Back to Mono

abbeyI went to a birthday party this weekend where I ran into a Facebook guy, a smart guy who asked me to go off the record. In fact, the whole party was supposed to be off the record. So I ignored the off the record part by insisting that I already knew the thing I was being told, and then I told him on the record what I thought was about to happen for Facebook. This being my usual m.o. which is to insist on not being NDAed except for things I don’t really want to talk about anyway, like the next version of Office.

That way, I can just make up what I want to have happen, never breaking any confidence and yet at the same time painting as plausible picture of assumed reality that it is hard to deny or in fact slow down. So here’s what I told the Facebook guy: the company has at most 3 months window to absorb FriendFeed and open the Everyone News Feed, and if that’s true (again, making all this up) then the messaging about how that’s going to work must begin immediately, like in two weeks. Then I went home and saw MG Siegler’s post and Scoble’s remake of Frenzy on FriendFeed.

OK, so I was off by two weeks. The noise about the death of FriendFeed is already off the charts, and the proof is in the lack of rejoinder from the FriendFeed team. As in: of course FriendFeed is not dead, and here’s what we’re going to do to remake Facebook in the next few weeks. Actually, that is indeed the message from Twitter, what with Lists and ReTweets and the return of Track just as soon as, well, sometime next year or so. No need for FriendFeed real soon now, because these Lists will soon be carved up and meshed together into an authority stream by the 3rd party developers.

Siegler nails the one provable negative about FriendFeed Facebook edition which is the lack of any innovation moving forward. The one thing the FriendFeeders didn’t get in under the wire before the money arrived was stream splicing, the ability to mesh together lists into an authority stream. Is that coming soon from Twitter either? Nope. So the antidote to FriendFeed stasis is Twitter right up until stream splicing is enabled… by who? As of right now, that would be via the FriendFeed APIs. If Siegler/Scobler are right, the danger of doing that is iterating on a dead API.

Here’s where the FriendFeed is dead rumor falls apart. OK, you’re Facebook and you’ve just spent $50 million or actually $15 million plus NetWhuffie stock. Now we sell the deal as a talent buy, which of course it is because the talent built the damn thing. We put Brett Taylor in charge of the platform (API) and Paul Buchheit in charge of something else he hasn’t said yet. So Taylor can still deliver stream splicing, just not in the FriendFeed context. Buchheit, the Gmail guy, now what do we do with him….

Remember, where were the FriendFeed guys when the clock was stopped. They were streamlining not just the API but the architecture of what used to be rooms and what was now groups. Subsumed into that construct was the wonderful Imaginary Friend notion, another way of saying how do we capture individual streams and normalize personal and group communications. Meanwhile, RSS is dying and with it readers of same, and we begin to see seedlings like Threadsy and Brizzly popping up to address the vacuum. Can we assume the Gmail guy might be in a good position to noodle down on this, particularly given the Google Wave fork under way across town?

Obligatory Beatle analogy: I’ve spent the last few weeks partying on the Beatles Mono Mixes, in particular the period that began with Rubber Soul, followed by Revolver, and then jumped ahead to the White Album and its companion Part II, also known as Abbey Road. In just two years, the group transformed themselves from lovable moptops into four individuals who wrote music to avoid board meetings with accountants (Something) and seers who mixed Tibetan chants with backwards tape loops (Tomorrow Never Knows.) At the time it seemed like a lifetime, but in mono as they blended this stuff together it takes on a sense of purpose and inevitability that belies the official storyline.

In effect, with Abbey Road they tied up the loose ends of the White Album’s API and architecture, injected the realtime nature of Revolver and the betrayal of Norwegian Wood, and sold the company to EMI on the promise of new records from a group that had already broken up. Listening to the mixes and reveling in the deconstruction of Beatles RockBand, you can see how intricate their alchemy became during those brief two years, and how valuable it would become for the ages. As David Crosby said recently of a CSN project to do cover versions of favorites, when it came to the Beatles, they’d work up one or another from memory and then give up after listening to the original.

This then is what Facebook bought, or rather invested in: the best work and that yet to come from this group of engineers, strategists, and explorers of realtime. It’s easy to forget how completely wrongheaded it was to attack the realtime experience, how to this day Wave is reviled as an unbridled solution to a problem that doesn’t exist. What are people afraid of to fuel this alleged stampede for the exits, particularly given that there is no credible replacement for many of its most addictive features? The very hysteria of the charge, a cry for no one, a love that should have lasted years.

More likely is that we’ll see a healthy battle for the legacy of FriendFeed between Facebook, Twitter, and Google. First will be the conversational flow, still a pathetic hack in Twitter, possible in Facebook only with a fast (three month) Everyone newstream rollout with 2 way Twitter/Everyone stream sync. Never mind the Web site; this has to come from the API, complete with granular tools for filtering the flow by data type and guaranteed RSS-hubbed delivery. Facebook must invest quickly in stream splicing to make filtering useful, and Twitter Lists give them the room to build it while no such tools exist on the Twitter side.

Second, if there is a second after that first takedown, is splicing email and personal FB data into the stream, through a combination of intelligent push into email for filtered stream data for mobile clients and harvesting of the Twitter follow and List social graph as an open directory outside Twitter’s control. Google has a play here with Gmail, but they need to drive here with Wave taking a backup role rather than the other way around. Buchheit is the key here, regardless of who does it, and Twitter’s lock on direct messages needs to be factored in in the Facebook sync planning, something that didn’t survive the cut in the FriendFeed lockdown.

Yoko didn’t break up the Beatles. The Beatles did. FriendFeed isn’t dead. It’s just getting started. And the walrus is Paul.