Thank you, HE and specifically KCochran !
Could you now think of a method for allowing the testing of ports above the first thousand ?
I assume you're concerned over use of your resources & possible denial of service. However, the users have to be identified and known to the system, right ? We could have a form entry for a range of ports to scan, that would accept maximum 1024 (or pick your number) consecutive ports at a time. If you're paranoïd even

an additional validation system or 'captcha' could be added.
What do you all think of this modest proposal ?