In this paper I argue that the kinds of literacy needed for making sense of information on websites is more nuanced and embedded in our everyday context that we are currently providing for learners. The kinds of analysis of websites which allow the processing of information in context are presented. This is demonstrated by an analysis of a scam site, which sold non-existent tickets to the Beijing Olympics and a description of a phishing attempt at Twitter. The skills required to understand information presented on the web have evolved far quicker than the parallel shifts in road safety skills, and people are now required to read web sites contextually if they are to be able to make informed decisions about information available on the World Wide Web. It is proposed that this is achieved through education rather than filtering out undesirable information. (Contains 4 footnotes and 4 figures.)