Mobile users of portable computers will soon have online access to a large number of databases via wireless networks. For example, while on the road, passengers will access airline and other carriers schedules, and weather information; investors will access prices of financial instruments, sales people will access inventory and other corporate data, commuters will access traffic information. However, access to an online database may be unavailable when the user is out-of-range of the wireless network. Furthermore, sending access-requests from the mobile computer to the online database may be expensive due to limited uplink bandwidth, and also due to the fact that sending messages is a significant drain on the portable battery. These two problems can be alleviated by maintaining a materialized view at the mobile computer. This view will be updated using wireless data messages, as the online database changes. This will also localize access, thus improving access time to the view. Therefore, to better deal with the problem of disconnection and to improve response time, the view should be materialized at the mobile computer. As explained above, sometimes communication cost can also be reduced by materialized views; but not always. For example, if the materialized view at the mobile computer is accessed very seldom, whereas the view is updated very frequently, then the cost of wireless messages that update the view may become excessive. (For example, currently RAM Mobile Data Corp. charges on average \$0.08 per data message, the exact charge depending on the message size). In this case, assuming that cost minimization is the major consideration, the view better stay virtual (i.e. nonmaterialized) at the mobile computer; the view will be requested from the fixed network, when accessed at the mobile computer. In other words, the optimal {\it allocation scheme}, i.e. the set of mobile or stationary computers at which the view is materialized, may change dynamically as the access pattern to the view changes throughout the network. The optimal allocation scheme also depends on the cost structure in the fixed and wireless network. Thus, we are also conducting research into {\it dynamic views}, i.e. views that are materialized according to different allocation schemes at different points in time. In conclusion, the objective of this paper is to outline the major issues related to maintaining personalized views in a network of fixed and mobile computers. Although the issue of view maintenance has been studied extensively in the literature (see [1] for surveys), our research concentrates on new problems introduced by distribution and mobility. New problems are introduced by the fact that the connection between the materialized view at the mobile computer and the online database may vary widely in terms of cost, reliability, and capacity. Specifically, wide-area wireless networks are costly, unreliable, and have a small bandwidth, whereas local-area wireless networks suffer much less from these limitations. We are currently building a software system for view maintenance in mobile computing that addresses these problems.