Mozilla on Thursday said that its Persona Web authentication system, which eliminates the need to enter site-specific passwords at participating websites, has entered its beta phase and is ready for general deployment.
Passwords should have died long ago. No less than Microsoft chairman Bill Gates declared them dead in 2004. But as a Microsoft researcher noted last year, passwords still have a pulse.
"Despite countless attempts to dislodge them, passwords are more widely used and ﬁrmly entrenched than ever," said Cormac Herley, principal researcher in the machine learning department at Microsoft Research, and Paul C. van Oorschot, professor of computer science at Canada's Carleton University, in a research paper.
Were passwords to die, online security might improve. That's the theory anyway, if you assume that there's nowhere to go but up, that cybercriminals won't adapt, and that technology rather than humanity represents the weakest link in the chain. At the very least, the end of passwords would eliminate the possibility of having your password posted on pastebin.com following a data breach and having to explain to friends and colleagues why choosing "password" as your password seemed like a good idea at the time.
Like other single sign-on systems--and there are many--Persona promises to allow third-party websites to authenticate users without requiring them to ask for usernames and passwords.
"Instead of per-site passwords, Persona lets users log into sites with just two clicks after completing a simple, one-time process for each of their identities," Mozilla explains on its website. "This is safe, secure, and built on top of public key cryptography. Instead of a password, the user's browser generates a cryptographic 'identity assertion' that expires after a few minutes and is only valid on a single site."
Persona, says Mozilla, is easier to use than OpenID, another authentication system, because it is based on the user's email address rather than a URL generated by OpenID.
Email addresses have two distinct advantages. First, they're pseudonymous and thus provide more privacy than single sign-on systems used by Facebook and Google+, which require the use of real names. Second, they're subject to greater user control: Individuals can register and operate their own Internet domain, to more or less own their email identity. At services such as Gmail or Yahoo Mail, users cannot take their email address to another service provider if they're dissatisfied.
Persona's architecture also provides a privacy advantage. Whereas systems like OpenID require third-party websites to contact an authentication provider, Persona makes the user's browser the intermediary, passing credentials from the email provider to the third-party website. This exposes less website visit data to potential tracking.
There are benefits for businesses that adopt Persona, too. Mozilla's system provides developers with access to email addresses, enabling websites and app makers to contact their customers and eliminate the friction of soliciting an email address separately. Email addresses also play nicely with most existing login systems.
The problem with authentication systems is business buy-in. Online content and service providers might choose to integrate an authentication system like Persona, but they have tended to avoid doing so. A 2010 study conducted by the University of British Columbia's Laboratory for Education and Research in Secure Systems Engineering on the failure of single sign-on technology, A Billion Keys, but Few Locks: The Crisis of Web Single Sign-On, found that the among websites that could implement OpenID, the adoption rate was less than 0.02%.
The paper cites a number of reasons for disinterest in the technology: lack of business incentives, competitive concerns, usability, security, privacy, trust, and legal issues. The lack of business incentives for becoming a "relaying party"--that is, integrating an external authentication system--appears to be the biggest impediment to making single sign-on systems more appealing.
Although adopting a technology like Persona might have operational benefits, such as reducing password maintenance and recovery costs, doing so often doesn't have competitive benefits. Indeed, relying on an external authentication system might impose a competitive disadvantage if it deprives a company of potentially valuable user data or makes the company dependent on a potential competitor. The paper characterizes the situation as an "identity war" that "has been ongoing since the beginning of the Web" to build walls to keep customers and competitors apart. Imagine Google allowing users to sign in using a Facebook ID or vice versa and the problem becomes clearer.
However, Persona's association with Mozilla, a non-profit organization, might work to its advantage. Companies that might have balked at becoming reliant on a platform-based authentication system from the likes of hyper-competitive Facebook or Google could be more willing to trust Mozilla and its open Web ethos.
IDC analyst Sally Hudson characterized the issue as a balance among cost, convenience, and risk. "Consumers (or even corporate employees) will try to bypass or avoid using complex authentication mechanisms," she said in an email. "In a world where revenues are often driven by eyeballs and mouse clicks, anything that annoys or slows the consumer down in their quest to purchase a product or apply for a service is quickly discarded."