Experiencing Information

Daylife, one of my favorite news aggregator sites (probably because it’s one of the best designed news sites), is now offering news widgets. They have some of their own, but also invited anyone to create a share widgets too. The News Index widgets shows change in overall mention for various topics and people from week to week.

We’ve been getting a fair amount of information in the European media about the presidential race in the USA, but I’ve still not been following things as closely as I should. Came across this resource recently that could help out: Presidential Watch 2008 from Linkfluence. Two interesting things here:

First, the use of information visualization is quite good, I believe. The graphs could be drawn a little better, but overall it’s fairly intuitive to use and provides a good amount of control. The main focus is to show trends–mostly at the source level. And it does this good. I like the Trends Monitor, where you can put two candidates head to head on a chart. (I assume this is a Flex application rendered in Flash on the browser.)

Second, the use of analytics to measure influence is interesting. The Watch includes blogs and communities as well as traditional online media, so you get a fairly broad picture. Looks like they are using volume of links to measure influence, which is a good start. To some degree, they may also be analyzing who is saying what and how they are saying it.

What you don’t get is how much the leading sources of information in the presidential campaign change opinions. Just because lots of people link to a certain political blog, for instance, doesn’t indicate whether others are persuaded to change their opinions. That’s hard to measure, but when talking about influence you ultimately need to know that.

Just came across GoLexa. The interesting thing about this is the search results. They provide quite a bit of context, including links to bookmarking sites, page data, page previews, etc. And there are also plenty of other tools, like direct links to analyze keywords and refine your search.

This brings up the point of the Navigation Layer that I made in my presentation at the Euro IA conference in Barcelona. Navigating the long tail of online information isn’t necessarily about having content or even just finding it. It’s about making sense of it and understanding it. In order to do that, you have to provide structure to both the tools and the content, which is what GoLexa does. There is a lot of hand-crafted IA work on the search results page for GoLexa, even though the content is all dynamically populated.

Check it out–it’s quite interesting.

Just came across a new book on text mining: Tapping into Unstructured Data: Integrating Unstructured Data and Textual Analytics into Business Intelligence, by William H. Inmon and Anthony Nesavich. I previewed it on Safari and downloaded a few chapters.

The book is not technical in the sense of showing programmers how to code, but it does focus on database architectures and the like. And when they talk about structured vs unstructured, they are really referring to database structures, not necessarily information architectures.

There is a chapter on visualization, but this is disappointing: it’s more about the process of creating visualizations than about whether the visualizations will be meaning to any human being. In fact, one of the examples used is a bar graph, where the bars themselves are blocks and they are stacked in a three-dimensional arrangement—two no-no’s.

One key point they make—a point I made in my presentation at the Euro IA Summit this year in Barcelona—is that for unstructured data to be useful, it often makes sense to bring it into a structured environment. This makes possible analysis and understanding that would otherwise not be possible.

The penultimate chapter is a brief case study on creating a corporate taxonomy. This company in question created one to help them tie together disparate IT systems and to allow analytics to take place at all. Taxonomies still have a place in the unstructured world.

The writing style is dry and not very engaging. And the summaries for each chapter (which I hoped to give me a better overview of the content) are very thin. So, I’m not sure I’d recommend you run out and buy the book, but since I have a Safari account it was certainly worthwhile to go over the content quickly. I plan to read a few key chapters in full later.

My ex-LexisNexis colleague Kevin Simons tipped me off to a new news service. David Winer created a way to browse the New York Times by topic. See his announcement of this serivce and the topic tree (i.e., an outline) for the NYT.

This outline isn’t really a taxonomy, but rather a list of keywords. I’m not sure where the keywords come from, however. Are they extracted for the stories algorithmically, or did Mr Winer set up queries behind each keyword ahead of time? Looks more like the former to me.

There are three things interesting with this that point to where I think online news content is going in the future:

1. Merely aggregating content will become less and less important as more and more of it becomes available on the web. The FT recently announced they will be making some its content available for free. This, on the heals of similar announcements from the NYT and even the WSJ.

2. With access to content essential equal, being alerted and making sense of information will both become more and more important. Different forms of text analytics will proliferate, as well as alerting services. And things like categories, taxonomy, and other IA artefacts help in both respects. Post-coordinated (pre-determined) structures will help make sense of it all. And you’ll pick a topic–perhaps a hand-crafted topic–within a meta-RSS feed mashup to be alerted on.

3. News for mobile devices makes a lot of sense. People read the morning news on the go. Executives get more information from their Blackberrys than from their laptops. And the NYT outline works great on normal cell phones.

Chris Anderson, of Long Tail fame, has this post on his blog about free chapters from a new book called Blogging Heroes: Interviews with 30 of the World’s Top Bloggers by Mike Banks.

The idea from the publisher (Wiley)  is that each of the 30 interviewees gets to give away his or her chapter. Interesting marketing scheme. Sure, the entire book is now available for free on the web, but you’d have to do some scavagering to it all. And along the way you’d be exposed to messages from the authors about their work and about the book. So there may be a powerful marketing effect here.

Others who are promoting their own chapter include Mark Frauenfelder at BoingBoing, David Rothman at TeleBlog and Steve Garfield.

Not sure if I’m going to buy the book. It’s only $17 on Amazon.com, but €26 on Amazon.de.

Jan tipped me off to FeedHub. This is a beta attempt at filtering lots of RSS feeds. I have to admit I’m not 100% what it tells me, but it appears to be doing some kind of text analytics on my feeds. It then personalize a structure around those feeds. The goal is to reduce RSS clutter and noise, so I can focus on the topics and subjects I want to (so they claim).

Here is what mSpoke, the creators of FeedHub, have to say about its inner workings in a blog post:

“Very simply, we learn about you based on the implicit usage of your personalized feed and any explicit gestures you choose to share with us. We use this information to distill a set of “memes” that describe your preferences. Each meme represents some characteristic of a post, like its topic, popularity in del.icio.us, or number of Diggs. Each meme also has a strength that indicates how predictive FeedHub expects it to be in choosing content you’ll like. As we learn about you, FeedHub automatically discovers new memes for you and strengthens or weakens memes appropriately.”

So you basically give FeedHub your feeds as a OPML file, it analyzes them for you, and then builds a profile of your interests that you can manage and customize.The basic building block of all of this is what they are calling a meme, or an extracted category.

I’m quite confused about the overall experience and how this really helps me make sense of the feeds I currently subscribe to. If anyone has more experience with it, I’d like to hear about it.