That's a good point. (Neat routine, too.) But consider this:

Before coming here, and before CGI.pm matured, I was a notorious advocate of cgi-lib.pl. As a matter of fact, part of the reason that the Vars method in CGI.pm was added was to facilitate the transition of cgi-lib.pl users over to CGI.pm. In cgi-lib.pl you would do something like this:

require "cgi-lib.pl";
&ReadParse(*input); # store form values in the glob "input"

print &PrintHeader; # print HTML header
print "Name: ", $input{name}, "<br>\n";
[download]

This was very nice syntax and worked well for me for many moons. =) When the form parser would encounter multiple values for a single field name, it would string them together with the NUL character (by default), so you could do this:

print "First value: ", $input{valueList}, "<br>\n";
print "All values: ", 
      join("<br>\n", split(/\0/, $input{valueList}));
[download]

And it would do exactly what you'd expect it to do. Neat, huh? Well, conveniently enough, CGI.pm inherited this behaviour. But don't take my word for it:

#!/usr/bin/perl -w
# /~alakaboo/cgi-perl/multi.pl

use strict;
use CGI;

my $q = new CGI;
my %v = $q->Vars();

print $q->header('text/plain');

{
    local $" = "\n";

    my @tmp = split /\0/, $v{ar};
    print "@tmp";
}
[download]

Voila.