What About Biometrics?

I mentioned earlier how authentication hasn't traditionally served the user's interest. It was built for, and sold *to* the back ends, and it's resulted in significant security issues, forced us to deal with multiple authentication schemes, and created serious confusion.

This brings to mind universal authentication schemes. How can we embed authentication in the middle of these Web and enterprise applications in a standard way?

Biometrics is the obvious choice, since virtually all users have fingerprints. But how can we integrate fingerprints in a *standard* way so that Web and enterprise applications can take advantage of them? How can we build middleware in a standard way so that anybody could log into any system with a fingerprint, so that applications could attach access control to different resources based on the identity associated with the fingerprint?

When we answer these questions, we will have in our midst a standard for authentication that makes a far simpler demand of the user than any currently being made. We will also have quite an interesting way to boost the overall authentication level of the Internet. And I've read more than one study that suggests users would warm up to this biometric method more quickly than to others.

Iris scans, for instance, are possible, but a bit invasive and probably not ideal for a universal authentication scheme.

Voice recognition isn’t invasive necessarily, but if the user catches a cold, gets hoarse, or attempts to log in at a loud event, all reliability is lost.

Most laptops and phones come with built-in cameras today, so facial recognition might work, but that's reliable only if the software can distinguish between a still photograph and a real person, and since most standard webcams on laptops typically capture at relatively low resolutions, this is a problem, too.

Whatever our endgame, our opening gambit is clear: we need to drive competition between vendors, and we need to do this by driving standards, because vendors of different equipment (i.e., the application developers and the hardware manufacturers) will need a standard way for the applications to talk to the hardware.

It is important, when we’re thinking of these things, to simplify solutions, but with authentication involving biometrics, it's very important that we design the systems correctly, to not oversimplify solutions.

One of the things that creep people out about biometrics is this sense that, "Oh my God, some server on the Web is going to store all the fingerprints for all the people in the world. What’s next?"

Obviously, that's not a great idea, but what is a great idea is the continued fostering of efforts to come up with protocols that actually make it possible to recognize the same fingerprint in a strong cryptographic way without ever sending any of the actual fingerprint images anywhere.

Different entities can still add things on top if there is a need (e.g., adding PKI to enable signatures and encryption). I'm fine with that. If they want to add one-time passwords because they want to be sure that people don’t replay things -- I'm fine with that, too. But even without these extras, a minimal level of biometric authentication will, in my prediction, prove quite resilient.

And what a bad day for the bad guys that's going to be.

Recognized in the industry as the "inventor of SSL," Dr. Taher Elgamal led the SSL efforts at Netscape. He also wrote the SSL patent and promoted SSL as the Internet security standard within standard committees and the industry. Dr. Elgamal invented several industry and government standards in data security and digital signatures area, including the DSS government standard for digital signatures. In addition to serving on numerous corporate advisory boards, Dr. Elgamal is the Chief Security Officer at Axway, a global provider of multi-enterprise solutions and infrastructure. He holds a Ph.D. and M.S. in Computer Science from Stanford University. View more of his blog posts here.