I have been thinking a bit about this notion of a real-time web. Having access to real-time information, as soon as it is published, seems to be a possible Achilles heel for Google according to some (here and here). People who say that do not understand the real strength of Google or it’s possible innovator’s dilemma. But the question that interests me is the user value question. Does it provide us value to have access to information, the moment it gets published? The answer is that it isn’t nearly as important as something else (will get to that).
I guess there are cases where this can have value. An area that comes to mind is big events. The Obama inauguration, a plane crash, earth quakes, the super bowl final.
I’ve tried to use Twitter search and Friendfeed’s real-time options, and honestly, I find the experience mediocre. A bit of nuance might be in place here as we are only discovering the first potential of such services. However, I am trying to grasp what the specific real-time component adds to the experience. And I can’t put my finger on it. I can think of a few reasons why:
- Life doesn’t jump from one big event into the next one. When watching the Obama inauguration, seeing the Twitter community discussing and commenting it gave a sense of added value. The information added value to the experience at that moment. If I look for Obama on Twitter now I get an incredible amount of useless information. The context defines value. Currently is no context in which real-time search results on Obama now provide me much value. There are times when there is such a context, but most of the time life goes on.
- Immediate knowledge doesn’t always add value. If there is an earthquake in San Francisco (or anywhere else for that matter) we now see Tweets reporting in within seconds. But that information is only relevant if you are in it (you didn’t need a Tweet to tell you about it), you have people you know live in that area, or you need to know it for professional reasons (e.g a reporter). The randomness of the waterfall of information getting through makes it hard to understand what is really happening out there. A recent plane crash in Amsterdam appeared within a few minutes on Twitter. It gives people a reason to discuss it (terrible tragedy) at the coffee corner, but did it really provide value? Not unless you had a relative in that plane crash.
- Real-time information is hard to verify and trust. People are saying a lot of things on services like Twitter. Without context or understanding more about the people tweeting, it can be really difficult to understand the trustworthiness and accuracy of the information. You can already see the algorithms being drawn up that take reputation, reliability and trust into account, but this problem can’t be solved easily. Reputation, reliability and trust aren’t real -time characteristics. They take years to build. The only way these characteristics can be determined on information is for that information to be published, read, and responded to by large amounts of people. A blog post can build up trust, reputation and reliability if it has been exposed to readers, critics etc. But a tweet that appears in seconds doesn’t follow that process, no matter what the reputation of the person is that sends it out.
Does all of this means that the real-time web and search has no value. Off course not. Getting the news out fast is important, and it has caused many of he traditional media to get online to join this rat race. But in my opinion speed really isn’t the most important factor.
I do think that it becomes increasingly difficult to find information with enough relevance. There is just too much out there. Google can’t index the entire web fast enough, nor is it able to display the most relevant links in any particular situation. Aggregators, no matter what kind, tend to do a pretty poor job of aggregating relevant information timely for us (yes that includes Friendfeed, Digg, Reddit, and most of the major tech blogs). If you want to know more about that, then read this excellent post by Paul Graham who talks about his experiences with setting up and running the Hackernews community. Excellent read.
It seems we do a much better job at storing and retrieval of information that doesn’t lose value as time passes by. Encyclopedia’s, history, arts, dictionaries, etc. There are however some experiments that try to approach the problem of information organisation very differently. I’ve always been very font of the work that Jonathan Harris is doing this area. Check out his universe demo, and his “We feel fine” project. Seriously, give it a spin and then come back. I’ll hold.
Jonathan’s work proves to me that we haven’t reached the depth of possibilities to handle information. I’ve said this before, but if I were Google or anyone else interested in organising the world’s information, I would definitely get someone like Jonathan on board. His work actually makes me crave for more information. I can get lost in the universes he has created and I return frequently to dive in for some more.
The real-time web sounds cool, but right now it isn’t much more than another technical capability. I don’t really get passionate about that. Instead I’d like to see what happens if we let non-tech people like Jonathan redefine the way we would be able to access information. I’d say we would find some more ground-breaking and relevant ways of information organisation and retrieval than the “real-time” web. I’d take this one step further and say that it isn’t relevant if published information gets indexed and found in real-time. The only relevance we should be focusing on is getting the user the right information at the exact right time!










