Wednesday, November 7, 2007

Search for Opinions

This is something I have been thinking about ... Most search that we are doing today, is based on the content. When you search the blogosphere, or any part of the internet, you are searching primarily for keywords. These keywords could either be words which occur in the document (this could be a document, or a site, a blog, or a wiki entry). Or, these could be corresponding to tags which are attached to the document by others.

This goes to show some of the things social computing has changed. In fact, shows the heart of the thing. Earlier, when you would search (remember those days when google wasnt there?), you would search only for keywords, which would be matched against those words occuring in the site, or document, and based on this search results would be thrown up. This has been extended by the evolution of the web, into web 2.0, where we can now search for things based on the way others have classified them. So now, we have two ways of searching. Well, simplistically speaking, that is ... But, thats the bottomline. Google changed the name of the game with some highly sophisticated algorithms.

How about giving some more? With the explosing of blogs, I think this is a requirement which is bound to come up sooner or later. Today, we search for words. But, blogs are more than words. Blogs represent opinions, and these opinions vary with people. Opinions are like noses (the original one is unmentionables, so I found someone change it to noses, and will stick by that), in that everyone has one. Fact is, what I find, at times, we need a way to search for opinions of people. True, these opinions are made up of words (and hence, we can search blogs), but what I am talking about is the aggregate. A way of searching for the aggregate opinion. Much has been written about prediction markets, and the fact is, groupthink notwithstanding, prediction markets have a way of getting things right more often than not. Not that groupthink would play a role in a diverse group like a prediction market, but the point I am trying to make here is, to make more sense out of the divergent voices which are emerging from the blogosphere, we probably are soon going to need a tool which can help us make sense of them, and bring out a picture of what the blogosphere is saying, on the aggregate. Dion Hinchliffe has mentioned sensemaker as a tool which can do this, but I couldnt really make too much sense from their site. A tool for this has to go beyond RSS, and needs to identify meanings from what people are writing, and bring this to a single platform.

Let me take an example ... When a new movie is released, you can get a general sense of how the movie is being received if you hang around the theatre. I am talking about something similar ... That, if you hang around the blogosphere, for a particular topic (as opposed to keyword), then you should be able to make sense of the conversation that is happening, and the general direction the conversation is taking, and this should be much simpler than it is now. Maybe something which take wikimindmap to a new dimension? If there are tools out there, would like to hear about them.


Unknown said...

i know not of any tool but remember we were talking about auto summarizing blogs AFter every tenth entry or so so that when the blog becomnes unwieldy it becomes manageable and information also is easily recoverable.
Remember auto summary in MS word.
probably if we have something like this and a google like search tool we can make general sense about blogs too.

Anonymous said...

Politeness costs nothing and gains everything.

Massachusetts Institute of Technology

Anonymous said...

I am speechless. This can be a good blog page and quite engaging too. Outstanding function! That's not really significantly coming from an beginner writer like me, but it is all I could assume soon after enjoying your posts. Excellent grammar and vocabulary. Not like other blogs. You seriously know what you are speaking about very. So considerably that you just produced me would like to discover far more. Your website has become a stepping stone for me, my friend. Thank you for that articulate journey. I genuinely loved the 27 posts that We have learn so a lot.