commit()
             new_rows[:] = []
 
-            progress = "%d%%" % (100 * csvfile.tell() // csvsize)
+            progress = "%d%%" % (100 * csvpos // csvsize)
             print_status(progress)
 
+        csvpos = 0
         for csvs in reader:
+            csvpos += 1
             row_data = {}
 
             for column_name, value in zip(column_names, csvs):
@@ -271,7 +279,7 @@ def load(session, tables=[], directory=None, drop_tables=False, verbose=False, s
                         value = False
                     else:
                         value = True
-                else:
+                elif isinstance(value, bytes):
                     # Otherwise, unflatten from bytes
                     value = value.decode('utf-8')
 
@@ -394,7 +402,7 @@ def dump(session, tables=[], directory=None, verbose=False, langs=None):
         else:
             filename = '%s/%s.csv' % (directory, table_name)
 
-        writer = csv.writer(open(filename, 'wb'), lineterminator='\n')
+        writer = csv.writer(io.open(filename, 'w', newline=''), lineterminator='\n')
 
         columns = [col.name for col in table.columns]
 
@@ -434,7 +442,7 @@ def dump(session, tables=[], directory=None, verbose=False, langs=None):
                     elif val == False:
                         val = '0'
                     else:
-                        val = unicode(val).encode('utf-8')
+                        val = six.text_type(val).encode('utf-8')
 
                     csvs.append(val)
 
diff --git a/pokedex/db/translations.py b/pokedex/db/translations.py
index 39dc07c..05a04fa 100755
--- a/pokedex/db/translations.py
+++ b/pokedex/db/translations.py
@@ -25,6 +25,7 @@ from __future__ import print_function
 
 import binascii
 import csv
+import io
 import os
 import re
 from collections import defaultdict
@@ -257,11 +258,11 @@ class Translations(object):
     def reader_for_class(self, cls, reader_class=csv.reader):
         tablename = cls.__table__.name
         csvpath = os.path.join(self.csv_directory, tablename + '.csv')
-        return reader_class(open(csvpath, 'rb'), lineterminator='\n')
+        return reader_class(open(csvpath, 'r'), lineterminator='\n')
 
     def writer_for_lang(self, lang):
         csvpath = os.path.join(self.translation_directory, '%s.csv' % lang)
-        return csv.writer(open(csvpath, 'wb'), lineterminator='\n')
+        return csv.writer(io.open(csvpath, 'w', newline=''), lineterminator='\n')
 
     def yield_source_messages(self, language_id=None):
         """Yield all messages from source CSV files
@@ -302,7 +303,7 @@ class Translations(object):
         """
         path = os.path.join(self.csv_directory, 'translations', '%s.csv' % lang)
         try:
-            file = open(path, 'rb')
+            file = open(path, 'r')
         except IOError:
             return ()
         return yield_translation_csv_messages(file)
@@ -353,11 +354,11 @@ class Translations(object):
                 count += 1
             if count > 1000:
                 for translation_class, key_data in everything.items():
-                    yield translation_class, key_data.values()
+                    yield translation_class, list(key_data.values())
                 count = 0
                 everything.clear()
         for translation_class, data_dict in everything.items():
-            yield translation_class, data_dict.values()
+            yield translation_class, list(data_dict.values())
 
 def group_by_object(stream):
     """Group stream by object
diff --git a/pokedex/main.py b/pokedex/main.py
index 180248a..a8171a0 100644
--- a/pokedex/main.py
+++ b/pokedex/main.py
@@ -21,7 +21,7 @@ def main(*argv):
     # XXX there must be a better way to get Unicode argv
     # XXX this doesn't work on Windows durp
     enc = sys.stdin.encoding or 'utf8'
-    args = [_.decode(enc) for _ in args]
+    args = [_.decode(enc) if isinstance(_, bytes) else _ for _ in args]
 
     # Find the command as a function in this file
     func = globals().get("command_%s" % command, None)