Content Delivered on Internet - How Much Can We Trust?

In this paper, we investigate the trust problem that the data retrieval and reuse processes bring to web content. We argue that this trust problem is due to the inappropriate setup of web server, heterogeneity and presentation demand of web clients, value-added functions (including caching) in the network, and insufficient trust-related support in current web architecture and protocols. To quantify our arguments on the trust issue of web content, modeling and monitoring of real data on Internet was conducted to quantify the seriousness of its error situation.