Anne Truitt Zelenka

Adam Green questions the prevailing wisdom that a reading list should include less than ten blogs. I can see uses for both, but I’m far more excited about big reading lists right now.

Small reading lists match the current news reading paradigm, exemplified by the three-panel browsers and river-of-news displays. But big lists, lists with 100 or 200 or 1,000 feeds in them, those provide fodder for displays that aggregate, filter, summarize, pivot, contextualize, and personalize my news. I hope that such next generation meme trackers whether individual or community-oriented will replace the information-poor displays we suffer with now. It means almost nothing to me when I see I have 15 new posts in my momblog folder, 11 in my current favorites, and 65 in other places. Without punching into specific feeds, I don’t have any useful information there, other than knowing that, as usual, I’m overloaded. Contrast that to checking tech.memeorandum. I see at a glance what many of my favorite bloggers are blogging about, and which topics are most popular. I am exposed to topics I might not otherwise follow because they’re of interest to the greater tech blogging community in which I work. Memeorandum, Megite, and Chuquet work because they limit themselves to select sets of blogs–reading lists–instead of scanning the whole of the blogosphere. But these are not small reading lists, if they be reading lists at all.

Yesterday I started creating a Reflective Parenting reading list, an OPML file hosted by BlogBridge that includes feeds with which I’m familiar that address parenting beyond the bloggers’ personal experiences of it. It’s an embryonic list, because there is no existing community of parent bloggers that link to each other with the wild abandon of tech bloggers. The purpose of this list is to serve as a starting point for the Momorandum community meme tracker I’ve been thinking about. You can already see Megite’s parenting page based on the Reflective Parenting list. It only includes thirty or so feeds right now, and that’s not too many. It’s probably too few. For this sort of approach, big reading lists are better.

Perhaps it shouldn’t be called a reading list if it’s used as input for an app that filters, rotates, and distills the contents into something that’s information dense. It looks like the same thing–a dynamic OPML file hosted at a URL–but it’s not used in the way we’ve been talking about using reading lists. Whatever you call it, this is the sort of thing I’m seeing great use for, but only in combination with better options for finding important conversations and concentrating them into a format that readers can grok quickly. The big challenge I see now is developing ways beyond simply looking for links to identify important topics. That may work well for politics and tech, but different domains need additional filtering and identification mechanisms, like comment counting and text analysis, perhaps even some Bayesian classification to make the news as informational as possible.