Securing software with automated functional diversity

对于大多数计算机用户来说，计算机安全仍然是一个持续的挑战。现有技术，如网络防火墙、病毒扫描程序和自动软件更新，可抵御许多攻击。尽管如此，由于管理错误、用户错误和未打补丁的软件漏洞，几乎每台连接到互联网的计算机都有可能在任何时候受到威胁。由于没有计算机安全的“灵丹妙药”，在可预见的未来，这种情况将继续下去。因此，如果我们希望保护互联网的安全，我们需要开发能够容忍我们的计算机系统和与之共事的人的缺陷的防御策略。

Computer security remains a ongoing challenge for most computer users. Existing technologies such as network firewalls, virus scanners, and automated software updates defend against many attacks. Despite this, virtually every computer connected to the Internet has the potential to be compromised at any time due to administrative error, user mistakes, and unpatched software vulnerabilities. As there is no "silver bullet" for computer security, this situation will continue for the foreseeable future. Thus, if we wish to secure the Internet, we need to develop defense strategies that can tolerate imperfections in our computer systems and in the humans who work with them.