Well, just in terms of cosmetics, let me suggest that you place the subroutine definition outside the foreach loop -- it just makes more sense that way, and is less confusing.

As for the procedure, you say the input file contains (one or more?) line(s?) like "&name1=value1&name2=value2&...", and if that is true, can you say what this line of coding is doing, and why you're doing this:

($info) = split(/\    /,$i);
[download]

I'm not sure (anything containing strings of two or more white-space characters is visually ambiguous), but I think:

• The backslash is unnecessary
• You're only saving the part before the first white-space string into $info (is that your intention?)

As for your stated purpose (to "remove &"), that's not entirely clear either. If you just remove this character from the input, you'll get (one or more?) strings like "name1=value1name2=value2name3=value3...", and I don't think that's what you want.

Maybe you mean "how do I split the line at the "&" characters?" If so, you do it like this:

my %parhash;
for my $i (@data) {
    chomp $i;
    my @params = split( /\&/, $i );

    # and while you're at it, the "nameX=valueX" can be
    # stored in a hash like this:
    for ( @params ) {
        next unless ( /.=./ ); # make sure there's a name and a value
        my ($k,$v) = split(/=/);
        $parhash{$k} = $v;
    }
}

# print html header stuff, then:
print "user supplied data was:<br>\n";
for ( sort keys %parhash ) {
    print "$_ = $parhash{$_}<br>\n";
}
# print html trailer stuff
[download]