4 ms·
Excess in password complexity policy is bullshit. I'm surprised Atwood would write knee jerk nonsense like this. If you follow the NIST guidance, it's not a pr
by Spooky23 10y ago
Excess in password complexity policy is bullshit. I'm surprised Atwood would write knee jerk nonsense like this.
If you follow the NIST guidance, it's not a problem at all. The bullshit happens when you allow infosec people who lack applied skills and go overboard in their analysis of policy and standards documents. Those infosec people wrongly see increasing entropy as equating to high assurance -- in reality increased complexity leads to ad hoc "something you have" tokens (i.e. My complex password is written on a post it note).
You need length and complexity standards because users don't know how to measure risk, don't understand what a good enough password is and don't really care. So people do stupid shit like make their password "qwerty". It puts their data and the integrity of your system at risk.
Disregarding the navelgazing about cultural insensitivity re: character sets, a 10 character length + 3/4 of upper/lower/numeric/symbol, combined with lockout controls ensures a reasonably high level of assurance. Again, per the NIST guidance, if you need more trust, you need multiple factors.