The cryptographers and COMSEC engineers have given us an incredible number of fundamental security primitives. We now have good versions of essentially all the tools we know how to build at all. These tools are so good that attacks which are either impractical or entirely theoretical are nevertheless considered major successes. At the same time, the vast majority of traffic on the Internet is completely unprotected. These two phenomena are not unrelated. The flip side of the praise given for finding relatively small vulnerabilities is the massive amount of effort that developers feel they have to expend on fixing (and preventing) even quite small vulnerabilities. The inevitable result is that designers spend much more time enhancing security protocols than figuring out how to deploy them in real applications.
Gates 4B (opposite 490), 07/25/03, 4:00 PM