Paper Reading #21 - Towards a Reputation-based Model of Social Web Search

Comments
Jeremy Nelissen
Vince Kocks

Reference Information
Kevin McNally, Michael P. O’Mahony, Barry Smyth, Maurice Coyle, Peter Briggs, IUI’10, February 7–10, 2010, Hong Kong, China.

Summary
This team of authors has created a unique way in which search could be moving. The social aspect to the web has boomed in the last several years and this idea called HayStaks is trying to socialize web searching. It comes in the form of a toolbar in the users browser and will record all searching events. Friend's can be made through the service and will allow friend's to share their searches with others. This can create communities of similar searchers and will also create people who explore and give out recommendations and people who follow those recommendations. The HayStaks program also has implemented a recommendation algorithm that is similar in style to Google. It takes into account the user giving the recommendation, how many friends they have, how many its shared to, the applicability of the search, etc. This allows for easier and better searches for the users to find. Also, the most important aspect of socializing the web is what the users can do with the recommended links. There are features to vote on the links, comment on them, etc. to bring people together and discuss.

Discussion
The idea of a social search environment is definitely logical as all aspects of the web are going that direction. However, I think it might be slightly over-saturating. I want my searching to be quick and simple and not have to spend my time with voting, etc. I think the approach that Google is taking might be a little better as it helps both the user and all other users. They have simply implemented a voting system that if your results were good then the website that answered the question is given 1 upvote. This helps the future searches of the user and also anyone else that searches that topic. Over time this will create much better results even if the searcher does not get better. Also, it seems like this system could simply be combined with Facebook or some already established social network since creating a social network needs a critical mass of people to become viable and/or relevant.