GroupLens is an open source structure for collaborative filtering of netnews. It is similar to Netflix ratings, Amazon reviews, and Rotten Tomato. This paper studies the group benefits of shared rating system and other filtering techniques, for example, the ability to put certain topics or authors on a kill file. Results show that a 5 point scale system is easy to use and understand by users. Social connections are facilitated through this collaborative rating architecture. The author concludes that it would be wasteful to not utilize reader reactions to create a system to inform future users.
The thesis of this paper has branched off into many research questions since it was published. Search and filter, internet privacy, collaborative social media, and personalized advertisements all have theories rooted in grouplens. This subject is seemingly more sociological than technical now. Key components derived from the paper provided guidelines for social computing engineering and design. I recognized many categories from models of Human Computer Interactions and procedures for Human Centered Design.
Much of the techniques mentioned in the paper are still in use today. Granted, some are dramatically improved up, some even took up hypothetical suggestions actually envisioned in this paper. An interesting topic is kill filing, or blocking/ muting, of authors. I feel this is interesting because there are many possibilities of why one might want to kill a certain person’s work from appearing. The paper states the system should block off similar authors as well. However, I think many times, the person is interested in the subject category, but personally dislikes a writer for various reasons. Modern technologies have the option to block or “stop showing contents like this.” Of course, there are still outdated things, like software for soullessly counting and calculating ratings, which is much too trivial now.
Another aspect that I feel has come a long way is anonymity. At one point, as stated by the authors, people were asked to leave their real names. Growing up with the development of the Internet, I recall when I felt the need to naively sign my real name. As time passed, I started to lie about my identity. I feel anonymity was less a way to start protecting privacy and more a lie control mechanism, eliminating the need to lie. In this case, I wonder if one could study the impact of anonymity on groupware. Because any empirical test will likely reveal personability associated with people knowing the truthful self of others as opposed to mysterious persona. However, in real settings, it may be lies versus mystery instead.