Trust, regret, forgiveness and comfort for information management in autonomous networked devices.

这项工作涉及信息保护、隐私和信任、后悔、宽恕和舒适的计算社会规范。它解决了物联网等联网设备的信息保护中的不适当保护、脆弱的安全模型和上下文安全漏洞的问题。世界是相连的。设备，其中许多在我们手中，其他在车辆、建筑物、我们周围的传感器和全球服务器上，不断地共享、交易、分发、操纵、重新散列、存储和重新分发数据和信息。这些信息中的大部分都是敏感的--个人的、私人的、可识别的或具有其他价值的。有许多机制可以帮助保护这种信息--例如，传统安全试图创建一个我们必须通过的“看门人”系统，我们必须通过该系统才能访问信息；加密解决方案试图使我们更难(如果不是不可能的话)看到不是为我们设计的(但可能仍然是关于我们的)信息或数据；存在基于信息持有者的安全要求的最佳路由信息的算法。另一方面，有一个名副其实的行业在打破这些保护，以便获得这些信息--既有非法的，也有受到制裁的，即使被制裁的人有时合法性值得怀疑，就像斯诺登的爆料所表明的那样。后门、网络钓鱼、中间人攻击、破解密码等等都是“利益相关者”可用的许多不同工具和技术。当然，安全只会失败一次，而攻击者可以继续尝试。正是在这个领域，这项建议审查了信任、后悔、宽恕和“舒适”的使用，作为更好地评估、保护和确定共享协议的手段。其目标不仅是创建使用这些社会概念来更好地管理信息的系统。本研究的具体目标是：1.考察信任在网络设备信息管理中的作用；2.将我们现有的信任、后悔和宽恕模型提炼为一个组合模型，将舒适度作为沟通和推理的工具；3.为组合模型开发适当的算法，用于共享、保护、过滤、警报和以其他方式管理设备上和网络中的信息；4.在网络设备上测试算法，包括信标、移动个人设备和更强大的计算设备；这项工作不仅有可能深刻地影响信息的存储、管理和共享方式，而且还会深刻影响使用它的人和它涉及的人的理解。

This work is situated in the areas of information protection, privacy and computational social norms of trust, regret, forgiveness and comfort. It addresses the problems of inappropriate protection, brittle security models, and contextual security holes in information protection for networked devices such as in the Internet of Things. The world is connected. Devices, many of them in our hands, others in vehicles, buildings, sensors around us and servers globally, continually share, trade, give away, manipulate, rehash, store and redistribute data and information. Much of this information is sensitive – personal, private, identifiable or of some other worth. Many mechanisms exist to help protect this information – traditional security, for instance, seeks to create a system of ‘gatekeepers’ through which we must go in order to access information; Cryptographic solutions seek to make it harder, if not impossible, to see information or data that is not intended for us (but may still be about us); Algorithms exist to best route information based on the security requirements of the information holder. On the other hand, there is a veritable industry in breaking these protections in order have access to this information – both illegitimate and sanctioned, even if the sanctioned is sometimes of questionable legality, as the Snowden revelations have shown. Back doors, phishing, Man in the Middle attacks, cracking passwords and so forth are amongst the many different tools and techniques available to the ‘interested.’ Naturally, security has to fail only once whereas attackers can keep trying. It is in this space that this proposal examines the uses of trust, regret, forgiveness and ‘comfort’ as the means to better evaluate, protect, determine sharing protocols. The goal is not only to create systems that use these social concepts to better manage information.  The specific objectives of this research are to: 1. Examine the role trust plays in information management for networked devices; 2. Refine our existing trust, regret and forgiveness model to a combined model incorporating comfort as a tool for communication and reasoning; 3. Develop appropriate algorithms for the combined model for sharing, protection, filtering, alerting and otherwise managing the information on devices and in networks; 4. Test the algorithms in networked devices, including beacons, mobile personal devices and more powerful computational appliances;  The work has the potential to profoundly impact not only how information is stored, managed and shared, but also the understanding of the people who use it and whom it concerns.