Saturday, July 30, 2005
Smart Aggregators and Efficient Attention Allocation
When I talk to people about Media 2.0, one of the questions I'm often asked is what's the difference between a vanilla aggregator and what I've termed a Smart Aggregator.
The answer's simple: the former are about syndication scale economies - grabbing and centralizing a lot of stuff - whereas the latter do that and also allocate attention to it efficiently.
Now, efficient attention allocation requires some thought: it's tricky to put into practice, because old-school concepts like 'personalization' cloud the picture. Personalization is like 'broadcatching' - a nice term, but strategically kind of meaninglessness
Without going into too much depth, I think there two dimensions to what Smart Aggregators should be doing right now: filtering the right content from the wrong content (what most people in the industry unfortunately call 'relevance'), and then filtering again within the right content (for freshness, oldness, whatever).
Some folks get this intuitively, like Russell:
"... Though it's nice to see that mainstream portals are jumping onto the RSS bandwagon (lead by my employer), the problem is that all of these services generally suck.
Why? Because they all break a very simple rule: You should only see an RSS item once.
Read marks and session management is the key to aggegators IMHO. As a person who scans almost 400 feeds daily, I can tell you this is the only way realistically keep up. Even aggregators made for the general populace, who may only keep track of a dozen or so news sources, not providing this functionality is just wasting their time. Though actually, I think that many people start out with a small list of feeds and just keep adding to them. Why not give them a scalable solution right away?"
I really recommend reading his post - it's killer.
The comments also debunk a lot of not-so-sensible numbers coming out these days, which should be intuitive to people involved in this space (Yahoo dominant in the feedosphere...come on):
"...But let me throw some cold water on those Feedburner market stats. A couple of weeks ago Datamation did a deconstruction of Feedburner measurement (confirmed by Feedburner) that determined that MyYahoo's share was inflated because of the default RSS subscriptions that are loaded into every subscriber's account. Whether they know about them or use them or not. If you even the playing field by ignoring the Default 10, MyYahoo plummets to 6% and Bloglines is in the lead with about 20%."
It's amazing that Bloglines still offers the superior value proposition, despite the resources that Google and Yahoo (etc) are throwing at becoming feedosphere players (and despite the fact that Ask now owns it). And if you think about it, even Bloglines hasn't really added much to it's value prop in a long time - for some reason, people are struggling in this space.
I think it's because most of these players fundamentally don't understand Media 2.0 economics, but you knew I was gonna say that :)
Umair, if you're looking for something that filters the right content from the wrong content, I'd be interested in hearing your thoughts on Findory.
Findory learns from the articles you read, searches thousands of feeds, and surfaces other interesting articles. It just takes a couple clicks on articles to get it started.
// Greg Linden // 4:46 PM
Post a Comment