The dominant theme of data protection discussions is one of preserving secrets. Much less attention gets paid to the problem of ensuring data's availability—not to mention verifiability—rather than limiting it. I spoke this month with representatives of Surety about their technology for addressing all four elements of affirmative data access management: confidentiality, authentication, integrity and nonrepudiation.

Surety describes its product as "trusted time-stamp service"—which is sort of like calling a Porsche Carrera "personal transportation." The company's AbsoluteProof service offers its own four-point list of key criteria for data assurance that people will actually use: "independent, portable, persistent and transparent." The service is independent of the party that has a stake in proving a data item's attributes; the authentication can travel with the data and be verified by the recipient; the verification is durable, and the use of the technology can be made invisible to the user.

For anyone who actually knows anything about encryption technology and practice, I hardly need to say anything more about Surety's ideas: The concept isn't nearly as hard to grasp as, say, string theory. In crypto, the critical thing these days is less often the concept than the implementation. As noted in eWEEK.com's widely viewed slide show, "The Dirty Dozen IT Embarrassments," even something as simple as correctly seeding a random number generator may be the difference between potential and actual crypto strength.

That's why I was struck, in particular, by one element of Surety's approach. The company periodically publishes a hash value, as plain old text in a classified ad in The New York Times, that reliably demonstrates the integrity of its entire process. This is a rare concrete example of the often vacuous phrase, "thinking outside the box"—which brings to most people's minds, I suspect, an image of climbing out of a metaphorical box of confined perceptions, but which most likely derives from the famous puzzle of the nine dots.

The paradox of most digital systems is that people try to use the system, or a parallel system with shared failure modes, to monitor and maintain function: a proposition that has a certain hall-of-mirrors paradoxical feel, and that leads to painful calculations of just how much safety is needed -- and how much it's worth to achieve that level of protection. Surety jumps out of that self-referential paradox by using another medium, one with completely different failure modes, as an outside check on its process. Well done.

Also helping developers to avoid getting boxed in by data assurance difficulties is PGP Corporation, with its announcement this month of a massive portfolio of updates to its encryption product line—with a general theme (it's always wrong to generalize) of making enterprise encryption more of a unified environment rather than a cobbled-up collection of separate tools and doctrines. There's not much more that I can say that's not addressed in my August interview with PGP director of product management John Dasher, but there's more from PGP at its site.

Tell me what kinds of assurance you wish you could get at peter_coffee@ziffdavis.com

Check out eWEEK.com's for the latest security news, reviews and analysis. And for insights on security coverage around the Web, take a look at eWEEK.com Security Center Editor Larry Seltzer's Weblog.