OneRiot Solves The Realtime Search Problem
Last October I wrote an article entitled “The Real Problem With Realtime Search” in which I explained one of the major problems of realtime search revolved around the lack of relevancy. In addition to the lack of relevance within results, we were inundated with unfiltered streams of the same data and most of the time trending topics revolved around a single source, of course, that being, Twitter. As I pointed out in the previous article this was a problem. However, as of today, the problem is being resolved. OneRiot has officially launched out of beta with a completely new approach, which I think, is ground breaking. I know I wrote them off in the previous post, however, I must say, my faith has been restored.
The latest release of OneRiot is a completely innovative and fresh approach to realtime search. Let me explain why OneRiot has it right and everyone else has it wrong. The major thing OneRiot is doing right is looking at the web as a whole as realtime. They are not solely focusing on Twitter. Currently, it appears based on their header that they examine Twitter, Facebook, MySpace, and Digg to see what’s trending on the web. This is a great start and I hope to see it expand to even more sites. Breaking away from the fold and understanding that realtime news can be other places then Twitter is key. No one else is doing that, everyone else (Google, Bing, and all the other guys) is so focused on Twitter as the only source – they are missing pieces of the puzzle.
This is exactly what I have been saying all along – how can you piece a puzzle together without all of the pieces? You can’t. By using Twitter as your sole source for a realtime search stream, you are literally missing a huge chunk of the puzzle. What do I mean? Let’s use the recent Times Square bombing attempt as an example, keep in mind this is just one of the many examples that could be used: The story breaks (either on Twitter or off – it doesn’t matter), suddenly you have a flood of tweets about it, some provide relevant information and links, but most do not (e.g. people just talking about it). As a user you will become completely lost in the stream of tweets. It’s literally impossible to sift through the good from the bad on Twitter. Now you search it on Google, you find a story, but it’s old and the realtime search stream in the middle of the results page is completely irrelevant. Now what? You are now forced to search news sites, blogs, etc. manually for information. Realtime search failed you.
What OneRiot is doing is genius and will fix this problem. Again, by examining multiple sources of information and using their algorithms to break down what’s relevant and good, and what’s not, you will be presented useful links and information. They can see what’s being linked to on Facebook and MySpace and what’s being Dugg on Digg in addition to what’s being tweeted. Not only will you have better realtime results, but you won’t be presented with the same links several times (a real time saver – pun intended).
Additionally, OneRiot will now be able to essentially determine what’s going to be “hot”, before it actually is trending. This could be done to an extent if you only focus on Twitter, but, to see what the web is truly talking about you need to look outside of the Twittersphere and I think OneRiot realized that before anyone else – or at least, so it seems. Not to mention, I see they want to license that technology to sites, boy would I love to have access to that kind of information. Nonetheless, if it works as well as their site does, it will be a really valuable tool for many major news sites and blogs.
This is what I expected to come out of Mountain View and Redmond, but it hasn’t. I guess I understand why Twitter hasn’t done it, but then again, if they want to be the pulse of the planet, maybe they should be looking in this direction. Nonetheless, OneRiot has nailed it – they deserve a big hand and a lot of recognition.
Sign in or become a BestTechie member to join the conversation.
Just enter your email below to get a log in link.