Feeder's Digest

feedersNow that Facebook has jumped into the activity stream, how long will it be before major platform vendors do the same? Google seems strangely quiet except for a few retracted comments from Eric Schmidt about Twitter being a poor man’s email. Speaking of poor man’s email (aren’t we all these days) Microsoft has taken a huge chunk out of Notes engagements with its on-demand Exchange Online product. But so far, no direct attacks on Twitter et al.

The Facebook announcements seem most threatening to FriendFeed, at least from a feature perspective. In recent weeks, Facebook has sprouted a number of FriendFeed constructs – Likes and Comments the most obvious. In a few cases, the new tools go a step beyond FriendFeed by making it easier to post rich media types. But there’s a catch in this ease of use, namely the Roach Motel-ian question of whether the data ever comes out.

The Facebook premise remains intact for the moment – that the visibility of data is tied directly to the social graph. As a competitive assault on Twitter, this stands a chance of working: Twitter is as closed outbound as Facebook. Both companies make the same noise about protecting users from unfettered access to user content. Twitter doesn’t tie it to a Twitter Connect the way Facebook does, but they use rate limiting and obfuscation about Track (realtime search services) to keep the data scoped to users’ Follow clouds.

No matter how fast Facebook opens the stream, at some point they have to allow the data to travel into the river and down to the sea. In doing so, they would cross the chasm so many see as daunting, the one from a walled garden to a feel-good Internet freedom approach. Never mind that we still see no music business emerging from such an emancipation, no federated open source micromessaging model, no media disruption, no free lunch.

That doesn’t mean disruptive business models will or are or have emerged. The App Store may be our first glimpse of the Phoenix formerly known as the newspaper business. Amazon’s Kindle/iPhone app is the harbinger of the music business salvation, when the Gang of Four realizes they can give away the music if they license the liner notes. Right now the iPhone/Kindle platform is a hardware-dongled version of what Live Mesh will soon spread across the the majority of devices.

This is analoguous to the late great print magazine, where editorial decisions and advertising dynamics create a portable container for information. The Kindle by itself suffers from being tethered to the long form of books and deeper analysis, but if you add the iPhone as a way of integrating Web media and upselling updates and breaking news, you achieve something similar to the purpose if not the actuality of the previous magazine era.

In this mashup, the stream becomes more central, but it also introduces the bifurcation of publishing and messaging. We spend a lot of time hemming and hawing about Twitter/Facebook business models, but here we have Amazon apparently giving away Kindle sales to the iPhone population. Do we have to own the Kindle to download and consume the books? No. Do we have to pay subscriptions for Web content? No. What gives?

What emerges is a new magazine model, where aggregators provide efficiency and editors context for the combination of static and dynamic information flows. It’s a new kind of album format, not the quick bursts of hit singles but the loosely-bound orchestration of themes and releationship that informed Blonde on Blonde, Revolver, and Electric Ladyland. The iPhone becomes the razor blade, the Kindle the razor.

If that’s true, look for the razor to be subsidized. Look for the stream to be subsidized. Let’s say Facebook and Twitter keep control of the outbound data, tying it to permissions either from the social graph or in Twitter’s case, the business alliances with a group of favored developers who share the 60 requests an hour allowed to each customer.

In the Facebook example, the user must do the work of in effect creating their own magazine, structuring Friend Lists and filters to produce an efficient stream. Of course, that is exactly why magazines emerged, to proxy that service to specific groups. In the Twitter example, the company becomes the publisher, creating the kind of channel conflict that has challenged most tech companies with the conspicuous exception (so far) of Google.

The net result of Twitter’s strategy – playing favorites with developers who protect a business model that hasn’t been clearly vetted with customers who aren’t identified yet – is that Facebook actually has an opening to provide publishers with a new store front that isn’t shrouded in mixed signals. This is Facebook we’re talking about – Beacon, TOS, yadda yadda – suddenly wearing the white hat by comparison. Wow. They should be paying Twitter the $500 million just for being the Washington Generals.

But Facebook has a big problem with the volume and weak typing of its notion of friendship. Zuckerberg’s blog post aptly illustrates this with an endless stream of fweets that takes some 30 seconds to load the page. For all of Facebook’s cloning of FriendFeed, they have no way of providing what Twitter used to create a much more valuable social graph: Track.

Where Facebook’s new strategy allows constrained Friend Lists for realtime updates and unlimited Follows of publishers, Track allows users to maintain an asymmetrical balance with both specificity and unanticipated messaging. We get the best of both worlds, realtime messaging and social graph-aware editorial context – and with Track comes the filtering tools we need to reduce broadcast noise.

Can Facebook do this given its fundamental security by social graph model? Can Twitter do this without undermining its simpler value proposition and forcing the company to make the case that it owns the data it’s collected and can control its spread across the network via its third parties? Recent rumblings about applying Creative Commons licenses to incoming content have to be sending shivers up Ev and Biz’s spines.

Meanwhile, FriendFeed continues to provide unfettered access to its engine and increasingly powerful tools to model the kind of filtering Facebook users will need to employ to reduce the flow of the new stream. Already data is flowing slowly (~30 minutes at times) from Twitter to FriendFeed, while virtually immediate in the other direction. If observers like Marc Canter and David Recordon are right, FriendFeeders might benefit from Facebook opening up its outbound update stream at realtime speeds to encourage people to multiplex Twitter posts into Facebook and then to FriendFeed.

Whoever solves the problem of freeing the stream while protecting the user’s right to be in control of an efficient digest will surely plug into the Kindle/iPhone Mesh or whatever competes with it. For now, let’s call that publication Feeder’s Digest.