Nearly seven years ago Jeremy Keith wrote an eloquent article on why passwords are an anti-pattern. His argument was predominantly around the need for OAuth and not entering passwords into third party sites to allow access to users data. We are through that argument now and for all its faults OAuth has been widely adopted.
But passwords are still an anti-pattern that deliver terrible experiences to users and a poor security model day after day.
Stuart and Lynne live in Australia and are retired. They use email heavily to communicate with their family and rely on their ISP to provide their email service. They use the same password for all of their services. It is a memorable word mixed up with some numbers. This is what they were advised to do. On Tuesday they wake up to some emails from friends that they have been hacked and that they have received spam emails from them. They are immediately very embarrassed and email all of their contacts to apologise. Later they start to feel very unsafe and concerned that the contents of their email have been viewable to a hacker. The word hacker means something close to criminal for them. They begin to feel scared of technology and question whether they will ever be safe.
From a customer experience the example is clearly terrible. A study from Ofcom suggests that Stuart and Lynne are in the majority in terms of using the same password for everything. The study notes that one in four use birthdays or names as passwords.
Do Stuart and Lynne care that they were the victim of a dictionary attack? No. Do they feel unsafe because of the authentication model they have to use? Yes.
In the physical world setting a password with a service provider would be the same as handing over the keys to your house to someone you don’t know. Every time you want to go into your house you need to visit this person and get the keys. When you are finished entering or leaving your house you have to hand the keys back. The security of everything in your house is the responsiblity of someone else.
How the third-party manages security around your keys is completely unknown. They might throw them in a box at the bottom of their garden or store them in a vault. But you are not able to know that. The third-party is also storing keys for other people. Occassionally people break in to the third-party and steal keys. It is then your responsibility to change the locks on your house, get new keys and give it back to the third-party.
The analogy is a little contrived but the point is that by setting a password and handing it over you have no way of knowing if the application provider is using plain text, or encrypting it with an insecure cipher to hash your password. You have no idea about the security environment around storing the password.
This might seem like complete insanity but in fact it is accepted behaviour on the web. Read any Terms of Service.
Google’s current Terms of Service are quite clear about passwords
To protect your Google Account, keep your password confidential. You are responsible for the activity that happens on or through your Google Account. Try not to reuse your Google Account password on third-party applications. If you learn of any unauthorised use of your password or Google Account, follow these instructions.
What this says is that you are responsible any breaches of the account and for anything that happens afterwards.
With certainty I can predict that about every other month a major service provider will have a security breach around passwords. A recent high-profile breach was Ebay where a statment noted that encrypted passwords had been compromised. They were quite clear in their statement about what had been stolen.
customers’ name, encrypted password, email address, physical address, phone number and date of birth
Today another site has announced a breach - Office a clothes retailer in the UK. The reaction of customers reflects how broken the current password model is
Just got a message from office shoes asking to change me password! These hackers are ruining the whole world god dayumn in!!!!!!
Whilst I respect the sentiment of @MissionD3 hackers are not ruining the world. The security model powering the web is the problem.
We have grown up with passwords on the web but it is time to admit that the model used by the majority of applications does not work. We are humans. We fail at remembering passwords so we make ones that are easy to remember and we use the same ones over and over again. We hand passwords over to a third-party willingly even though there are regular, predictable breaches.
The fact that power users resort to password managers or even use password reset to access their accounts is a symptom of the problem.
I don’t have the answer to where the web goes with security. Private keys, two-factor authentication would seem to be good starts. But the current model seems inherently broken and consistently delivers terrible experiences for users. Enough is enough - we need to rethink this.
Linux and Unix sha1sum command tutorial with examples
Tutorial on using sha1sum, a UNIX and Linux command to compute and check a SHA-1 message digest. Examples of reading a SHA-1 message digest, writing a SHA-1 message digest to a file, and checking a SHA-1 message digest.
Linux and Unix pwd command tutorial with examples
Tutorial on using pwd, a UNIX and Linux command for printing the name of the current working directory. Examples of printing the current working directory, avoiding symlinks and how to get the current working directory in shell scripts.
Configuring and working with Cloudfront Logs
Example of how to setup Cloudfront to log to S3, enable log rotation and how to download and work with combined Cloudfront log files.