Wouldn't it be more sensible to learn how to design security systems that are hack-proof than to learn how to identify security systems that are dumb? Isn't th

Wouldn't it be more sensible to learn how to design security systems that are hack-proof than to learn how to identify security systems that are dumb? Isn't this line inherently contradictory, how can a person make a system hack-proof ( as if it is possible ) if he doesn't know how to identify the vulnerabilities he could be introducing into it by doing something a particular way. Anyway the only good thing I got out of it was Personal Observations on the Reliability of the Space Shuttle by Richard Feynman(http://www.ranum.com/security/computer_security/editorials/dumb/feynman.html http://www.ranum.com/security/computer_security/editorials/d...), its a better read than this article.