Microsoft Makes Passwords Easier AND Safer

HardOCP News

[H] News
Joined
Dec 31, 1969
Messages
0
Microsoft has developed a system of creating easy to remember passwords without making you more vulnerable to hackers.

Instead of enforcing complex passwords, as many organizations do, the new scheme makes sure than no more than a few users can have the same password, which has a similar overall effect on security. Further research from Microsoft also reveals why only some organizations insist on very complex passwords.
 
No I think the point is that the passwords themselves are monitored and unique so that it's more obviously when something it up.
 
so they are making same password to be limited to only a few accounts. So this will only lead to a few people getting hacked compared to millions
 
So I should allow “weak” passwords yet make it so only a predefined amount of users can use the same password? I'm confused how this is logical for security. Eventually a key word generator is going to pick up “weak” password regardless if 5 people or 5,000 people use the same password. My thought is If you want to be secure, disable the root or admin account. Then, enforce strong password and make the system lock down the account or block access after x amount of failed attempts. Again, this is not totally foolproof, but its way better than allowing weak password to begin with...
 
hmph.. yea that doesn't make alot of sense...

Retinal Scanner or Finger Printer scanner FTW!
 
Yeah, I didn't think the volume of successful hacks into a company is as much of a problem as the level of authority the first successful hack gets.
 
Ah, I see their justification: Complex rules make remembering passwords more difficult--resulting in lots of lockouts for legitimate users--so companies forgo complex rules entirely.

A better solution would be to stop pushing password expiration. Have a user pick a strong, single-signon password that is good for their career with the company. Even if it were 20 characters long, the user is typing it in every day; they will never forget it. If someone ever susses it out, that's ONE account you need to deal with.
 
hmph.. yea that doesn't make alot of sense...

Retinal Scanner or Finger Printer scanner FTW!

Fingerprint scanners. haha. Get some clear plastic and push it down on it. The imprint of the users fingerprint is left on the sensor in finger dirt. Cracked.
 
It sounds pretty stupid to me. If a site uses this, as a password becomes unavailable because the limit for its usage have been reached, you will know someone is using it because you can't sign up with it. You now know what someone's password is. You just have to figure out who.
 
Smartcard coupled with webcam. When you step away, it locks the PC. When you come back, you have to reswipe the smart card. Windows 8 is rumored to support that. Or face recognition.

I wonder if you could bypass it with a printout of someones face (2D image)?
 
Smartcard coupled with webcam. When you step away, it locks the PC. When you come back, you have to reswipe the smart card. Windows 8 is rumored to support that. Or face recognition.

I wonder if you could bypass it with a printout of someones face (2D image)?

Or a biometric device such as a fingerpring scanner. They make those, y'know. Nobody can steal your fingerprint.
 
I wonder if you could bypass it with a printout of someones face (2D image)?

They had those for vending machines. You had to have your face scanned to prove you were old enough to buy things. And people did exactly that. Toshiba laptops have face recognition.
 
Or a biometric device such as a fingerpring scanner. They make those, y'know. Nobody can steal your fingerprint.

Fingerprint scanners are extremely easy to fool. You leave your fingerprints on everything you touch, which can then be lifted and used to bypass the fingerprint scanner.

There's also the option of slicing someone's finger or hand off, that's a viable option for getting around a fingerprint scanner too.
 
Fingerprint scanners are extremely easy to fool. You leave your fingerprints on everything you touch, which can then be lifted and used to bypass the fingerprint scanner.

There's also the option of slicing someone's finger or hand off, that's a viable option for getting around a fingerprint scanner too.

What do you mean you lost your security clearance?

*Shows amputated hand*

...oh
 
It sounds pretty stupid to me. If a site uses this, as a password becomes unavailable because the limit for its usage have been reached, you will know someone is using it because you can't sign up with it. You now know what someone's password is. You just have to figure out who.

As the article says, it only works when you have millions of users. In the end you'll have to use brute force to find 3 names out of say 25 million members. Not impossible but still a difficult task.

A better solution would be to stop pushing password expiration. Have a user pick a strong, single-signon password that is good for their career with the company. Even if it were 20 characters long, the user is typing it in every day; they will never forget it. If someone ever susses it out, that's ONE account you need to deal with.

That would be the same as using one long password everywhere on the web. What happens if one site is compromised. Now they have access to everything else on the intranet. Never assume that anyone has safe data practices. For all you know some program could be holding them all in clear text, even if say could just check with Active Directory to authenticate. The only part I agree with is that expiring passwords and remembering multiple passwords back can sometimes do more harm than good, and still isn't fool proof.
 
