The ~M and ~M` EWTS change had not fully been made. Someone submitted a bug report 911472 that alerted me to this.

source/org/thdl/tib/text/TibetanMachineWeb.java

@@ -1122,7 +1122,8 @@ public static boolean isWylieVowel(String s) {
 public static boolean isWylieAdornment(String wylie) {
     return (vowelSet.contains(wylie)
             || (wylie.equals("M") /* U+0F7E */
-                || wylie.equals("M^") /* U+0F83 */
+                || wylie.equals("~M") /* U+0F83 */
+                || wylie.equals("~M`") /* U+0F82 */
                 || wylie.equals("iM")
                 || wylie.equals("-iM")
                 || wylie.equals("eM")
@@ -1137,7 +1138,8 @@ public static boolean isWylieAdornment(String wylie) {
 public static boolean isWylieAdornmentAndContainsVowel(String wylie) {
     return (isWylieAdornment(wylie) &&
             !wylie.equals("M") /* U+0F7E */
-            && !wylie.equals("M^") /* U+0F83 */);
+            && !wylie.equals("~M") /* U+0F83 */
+            && !wylie.equals("~M`") /* U+0F82 */);
 }
 
 /**

source/org/thdl/tib/text/tcc_keyboard_1.ini

@@ -113,6 +113,6 @@ $=
 )=)
 H=:
 M=*
-^M=`
+~M=`
 // DLC FIXME: \tmw7095=__TILDE__
 &=$

source/org/thdl/tib/text/tcc_keyboard_2.ini

@@ -1,112 +1,112 @@
-Tibetan Computer Company Keyboard #2
-
-<?parameters?>
-has sanskrit stacking=true
-has tibetan stacking=true
-is stacking medial=false
-stack key=a
-has disambiguating key=false
-disambiguating key=
-needs achen before vowels=true
-has 'a' vowel=false
-is achung consonant=true
-
-<?characters?>
-k=q
-kh=w
-g=s
-ng=e
-c=b
-ch=n
-j=m
-ny=,
-t=o
-th=p
-d=j
-n=k
-p=r
-ph=/
-b=d
-m=f
-ts=;
-tsh='
-dz=[
-w=]
-zh=z
-z=x
-'=c
-y=g
-r=h
-l=v
-sh=.
-s=l
-h=G
-a=H
-T=O
-Th=P
-D=J
-N=K
-Sh=>
-0=0
-1=1
-2=2
-3=3
-4=4
-5=5
-6=6
-7=7
-8=8
-9=9
-<0=
-<1=
-<2=
-<3=
-<4=
-<5=
-<6=
-<7=
-<8=
-<9=
->0=
->1=
->2=
->3=
->4=
->5=
->6=
->7=
->8=
->9=
-
-<?vowels?>
-a=
-i=t
-u=u
-e=y
-o=i
-I=
-U=
-ai=Y
-au=I
-A=
--i=T
--I=
-
-<?punctuation?>
-_=
- = 
-/=\
-|=+
-!=|
-:=%
-;=
-@=!
-#=@
-$=
-%=
-(=(
-)=)
-H=:
-M=*
-`=`
-&=$
+Tibetan Computer Company Keyboard #2
+
+<?parameters?>
+has sanskrit stacking=true
+has tibetan stacking=true
+is stacking medial=false
+stack key=a
+has disambiguating key=false
+disambiguating key=
+needs achen before vowels=true
+has 'a' vowel=false
+is achung consonant=true
+
+<?characters?>
+k=q
+kh=w
+g=s
+ng=e
+c=b
+ch=n
+j=m
+ny=,
+t=o
+th=p
+d=j
+n=k
+p=r
+ph=/
+b=d
+m=f
+ts=;
+tsh='
+dz=[
+w=]
+zh=z
+z=x
+'=c
+y=g
+r=h
+l=v
+sh=.
+s=l
+h=G
+a=H
+T=O
+Th=P
+D=J
+N=K
+Sh=>
+0=0
+1=1
+2=2
+3=3
+4=4
+5=5
+6=6
+7=7
+8=8
+9=9
+<0=
+<1=
+<2=
+<3=
+<4=
+<5=
+<6=
+<7=
+<8=
+<9=
+>0=
+>1=
+>2=
+>3=
+>4=
+>5=
+>6=
+>7=
+>8=
+>9=
+
+<?vowels?>
+a=
+i=t
+u=u
+e=y
+o=i
+I=
+U=
+ai=Y
+au=I
+A=
+-i=T
+-I=
+
+<?punctuation?>
+_=
+ = 
+/=\
+|=+
+!=|
+:=%
+;=
+@=!
+#=@
+$=
+%=
+(=(
+)=)
+H=:
+M=*
+\u0F13=`
+&=$

source/org/thdl/tib/text/tshegbar/UnicodeCodepointToThdlWylie.java

@@ -218,8 +218,8 @@ public class UnicodeCodepointToThdlWylie {
 
         case '\u0F80': return "-i";
         case '\u0F81': return "-I";
-        case '\u0F82': return "\\u0F82";
-        case '\u0F83': return "^M"; // DLC unsupported in Jskad
+        case '\u0F82': return "~M`";
+        case '\u0F83': return "~M"; // DLC unsupported in Jskad, and 0F82 too probably
         case '\u0F84': return "?";
         case '\u0F85': return "&";
         case '\u0F86': return "\\u0F86";

source/org/thdl/tib/text/tshegbar/UnicodeReadingStateMachineConstants.java

@@ -41,7 +41,7 @@ interface UnicodeReadingStateMachineConstants {
     static final int CC_MCWD = 1;
 
     /** for combining marks in the Tibetan range of Unicode, minus
-     *  CC_MCWD, U+0F82, and U+0F39 */
+     *  CC_MCWD, U+0F82, (DLC U+0F83???) and U+0F39 */
     static final int CC_CM = 2;
 
     /** for combining consonants, that is to say U+0F90-U+0FBC minus
@@ -62,7 +62,7 @@ interface UnicodeReadingStateMachineConstants {
     static final int CC_0F8A = 6;
 
     /** for U+0F82, which is treated like {@link #CC_CM} except after
-     *  U+0F8A */
+     *  U+0F8A (DLC FIXME -- do we need similar treatment for 0F83???) */
     static final int CC_0F82 = 7;
 
     /** for U+0F39, an integral part of a consonant when it directly