 |
v1
v2 |
Check out the new website!
Echo Chamber is a link aggregator that suggests links to you based on your shared likes and dislikes with other users. You can currently post new links by using our Chrome extension or Android app. If the link is new, a Posted by [your username] byline will be shown under the link title everywhere it appears. All posted links support threaded discussion, and individual comments can be liked. You can add friends via their user page to see everything they like in the friend links section, and choose to get notifications when they like something new or post a comment. You can also use the Android app to add your own RSS/Atom feeds, and you’ll be able to see links from those feeds in your feed links section as well as mixed into your suggested links section on both the website and in the app. If you’re the first to like a link in your own feed, you’ll get Posted by credit, and it will be shown to other users in your Suggestion network.
The short of it is that Echo Chamber averages out your agreement ratio with every user who voted on each new link, and presents that average percentage as a match score. Links with a match score of 60% or above are shown in your suggested links section. Links with a score of 40% or below are shown in the rejected links section. Anything in between, including links with no relevant vote correlation data are placed in the new links section. The more you vote, the more refined your agreement scores with other users become. New suggestions are generated once an hour, and to keep things more relevant to your recent choices, only votes from the last three months are factored into the calculation.
Indeed it is. Since Echo Chamber is new and there are not a lot of us using the site yet, we are sort of simulating a situation where a lot users are posting a lot of content by scanning RSS feeds hourly, and “liking” each new link by users corresponding to the author and each category the link appears under in the RSS feed. This is not quite as good as having real people liking or disliking links based on whether they enjoyed them or not, but it already does a pretty good job of suggesting a lot of good stuff if you like any of the auto-generated content. And of course, the more people use Echo Chamber, the more realistic these correlations should become.