The underlying principle behind every query answering system is to provide a query describing accurately the information of interest. The desire to bring query answering systems to a broader audience made clear the need for relaxing this requirement. Keyword queries were a first step towards this direction, but was not enough. Keyword queries are still machine oriented, i.e., few people talk to each other in keywords. We try to push the motivation of keyword queries even further and build query answering systems that are closer to the way humans interact and search for information in their daily lives. We present two works that bring two important features to existing the search engines: Interactivity and Examples. The first allows the system to interact with the user when the systems is not in position to provide an answer, in order to help the user find what is searching for in the shortest possible time. The second, introduces a novel form of query answering in which queries are not any more specifications of what the user is searching for, but simply a sample of what the user knows to be of interest but cannot provide a description that includes all the elements of interest she does not know about. We also talk about the problem of answering keyword queries over relational data.