How Knowledge Workers Gather Information from the Web: Implications for Peer-to-Peer File Sharing Tools

Abstract

The success of peer-to-peer (p2p) music sharing has no doubt contributed to assumptions that individualsﾒ PCs are a vast untapped resource of assets just waiting to be unlocked by such systems. This includes the push for opening up our file spaces at work to allow peers access to previously inaccessible information with minimum effort. We wished to explore the potential value of these ideas and to test some of the assumptions underlying them, the motivation being that we believed the issues raised by this investigation would be important to those developing p2p information sharing tools. We do this by looking at the flow of information in and out of 16 knowledge workersﾒ file spaces in the context of carrying out Web information gathering tasks at work. In doing this we find that the file spaces used for knowledge work are more like ﾓworkbenchesﾔ than ﾓarchivesﾔ and that the information held within them is fundamentally different in content and organisation to that which knowledge workers place in shared information spaces such as the Web. Knowledge workers work on their information to make it shareable to specific audiences yet this information is found side by side on the ﾓworkbenchﾔ with unshareable information. This leads us to question the potential value of enabling people to open up their file spaces without having regard to the reusability of this information for others.