Showing papers by "Michael S. Bernstein published in 2007"
07 Oct 2007
TL;DR: The design and implementation of Jourknow is described, a system that aims to bridge lightweight text entry and weightless context capture that produces enough structure to support rich interactive presentation and retrieval of the arbitrary information entered.
Abstract: Information cannot be found if it is not recorded. Existing rich graphical application approaches interfere with user input in many ways, forcing complex interactions to enter simple information, requiring complex cognition to decide where the data should be stored, and limiting the kind of information that can be entered to what can fit into specific applications' data models. Freeform text entry suffers from none of these limitations but produces data that is hard to retrieve or visualize. We describe the design and implementation of Jourknow, a system that aims to bridge these two modalities, supporting lightweight text entry and weightless context capture that produces enough structure to support rich interactive presentation and retrieval of the arbitrary information entered.
30 citations
28 Apr 2007
TL;DR: Designs and prototypes for information scrap capture and access tools for short, self-contained personal notes that fall outside of traditional filing schemes are described.
Abstract: We introduce research on information scraps. short, self-contained personal notes that fall outside of traditional filing schemes. We report on a preliminary study of information scraps. nature and outline plans for the next phase of our user study. Based on ongoing study results, we describe our designs and prototypes for information scrap capture and access tools.
18 citations
01 Nov 2007
TL;DR: The contribution of this work is the demonstration that a simple property/value extension to RSS feeds enables a new kind of interaction with information: even non-specialist users can define precise rules to take control of or successfully delegate the handling of the high volume of both personal and public information the authors produce and must process.
Abstract: Over the past two years, social networking sites have fostered a new kind of data publication: personal feeds about schedule, location, music playing, activity and so on. While these feeds have been mainly used at face value as status reports for human readership, we present AtomsMasher, a tool that uses these feeds as a computer's context to inform automatic actions: an update of a current location query, compared with a calendar entry's meeting location and time can trigger an automatic "I'm late; I'm on the way" to necessary parties. This light-weight (though surprisingly complex) automation frees us from manually updating multiple sources; likewise the information context can privilege the presentation of other sources: if the news is not about a band i listen to, don't show me upcoming gigs. To deliver this utility however, we have needed to address two key challenges: operationalizing data sources with little original structure and providing interaction approaches to support non-specialists defining rules for these sources' interaction. The contribution of this work is the demonstration that a simple property/value extension to RSS feeds enables a new kind of interaction with information: even non-specialist users can define precise rules to take control of or successfully delegate the handling of the high volume of both personal and public information we produce and must process.
3 citations
01 Jan 2007
TL;DR: A user-controlled central database of personal information called Databasket is proposed as a potential reinvention of web personalization, and an interface drawing on research in usable privacy and security is designed to keep the user aware and in control.
Abstract: Web applications have put significant effort into personalization services to improve the user experience. The current personalization model suffers from two major drawbacks: each site has access to a very limited subset of information about the user, and the users themselves have little or no control about what data is maintained and how it is kept private. Users thus repeat personalizing rituals across a number of sites, specifying their names, email and shipping addresses, and interests; and web sites often make poor predictions, recommending items when inappropriate or the wrong items altogether. Web sites occasionally see privacy gaffes such as America Online’s in 2006, sharing personal data on the Web and exposing their users to fraud and identity theft. In this paper we propose a user-controlled central database of personal information called Databasket (Figure 1) as a potential reinvention of web personalization. We place the data locally on the user’s computer, ensuring that the user himor herself has primary control over how the data is shared. We provide a Javascript API for web sites to query over a range of this data once the user has granted permission, thus allowing web sites access to customize using broader, more up-to-date data. To control data access, we have designed an interface drawing on research in usable privacy and security to keep the user (arguably the most vulnerable link) aware and in control. To follow, we introduce the Databasket system and its design. We focus first on related work in centralized personal data repositories for the web. Then we describe a typical Databasket use scenario, the system’s user interface and developer API, and back-end implementation. We report on a first-use study of the interface using two web sites developed using the Databasket API, and finally focus on challenges and future work for the system.
2 citations