diff --git a/dawg/German/Makefile b/dawg/German/Makefile
index 62ec0be1e..b5e453d5a 100644
--- a/dawg/German/Makefile
+++ b/dawg/German/Makefile
@@ -1,5 +1,6 @@
-# -*- mode: makefile; coding: iso-8859-1; -*-
-# Copyright 2002 by Eric House (xwords@eehouse.org).  All rights reserved.
+# -*- mode: makefile; coding: utf-8; -*-
+# Copyright 2002 - 2010 by Eric House (xwords@eehouse.org).  All
+# rights reserved.
 #
 # This program is free software; you can redistribute it and/or
 # modify it under the terms of the GNU General Public License
@@ -17,7 +18,7 @@
 
 XWLANG = German
 LANGCODE = de_DE
-ENC = ISO-8859-1
+ENC = UTF-8
 
 TARGET_TYPE ?= WINCE
 
@@ -28,11 +29,11 @@ include ../Makefile.langcommon
 SOURCEDICT ?= $(XWDICTPATH)/German/HansGerman.dict.gz
 
 $(XWLANG)Main.dict.gz: $(SOURCEDICT) Makefile
-	zcat $< | tr [a-zäöü] [A-ZÄÖÜ] | \
-		LANG=$(LANGCODE):$(ENC) sed -e 's/ß/SS/g' | \
-		LANG=$(LANGCODE):$(ENC) grep '[AEIOUÄÖÜ]' | \
-		LANG=$(LANGCODE):$(ENC) grep '^[A-ZÄÖÜ]\+$$' | \
-		gzip -c > $@
+	zcat $< \
+		| tr [a-zäöü] [A-ZÄÖÜ] \
+		| sed -e 's/ß/SS/g' \
+		| grep '^[A-ZÄÖÜ]*$$' \
+		| gzip -c > $@
 
 # Everything but creating of the Main.dict file is inherited from the
 # "parent" Makefile.langcommon in the parent directory.
diff --git a/dawg/German/info.txt b/dawg/German/info.txt
index f6321981d..7c3ee588f 100644
--- a/dawg/German/info.txt
+++ b/dawg/German/info.txt
@@ -1,4 +1,6 @@
-# Copyright 2002 by Eric House (xwords@eehouse.org).  All rights reserved.
+# -*- mode: conf; coding: utf-8; -*-
+# Copyright 2002 - 2010 by Eric House (xwords@eehouse.org).  All
+# rights reserved.
 #
 # This program is free software; you can redistribute it and/or
 # modify it under the terms of the GNU General Public License
@@ -15,17 +17,18 @@
 # Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA  02111-1307, USA.
 
 LANGCODE:de_DE
+CHARSET: utf-8
 
 # deal with DOS files
 LANGFILTER: tr -d '\r'
 # substitute for sharfes-s
-LANGFILTER: | sed -e 's/ß/SS/g'
+LANGFILTER: | sed -e 's/ß/SS/g'
 # uppercase all
-LANGFILTER: | tr [a-zäöü] [A-ZÄÖÜ]
+LANGFILTER: | tr [a-zäöü] [A-ZÄÖÜ]
 # no words not containing a vowel
-LANGFILTER: | grep '[AEIOUÄÖÜ]' 
+LANGFILTER: | grep '[AEIOUÄÖÜ]' 
 # none with illegal chars
-LANGFILTER: | grep '^[A-ZÄÖÜ]\+$'
+LANGFILTER: | grep '^[A-ZÄÖÜ]\+$'
 
 # Until I can figure out how to force sort to use a locale's collation
 # rules we can't trust sort in the filtering rules above and so must
@@ -46,9 +49,8 @@ XLOC_HEADER:0x8300
 
 <BEGIN_TILES>
 2			0		{"_"}
-5		        1		'A'
-# A mit umlaut
-1			6		196
+5	        1		'A'
+1			6		'Ä'
 2			3		'B'
 2			4		'C'
 4			1		'D'
@@ -63,16 +65,14 @@ XLOC_HEADER:0x8300
 4			3		'M'
 9			1		'N'
 3			2		'O'
-# O mit umlaut
-1			8		214
+1			8		'Ö'
 1			4		'P'
 1			10		'Q'
 6			1		'R'
 7			1		'S'
 6			1		'T'
 6			1		'U'
-# U mit umlaut
-1			6		220
+1			6		'Ü'
 1			6		'V'
 1			3		'W'
 1			8		'X'