This patch for WWW::Mechanize stops these warnings from appearing when using
WWW::Myspace:

    Parsing of undecoded UTF-8 will give garbage when decoding entities at
     /usr/local/lib/perl/5.8.8/HTML/PullParser.pm line 83.

It is not necessary to use this patch, but you may wish to use it if you want
to stop those warnings from appearing.  It is provided for your convenience but
there is no guarantee it won't break functionality of other programs dependent
on WWW::Mechanize.  Use at your own risk!

It can be applied against an extracted copy of the WWW::Mechanize distribution
before testing and installing it, for example:

    tar -zxvf WWW-Mechanize-1.34.tar.gz
    cd WWW-Mechanize-1.34/
    patch -p1 < /path/to/WWW-Myspace/mechanize.patch
    make
    make test
    sudo make install

Alternatively, it can be applied to an already-installed copy of the module,
as long as you know where to find it.  If your all of your Perl modules are in
/usr/share/perl5 (as is the case for Debian-packaged Perl modules), do the
following:

    cd /usr/share/perl5
    sudo patch -p2 < /path/to/WWW-Myspace/mechanize.patch

If you don't know where the module is installed, you can do the following to
search for all installed versions of WWW::Mechanize and try to patch all of
them (at your own risk) :

    perl -e 'foreach $d (@INC) { print "$d\n" if ( $d =~ /^\// && -f "$d/WWW/Mechanize.pm" ); }' | while read d ; do cd "$d" && sudo patch --dry-run -p2 < /path/to/WWW-Myspace/mechanize.patch ; done

Note that this patch was created against WWW::Mechanize version 1.34.  The
patch may fail to apply against later versions.  Hopefully the problem will be
fixed in WWW::Mechanize itself and there soon won't be any need to patch it.

diff -Nru WWW-Mechanize-1.34.orig/lib/WWW/Mechanize.pm WWW-Mechanize-1.34/lib/WWW/Mechanize.pm
--- WWW-Mechanize-1.34.orig/lib/WWW/Mechanize.pm	2008-08-17 02:59:35.000000000 +0100
+++ WWW-Mechanize-1.34/lib/WWW/Mechanize.pm	2008-08-17 02:59:26.000000000 +0100
@@ -2146,9 +2146,7 @@
 
     # Try to decode the content. Undef will be returned if there's nothing to decompress.
     # See docs in HTTP::Message for details. Do we need to expose the options there? 
-    # use charset => 'none' because while we want LWP to handle Content-Encoding for 
-    # the auto-gzipping with Compress::Zlib we don't want it messing with charset
-    my $content = $res->decoded_content( charset => 'none' );
+    my $content = $res->decoded_content;
     $content = $res->content if (not defined $content);
 
     $content .= _taintedness();