Some thoughts on Kinja and the value of a blog post

I played around with Kinja a bit this morning (here's my public digest). As I was browsing through other people's digests, I got the same feeling I get from looking at the group newspaper in FeedDemon: it's all Boing Boing and Anil's daily links.

In a Kinja digest, link list posts and other outboard brain-type posts get the same treatment as regular blog posts. Which is fine for the aggregator--from an aggregator's perspective, a post is a post. But from a reader's perspective, some posts are more important than others.

I read Dive Into Mark and Boing Boing regularly, and I'd say they have roughly the same value to me as weblogs. But clearly a Dive Into Mark post is worth than a Boing Boing post. Over the last couple of weeks Boing Boing published with about 40 to 50 times the frequency of Dive Into Mark. Assuming a direct relationship between the post values (for the sake of argument, anyway) one Mark Pilgrim post would be worth 50 Boing Boing posts. My gut tells me that valuation isn't quite right, though I think the principle it implies is correct: I don't want a high-value post to get lost in a low-value linkstorm.

That's the reality with Kinja (and the FeedDemon newspaper, too). Part of the problem is that reverse chronological order is the only display mode, which means high-volume publishers will tend to get more screen space than low-volume publishers.

Having gotten over the hump of basic syndication and consumption, I think the next generation of aggregators will look at things like publishing frequency, inbound links, post length, recency and other factors to make what are in essence editorial decisions. Some sort of algorithm will determine the average post value for each weblog. Lower-value posts--like link lists--might be aggregated into a sidebar (link lists would be a perfect place to use an intersection filter). High-value posts would appear at the top of the page. Reverse chronological order would become less important. Redundant links would be filtered out.

As good as Kinja is (I found it very easy to use), I think it needs to be less bloggy and more like a newspaper to really succeed as a layperson's aggregator. Not literally like a newspaper, of course, but like a newspaper in that someone/thing decides which stories make the front page and which are buried on B17. It needs an editorial engine, if you will.