A non-breakable space is, by definition, non-breakable :) - \s simply doesn't match. You can substitute it with plain spaces if you want. The following snippet will work with both decoded and encoded chunks of text:

#!/usr/bin/perl
use strict;
use warnings;
use HTML::Entities qw( decode_entities );
 
my $EncodedDate = "Tue Sep 13, 2005  10:38
pm";
my $MixedDate = $EncodedDate . "\n------\n" . decode_entities($Encoded
+Date);

# $nbsp will contain the decoded version of  
decode_entities(my $nbsp = ' ');

# Now, substitute all flavours of non-breakable-spaces
$MixedDate =~ s/ |$nbsp/ /g; 

# Same split as before
my @date = split /\s+/, $MixedDate;
print "$_ $date[$_]\n" foreach 0 .. $#date;

__END__

0 Tue
1 Sep
2 13,
3 2005
4 10:38
5 pm
6 ------
7 Tue
8 Sep
9 13,
10 2005
11 10:38
12 pm
[download]

You'd probably prefer to split on /[\s:,]+/, anyway: it will get rid of the comma after the "13", and will split "10:38" as well.