Another variation on a theme. All strings assumed to be ASCII strings. All strings do not have to be the same length. Run under Perl version 5.8.9.

c:\@Work\Perl\monks>perl -wMstrict -MData::Dump -le
"my $x = 'YAYBOOAAFUMAAZOT';
 my $y = 'YAYFOOAABARA';
 ;;
 my $d = $x ^ $y;
 dd 'diff string: ', \$d;
 ;;
 my @d_seqs;
 push @d_seqs, [ $-[0], $+[0] - $-[0] ] while $d =~ m{ [^\x00]+ }xmsg;
 dd 'diff subsequences [offset, len]: ', \@d_seqs;
 ;;
 for my $s ($x, $y) {
   printf qq{'$s' diff subseqs: };
   printf qq{'$_' } for
     map { $_->[0] < length($s) ? substr($s, $_->[0], $_->[1]) : () } 
+@d_seqs
     ;
   print qq{\n};
   }
"
("diff string: ", \"\0\0\0\4\0\0\0\0\4\24\37\0AZOT")

(
  "diff subsequences [offset, len]: ",
  [[3, 1], [8, 3], [12, 4]],
)

'YAYBOOAAFUMAAZOT' diff subseqs: 'B' 'FUM' 'AZOT'

'YAYFOOAABARA' diff subseqs: 'F' 'BAR'
[download]

See "Variables related to regular expressions" in perlvar for  @- @+ arrays. (Update: The  $-[0] $+[0] terms in the code above access elements of the  @- @+ arrays respectively.)