Cadmus aims to filter spammy and repetitive tweets from your Twitter feed
There's no doubt that it's easy to become overwhelmed with the sheer volume of posts when you start to follow a lot of people on a social network. Cadmus is trying to help with that problem, specifically by filtering out similar toots from your Twitter feed, and grouping related toots.
My first impression of Cadmus was not good, but that was possibly due to Twitter's flakiness. I signed up for the service and linked it to my Twitter account using OAuth, and then ... nothing. It took about four hours before Cadmus was able to return any results. The results it now has are quite current, so maybe it needed some time to analyze my (oh no, I'm going to have to say it) "social graph". Or maybe Twitter was just being twittery and flaking out. It's hard to say.
So, now that it's returning results, how are they? Well, Cadmus reports that it is filtering out about 4% of my updates. Not bad, but at the same time it's not really going to save me a huge amount of time. The other thing Cadmus tries to do is to present the most relevant toots to you and group related ones together.
Maybe I'm not following enough accounts on Twitter, because Cadmus doesn't seem to be finding many related tweets among the 172 accounts I follow. This could also be due to the fact that it's the weekend, and there's no big news happening that many people would be likely to be tooting about all at once.
It's also tough to evaluate just what "relevant" means when it comes to ordering my updates. I couldn't care less about football, and three of the top ten updates on my Cadmus page were football-related. It's not really fair of me to expect Cadmus to know that I don't care about football, but at the same time it makes me suspicious of their use of the word relevant. Maybe "popular" would be more accurate.
My personal grouchiness aside, Cadmus seems like a well-designed and executed Twitter client that could prove to be useful, particularly during conferences or times of breaking news.