update man page

1 files changed, 6 insertions, 6 deletions

diff --git a/uxd.rst b/uxd.rst
index d4ad0b4..53a9789 100644
--- a/uxd.rst
+++ b/uxd.rst
@@ -99,9 +99,9 @@ as *K*, *M*, and *G* for power-of-10 based (e.g. *1K* is 1000 bytes).
 .. print number of bytes/chars/ascii/multibyte/bad sequences.
 
 -j
-  Java mode (aka MUTF-8). Identical to UTF-8 except it allows the
+  Java mode (aka MUTF-8). Identical to UTF-8 except the
   overlong **0xc0 0x80** encoding for codepoint U+0000 (aka NUL),
-  which normally would be considered an error.
+  is highlighted in purple and not counted as an error.
   This may be useful for looking at serialized data created by Java
   programs.
 
@@ -141,10 +141,9 @@ as *K*, *M*, and *G* for power-of-10 based (e.g. *1K* is 1000 bytes).
   highlighed in red.
 
 -r
-  Highlight multi-byte sequences in reverse video, in the hex
-  output. Ignored if **-m** given.
+  Don't highlight multi-byte sequences in reverse video.
 
-.. highlight multi-byte chars in reverse video.
+.. don't highlight multi-byte chars in reverse video.
 
 -s pos
   Seek in input before starting to dump. *pos* is bytes, not
@@ -187,7 +186,8 @@ as *K*, *M*, and *G* for power-of-10 based (e.g. *1K* is 1000 bytes).
 .. print version of uxd.
 
 -w
-  WTF-8 mode. Surrogates **U+D800** to **U+D8FF** will not be considered errors.
+  WTF-8 mode. Surrogates **U+D800** to **U+D8FF** will be highlighted in
+  purple and not counted as errors.
 
 .. WTF-8 mode (allow surrogates).