Communications Nerdvana: Next Year, Maybe

I have been smacked in the face with a battery of new flow apps. There are so many popping up I can't even adequately fool with them.

David Chartier turned me on to Mozilla's Snowl, and sets context pretty well:

[from Mozilla's "Snowl" hunts Twitter, RSS, and (soon) e-mail]

The web is an increasingly chatty place. Between following comment threads, checking in with friends on Twitter, reading a few blogs with RSS feeds, and conquering a mountain of e-mail, we have plenty of conversations to keep track of. Mozilla wants to help us on the conversational journey, which is why the company has launched Snowl, a new tool that offers a small glimmer of hope for those seeking communications nirvana.

But Snowl immediately disappoints. The post on the Mozilla Labs blog calls it an experiment, but it is a failed one, in my case, although I agree with the drivers:

[from Introducing Snowl]

Snowl is an experiment to answer that question. It’s a prototype Firefox extension that integrates messaging into the browser based on a few key ideas:

1. It doesn’t matter where messages originate. They’re alike, whether they come from traditional email servers, RSS/Atom feeds, web discussion forums, social networks, or other sources.
2. Some messages are more important than others, and the best interface for actively reading important messages is different from the best one for casually browsing unimportant ones.
3. A search-based interface for message retrieval is more powerful and easier to use than one that makes you organize your messages first to find them later.
4. Browser functionality for navigating web content, like tabs, bookmarks, and history, also works well for navigating messages.

I think that moving towards a messaging basis instead of a publishing basis is dead-on (see Blogging 2.0 Meme Doesn't Go Far Enough), but getting stuck in the email inbox or RSS reader paradigm is not helpful.

If it's all about conversations being supported by messages -- and who cares where the messages are generated, whether Twitter, blogs, email, whatever -- then we need to adopt a conversational paradigm, something more like comment threads, instant messaging, or Twitter.

I had stumbled across Rejaw yesterday, which is a Twitter application. They seem to have twigged on a useful distinction between 'shouting' -- where a user is speaking one:many -- in distinction with 'whispering' -- where two (or more?) people are speaking privately. Here's is a shout that I created, and a series of replies.

Rejaw: Shout from Stowe Boyd, originally uploaded by Stowe Boyd.

Rejaw goes farther than Twitter in tracking the context of a Tweet, and trying to manage visibility around the intention of the speaker. But I rapidly got lost in the twisty little tunnels of the user experience. I replied to a comment from someone, then couldn't figure out how to find the message I had created. Maybe a shout v whisper pane approach, like Tweetdeck, is needed?

Communications nerdvana was exactly what Greg Narain and I had proposed to AOL at the beginning of 2007. I was approached by various folks then working there, based on a series of posts I had written in 2006 and 2005. We kicked off a project, and designed a solution that has never been built, but I think some of its features -- although geared in principle to a large instant messaging network like AIM -- could help the conceptual mess we seem to be getting into with these conversational matrix applications. I offer these observations in open discourse, although if anyone would like to get our help, I am sure that Greg is as open to it as I am.

In the Nerdavana project, Greg Narain and I designed the basic backbone for the application around the model of a buddylist (specifically AOL's AIM buddylist). This lines up with my belief that all media is becoming socialized: all media will be perceived and delivered as relationship-based messages, where the source of the message is as important (and generally more important) as what is being said.

Nerdvana Conceptual Design - Buddylist, originally uploaded by Stowe Boyd.

The idea here is simply to aggregate all the myriad streams from all of your buddies, and to be able to respond to them in a simple way, even if the world is fragmented. Note that Nerdvana is status aware: it knows that you are reading something, or commenting, or whatever, and if you want that status is automatically transmitted as status. (I believe the conflation of status and messages is a mistake in the Twitter model, for example.)

If MadMonkNYC has made a comment on some third party's post then if I want to respond I am confronted with the option of a/ going back to the post, b/ sending a message -- email, Tweet, instant message, whatever -- to MadMonkNYC, or c/ creating a linked comment within a conversational matrix tool like Nerdvana (or Friendfeed, Rejaw, or Socialmedian). (One irreducible problem is that these world are not interoperational, and they involved divided communities, but that is a different issue, one that I started to address in yesterday's Blogging 2.0 Meme Doesn't Go Far Enough post, and one that I will return to in another post.)

Greg and I also devised ways to look at the 'pile-ups' in Nerdvana -- when various individuals begin to talk about the same 'topic' -- which could be an event, or a web post, or a direct interchange like an person-to-person message (like an a tweet directed to @gregarious in Twitter). We called this the 'Stories' view, which is similar to the Techmeme memetracking motif:

Nerdvana Conceptual Design - Stories View, originally uploaded by Stowe Boyd.

Here you see a group of Nerdvana users have commented on the same post, and rated it. We had also devised a model where users could dial up the degree of how wide the sources for their stories might be: 100% from my closest friends, 100% from the most popular in the entire user community of Nerdvana, or some degree in between.

The RSS reader model might seem to be almost the same, but it treats people as sources of RSS, which is the wrong layer of abstraction. People are conversational engines, and can opt to converse through any number of media, and these should be blended into a person's conversational flow. RSS readers generally do not support conversation, and may not even support finding them when they are happening, which is why reading Techmeme is often more illuminating that Google Reader.

In the final analysis, we have a number of contending models smooshing into each other, but some are inherently more rich and will therefore rise to become the defining warp and weft of the conversational matrix that will replace the publishing model of RSS. There are a large number of challenges, however, and a great deal of retrofitting will have to be contrived to keep the Web a contiguous, even if a fragmented, experience. Tools like Nerdavana don't solve the fragmentation problem, but they do elevate the experience of web involvement from reading to conversing, which is a millennial step forward.