Last Friday was the Techcrunch Crunchup conference about the Realtime web. It was followed immediately by the annual TC party at August Capital. Of the many conferences I’ve attended, Crunchup was more interesting and relevant than most. It kicked off on a high note with a discussion between Ron Conway and John Borthwick, who were pestered for financial secrets by Michael Arrington and the occasional bark from Steve Gillmor.
Why Steve must stand alongside a panel every panel I’ve seen him at escapes me — unless it is to exercise his signature interrogatives from sidelines at a distance from which it is safer to lob remarks. Neither Mike nor Steve are the most supportive interlocutors you’ve seen on stage. But the tech community sets its bar for live interaction pretty low… I digress…
I wanted to capture a couple reflections on the day’s demos and panels. Without naming companies by name, there was a clear preponderance of feed readers and aggregators. It struck me that the industry’s first response to the realtime web is to address disaggregation of conversation forced by the many sites, services, and practices that currently produce and support conversation. (Think commenting, status updates, twitter, and the fact that most of these can be commented on at their source, on sites they are syndicated to, or even in some third party apps).
In other words the problem created by the realtime web, and addressed by many of these apps, is an information problem. Clay Shirky has been on top of this of late, arguing that the realtime web creates an abundance of personal, social, and emotional news and content. Robert Scoble, whose behavior and lifestyle often speaks for itself, in volumes, gets the flow filtration problem.
Clay sees the distortive human bias that can creep into a pattern of social media use. Scoble sees the patterns in the content (earthquakes, Michael Jackson). For Shirky the pattern is in mediated audiences; for Scoble the pattern is in the content they circulate (news).
Both approaches clearly work at some level to reduce the complexity of massive amounts of conversational content. But the content produced by people talking on twitter, declaring their status in Facebook, or liking and commenting in Friendfeed is not just content like any other kind of content. It’s talk: often addressed to one or more people, or at least shared with an audience in mind (followers). Most tweets make some kind of appeal that audience: appealing for a response (RT, @reply), for a follow, a comment, a click. If not in the tweet itself, the practice of tweeting in general begs for social validation and acknowledgment.
In other words the realtime web, insofar as its content is the content of a strange kind of talk, is not the same kind of content as regular web content. It’s neither news nor information — though it may be news and it may be informative. Those forms of content hail from broadcast media. I would insist that talk produced by people aware to some degree of how they appear to others, engaged in the activity of maintaining a reputation, of sustaining relationships, and of talking in an interested way, leave behind content that’s of a different nature than the stuff that comes across the newswire.
Or just think subjective vs objective.
Point being that realtime web content issues may be addressed by filtering and aggregating (or re-aggregating, as JS-Kit calls it) incoming feeds to reduce noise and increase relevance. But does this approach leave the conversationality of realtime web interactions unaddressed?
The realtime web may mean access to social news (twitter, Facebook, etc) in realtime. A shift to realtime flows and content streams. To news and updates, and to the trends that will naturally accompany real-time distribution. But realtime access is not the same as synchrony. Synchrony is the simultaneity of two or more independent actors — synchrony in conversation allows us to coordinate communication. We all know that twitter is “realtime” but asynchronous. We get content in realtime, but don’t assume that others are in twitter at the same time.
Is this emphasis on the information problem of realtime just a common-sense approach, given that it’s an extension of conventional web search, filtering, sorting and results display practices? Or will there be another wave of companies — those that treat realtime web as a conversation space, focusing perhaps on presence, availability, relationship, similarity, affinity, graph and so on?
What’s in the Venn overlap of these two approaches, where content is people but people make content, and where affinities can produce trust, confidence, similarity of identity, interest, communication, and sometimes relationships? I didn’t see much by way of social browsing, social serendipity or discovery. Personally I don’t benefit as much from realtime content as I do from tweeting with friends.
I want to know more about the latent connections and relationships between people. When I’m realtime I drive, I do not use readers but tend instead to browse when I’m interested in finding something out. Constant exposure to realtime social commentary paralyzes me. Most panelists of course acknowledged that we’re in early days yet. The easiest problems to solve are indeed aggregation, filtering, and search — and that’s not to diminish the engineering or user experience challenges there.
I’m fairly certain that the treasure in realtime web is not in content but in the relationships, in the meaning of content as revealing disclosures of trusted, influential, respected, credible, authoritative, and liked relationships among friends, peers, and strangers. …Of course that content surfaces only if the tools supporting online social interaction preserve the human and personal aspects of the user experience.