Thursday, March 5, 2009

Aggregation Types

Great article in Wall Street Journal - Information Wants to Be Expensive (found via Big Dog Little Dog) suggests that more people should be charging for content online. There was definitely some good points about what people will pay for:
People are happy to pay for news and information however it's delivered, but only if it has real, differentiated value. Traders must have their Bloomberg or Thomson Reuters terminal. Lawyers wouldn't go to court without accessing the Lexis or West online service.
...
For years, publishers and editors have asked the wrong question: Will people pay to access my newspaper content on the Web? The right question is: What kind of journalism can my staff produce that is different and valuable enough that people will pay for it online?
...
American Lawyer founder Steven Brill argues that "local newspapers are the best brands, and people will pay a small amount to get information -- whether it be a zoning board or a Little League game -- that they can't get anywhere else."
So people will pay for differentiated, quality content that they can't get anywhere else.

Journalists as Human Aggregators

As part of working on Topic Hubs, I've come to realize that there's a lot of very high quality content already out there. It's free. But there's friction finding it, organizing it and making sense of it.

Many of the people who write the blogs who are included in Topic Hubs are the same people who are being interviewed by Journalists for articles. Take a look at the recent hub around Electric Vehicles. This includes folks like Chris Paine - Who Killed the Electric Car - who is regularly interviewed. In thinking about this, I realized that:
Journalists are human aggregators.
They go look at the information, often in areas they don't understand that deeply and pull it together into a meaningful piece. They are quite good at this aggregation role. And no current automation is able to produce as high a quality result as a good journalist. But ...
The information behind the article that a journalist produces is already available for free somewhere.
There are cases (the local little league game) where no one else has captured that information or where the journalist truly creates something new. But it's like the old adage ...
In order to bake a cake from scratch, you first have to create the universe.
It's pretty rare to be working on truly new, differentiated, high value content. Most of what we work with are derivatives. I think of everything I'm writing now as being new - because it's new to me - but I'm sure that there's discussion of all of these issues out there somewhere.

Aggregation

It's pretty rare when I disagree with Stephen Downes, see Stephen Downes is Wrong. But he left a comment on my Topic Hubs post:
Topic hubs are not the way forward. Focus on being a network, not being spikey.
I found this to be quite interesting. I think of Stephen as being one of the biggest topic hubs out there. His OLDaily is Stephen doing amazing things by finding interesting articles and tagging them; he also has technology that pulls it together and organizes it. This helps to make sense of a large network of bloggers and other information sources and organize it for consumption by folks like me who are not going to subscribe to all of those blogs individually. He also helps to organize the information for you via tags that allow you to find stuff on topics at a later time.

I think that Stephen provides tremendous value on top of a network (and is part of the network himself). And I guess I think of him as a human-centric aggregator. Maybe a better term is provided by Robin Good - he calls this - NewsMastering.
Newsmastering is the process by which a human being identifies, aggregates, hand-picks, edits and republishes a highly-focused, thematic news via RSS.
It's interesting to see the term "human being" - both human and singular.

I believe that Topic Hubs like eLearning Learning, Mobile Learning, Informal Learning Flow, Communities and Networks Connection provide a similar kind of value as Stephen and Robin - but attack it differently. In this case, social signals (human activity across the network) surface posts such as shown in eLearning Learning Hot List Feb 1-14, Work Literacy Hot List - Early February, Hot List from the Communities & Networks Connection, Mobile Learning Hot List.

Each of them combines human decision making about what should be brought in (Thanks Judy, Nancy, Jay) and social signals, activities across the network - thanks everyone!, to determine what's likely good stuff. This relies much more on automation and doesn't have the editorial that Stephen or Robin provide. There certainly is a difference when you have a single individual (or small group) providing editorial control. Robin puts it this way:
The real added value is specifically in the ability of the newsmaster to manually pick the very best and most relevant stories for its target audience.
You can argue that none of the information provided by a Topic Hub is new. However, it is new in that it provides value on top of existing information much like Stephen's OL Daily and a journalists article. They use different methods to surface what's interesting or relevant. They create additional information and structure on top. And there's value in that additional information and structure.

Aggregation Types

As I'm thinking about this, there's likely a few different forms of aggregation implied by looking at systems like Social Media Today, OLDaily, Communities and Networks Connection, Techmeme, Sphinn and Digg.
  • Centralized content or distributed content. Do they pull all the content into the central site or leave it distributed on the original source?
  • Organization and Access - how do they organize the content. Human tagging? Automated? How do you access it?
  • Editorial Distribution - Single person, small group or widely distributed control of what comes in and what is best?
Each of the different approaches has a reason and rational. I look forward to trying to figure out what makes sense in what situations.