We present a study of how awareness of presence affects content of instant messaging sent between students using WebWh o, an easily accessible web-based awareness tool. WebWho visualizes where people are located in a large university computer lab and allows students to virtually locate one another and communicate via an instant messaging system. As WebWho is there to be accessed through any web browser, it requires no programming skills or special software. It may also be used from outside the computer lab by students located elsewhere. The sender's user name is normally automatically added to the instant messages, but the messages can also be sent anonymously. We were interested in finding out if the sender's conscious hiding of his or her identity seemed to be reflected in the content of anonymous messages, and how these differed from those with identified senders. Awareness of presence seems to be one of several factors influencing message composition, both content and structural aspects. At this stage, we have primarily focused on examining how different factors affects the content of the messages. We cross-analysed the messages for content in relation to parameters such as sender location (collocated, distributed and distant) and sender status (anonymous vs. identified), in order to find out whether awareness of presence seems to be an influencing factor. Computer-mediated communication (CMC) is often claimed to be a sort of hybrid between spoken and written interaction c.f. Ferrara, Brunner & Whittemore (1991) and and several others]. We compared the messages that were sent using the instant messaging tool in WebWho with data from other types of CMC (email, chat) and also with corpora of spoken language and traditionally written language. The aim of the study was primarily to investigate awareness of presence affects on instant messaging, and only secondarily to investigate spoken vs. written features of the texts. Results show that awareness of both physical and virtual presence affects the content of the messages, and that these factors affects the text differently. Sender status, the nature of the computer-mediated medium, and the written mode shape the messages as well. Results show that the students use the messaging system to support collaborative work and coordinate social activities, and extensively for playful behaviour.