How's this going to stop people who have the current list of common passwords? All passwords less common = All passwords the same commonness.

If 5 out of 50 people can use the password [H]appy then that means that 45 people can't use the password [H]appy. Then it becomes a game of probability for most likely of a word/character to use that. So if you know that only 5 can have the same password, you know that X minus 5 won't have that password, so you've eliminated that password for X minus 5 people.
 
As the article says, it only works when you have millions of users. In the end you'll have to use brute force to find 3 names out of say 25 million members. Not impossible but still a difficult task.
Name a company that has millions of users with intranet access.

That would be the same as using one long password everywhere on the web. What happens if one site is compromised. Now they have access to everything else on the intranet. Never assume that anyone has safe data practices. For all you know some program could be holding them all in clear text, even if say could just check with Active Directory to authenticate. The only part I agree with is that expiring passwords and remembering multiple passwords back can sometimes do more harm than good, and still isn't fool proof.
So you're against single-signon entirely? Businesses would grind to a halt.
  • As an end user, all password expiration would do is to cost me hours going to each of these servers, changing my password to match the new one.
  • As a hacker, all I need is access to a user's e-mail account and then I can take advantage of forgotten password help to get all new passwords anyway.
  • As an IT administrator, if an account is hacked I'd have to assume the infiltrator has access to all those systems, so what have I gained by enforcing separate credentials?
Again, it would be a user being compromised, not an entire site or intranet (or even a class of users, as Microsoft's limited-shared-password scheme would allow). Seems to me that locking a centralized and replicated user account would be much simpler than trying to determine that user's individual access to any affected systems. If a user suspects they've been compromised, they can change their own password, effectively locking out the hacker without fear of missing any accessible intranet applications--not taking into account any code the hacker snuck in, of course.
 
For those of you that didn't read the article, here's the better-quoted bit that tells you the thinking:
One way that system designers try to defeat dictionary attacks is by temporarily disabling an account when a wrong password is submitted more than a few times. This is called account lock-out, and not surprisingly, attackers have discovered a simple way to defeat the approach. Instead of guessing thousands or millions of passwords for a single account, attackers simply guess the most commonly used passwords for thousands, or even millions, of different accounts.

It honestly makes sense... But one has to wonder on a massive service such as Gmail, Hotmail, etc... How many "Your password has already been used by another user" type things would exist. I could just picture some end-users trying dozens of simple passwords and none of them working.
 
I think the bottom line here, is that passwords are old school verification of identity.

Complexity, Lockout, Frequency, etc... They all have their flaws, the concept of a password is flawed.

I haven't seen anything proposed that would work BETTER though with the same amount of convenience.
 
For those of you that didn't read the article, here's the better-quoted bit that tells you the thinking:


It honestly makes sense... But one has to wonder on a massive service such as Gmail, Hotmail, etc... How many "Your password has already been used by another user" type things would exist. I could just picture some end-users trying dozens of simple passwords and none of them working.

Eh, I have never come across already-used password errors anywhere. Only user names.
 
For those of you that didn't read the article, here's the better-quoted bit that tells you the thinking:


It honestly makes sense... But one has to wonder on a massive service such as Gmail, Hotmail, etc... How many "Your password has already been used by another user" type things would exist. I could just picture some end-users trying dozens of simple passwords and none of them working.

*Four hours later*

Your password is still not valid! Please choose another password!
 
*Four hours later*

Your password is still not valid! Please choose another password!

I don't think it would be that massive. Go through the links a couple of layers and read some of the PDF where some of the info into the thought process of this came from. It's more about getting rid of passwords that while compliant still really aren't secure, such as using p@ssword as a password, and how Twitter banned 390 of it's most used passwords. If your password truly is random and not based from a dictionary or something else that is simple the chances of you getting multiple invalid password prompts is very unlikely.
 
If your password truly is random and not based from a dictionary or something else that is simple the chances of you getting multiple invalid password prompts is very unlikely.

If your password is truly random than this entire philosophy doesn't apply to you anyway, does it???

The purpose is to make simple passwords safer by limiting them, so I can still see Grandma Sue trying to use "billy" as a password, told it doesn't work, then using "joey", "skittles", etc...

It's a good idea in theory but users won't use a system if you make it too hard for them.
 
So now you can create a script to try to change your own password, and get a large list of passwords from the site.

If use profiles are publicly available, there's your user list too!
 
Back
Top