Fact versus Opinion

Information overload overload is becoming a serious problem for me. There are so many services out there that offer ranked lists of pages that I really need to read, that the task of evaluating these services is becoming almost impossible. Heck, I only just have time to get some coding done in between flicking between TailRank, Digg, Memeorandum, BlogPulse. Oh, and Slashdot. All of these services offer something - I almost always find something of interest on each of these sites on every visit. Interestingly, and naturally, all the lists are different, which begs the question: how do we evaluate these services? and, when will we see an aggregator aggregator?

At any rate, all this bouncing around the listosphere has resulted in some fresh perspective in terms of what I want - a tool that would help me manage all this stuff.

Take a very abstract view of the content: there is factual content and there is opinion/analysis/etc. Factual content has long been the primary intention of the newspaper. Of course, there is bias, misinformation, etc., but for the sake of argument let us believe in this model. Opinion, on the other hand, colours and textures factual content. Of course, any real content source out there is a blend of both fact and opinion - or at least the intentions to deliver fact and opinion.

So, here is what I want: a stream of factual data - not in the form of lists of links, but in the form of abstract factual summaries derived from textual content. In parallel, I'd like to have a stream of opinion data - blog entries, editorial essays, etc. I'd like to control my attention of these individually, according to my current working state and according to interactions I have with the streams as and when they are visible to me.

Comments

[Kevin Burton]

Interesting idea...

The problem is that in the current political climate the definition of trubth is a bit abstract:

http://en.wikipedia.org/wiki/Truthiness

RE quality of services I've been really eager to write some code to provably ensure that TailRank delivers results better than our competitors. I still think we have more work to do in this are but we're getting there :)

Kevin

[jackvinson]

Interesting indeed. Can you give us an example of how this might look? Is this based on a topic you claim as "interesting?" Or do you want a smart version of a commentary history: here is the source, and here is all the commentary.

[Ken Yarmosh]

Matt,I have been thinking about something very similar lately. I started working on a post yesterday and would like to put it up tomorrow. I hope to include some of your thoughts.