Community-Based Location-Aware Systems
The concepts behind Web 2.0 has shown the real power of the web through community based selection and ranking of content. The diversity of people and what content they like and dislike cannot be captured completely by editors, and increasing the role of editors is becoming less and less important as readers demand more and more information. Those who actively participate in community-based information selection sites such as Digg, Flickr and De.icio.us enjoy content filtered by the users themselves.
Now we look at the current state of mobile applications. I’m going to use the word application in a loose sense here; which describes any tool, whether it’s a downloaded Java ME program or a WAP-based application. Some are more useful than others. Games, lifestyle applications (like organisers, to-do-list makers), office applications, general communications applications (like Google’s mail application for Java ME) and similar applications have their own varying degrees of success. A specific form of WAP sites and SMS-based applications - the access mechanisms of online social networking sites such as facebook.com lets you use MMS to add photos to your albums. Great, you can let your friends know straight away what you’re up to by labeling the photos with comments as part of the MMS. Facebook also has a WAP site enabling you to access a simplified form of the website, allowing you to search for friends, add friends, message friends, post on friends’ “walls”, etc.
This is all fine and dandy, but there’s something missing - are you really taking full advantage of the power of mobile devices? Particularly, the fact that you carry around your phone into your everyday contexts. The places you frequent, and the things you do, I believe that this information can be harnessed to give people access to information, revolving around the places they visit and the people they see; and indeed, so do researchers in context-aware systems.
The main forms of context we try to describe are time and place. There is also intent, mood, season, social situation and emotion (to name a few), that would affect the understanding and relative usefulness of an item of media.
So what has been done so far? Plazes.com, an online social networking site, lets you put place-markers on real places, and media can be shared around those places. Like facebook.com, you can upload images via your phone through messaging services. The site is basically a community no matter from which direction you look at it, and you can see that it is lacking in the way it uses the power of mobile phones. I believe that we can observe the advantages of the community-centered content creation aspects of the recent web, and apply that to mobile systems. We should extend the web, not create access to what’s already there. Content should ideally be arbitrary, and search mechanisms should be used to explore those spaces. Content should be customised, like RSS, but additional parameters of context should be used. For example, content may become more useful if supplied by someone in your social network. Content will not compete with the information already on the web– instead, local knowledge - tips and tricks on how to get about, reviews on movies supplied by people at your local cinema, should have its place in this mobile network.
What we should study now is not how to create as ‘rich’ an experience as possible with existing technologies, like social networking websites with limited location-based functionality. There is always a compromise when that happens; although a relatively large user base (tens of thousands) can be used to gather data for analysis, it is of limited dimensions (just time, place, and comments). Instead, we should concentrate on studying the psychological needs of people, and create relatively smaller tests so that we can first identify the real uses of location-based systems, before the technology catches up when handset manufacturers realise that positioning systems on mobile phones has become a commodity.
No Comments Yet