BearHug Camp proved successful at bringing together a wide swath of what is known as the open micro blogging community. With companies including Twitter, Laconica/Identi.ca, Facebook, Google, Microsoft, bit.ly, and Seesmic/Twhirl, some 1700 of those who couldn’t make it in person followed over Leo Laporte’s TWiT-TV Live streamcast. While most media folks took their leads from Twitter execs Jack Dorsey and Biz Stone’s comments on the return of Track in the morning session, the work done in the afternoon may prove more significant in the long run.
The Twitter conversation took the form of a press conference, with BearHuggers firing increasingly detailed questions about Track functionality and its attendant XMPP transport. In a nutshell, Twitter execs flatly stated Track would return, but also committed to allowing third-party developers such as Dustin Sallings (TwitterSpy/IdentiSpy) to be whitelisted for 5-second interval access on an unlimited basis to the API. The mechanism for that permission involves an email to Twitter developer Alex Payne, who promised a 48 hour turnaround.
Although there was less definitive discussion about restoration of XMPP services, the release of terms and conditions for such access, and the business discussions required for use of the firehose stream of all status messages, the API whitelist gateway should prove sufficient in the short term to restore Track services to Twitter users in time for the first presidential debate on September 26. It was the the early primary dates and election night coverage that triggered Twitter’s earlier demise and subsequent loss of Track and IM.
Even those who see Twitter and similar services as the cornerstone of what they call the Now Web depend on the assumption that such real time services will have to achieve a context similar to that of email in order for the technology to prove its value at scale. Although IM has proven itself a powerful attractor of eyeballs and stickiness, it has failed to be directly monetized like messaging’s Microsoft and IBM juggernauts. By contrast, even open source browser projects such as Firefox have become powerful revenue engines in relatively short time.
BearHuggers seemed uneasy at the driveby status of the Twitter floks, a bemused combination of surprise that they showed up at all and irritation that they didn’t stay to hear from the rest of the community. Jack Dorsey had in fact made it clear privately that he could only stay for a brief time, and bringing Payne and Stone went beyond a mere photo opp. Perhaps more interesting than the public conversation were a series of casual huddles after the segment, as Dave Winer and other independant developers chatted casually with the trio. Winer played an important role in arranging the visit, and seemed engaged with the group.
I’ll leave a more detailed report to Jack Moffitt’s excellent notes, but even those observations carry with them the underlying sense that the heavy lifting will have to be done not by Twitter but the open community. The afternoon session drew out Evan Prodromou’s snapshot and roadmap of the Laconica effort, as well as Seesmic founder Loic Lemeur’s efforts to turn Twhirl from a client side to a server side hub.
But there was an odd ambiguity about much of what Prodromou committed to in sketching forward from the recent XMPP bandaiding he’s been doing to a more stable unified framework. Moffitt announced he was building an aggregation service using XMPP pubsub “for the OMB community” and layering Track on top of it. Prodromou continues to insist he will build Track into Laconica in spite of calls to leave that to an independent middle layer of developers, though he appeared to push it down the stack when challenged. And Sallings, Lemeur, and less visible players continue to work in groups of one or two on virtually the same strategies.
This may seem to be good news for Fred Wilson and others who hope Twitter’s lead in users will translate to more control over the marketplace when it emerges. But fast-moving startups such as Yammer could easily use a solid enterprise base to leapfrog the open source players and challenge Twitter in bringing developer resources to bear in the short term while the market is still susceptible to momentum plays. It’s this threat that I believe has encouraged Twitter to begin the process of re-engaging with third party developers, much like Apple has done with iPhone developers.
But unlike Apple, who has little to fear from anger about applications clearly designed to disrupt Apple’s iTunes and carrier bearhug fundamentals, Twitter does have to harness the open micromessaging space or risk its size advantage in market share being dwarfed by either Google or Microsoft or even Salesforce. Any one of these could realize the efficiencies of giving away an open service as an attractor for next generation relationships with key affinity groups now reaching maturity as market forces.
Google Chrome could be the most obvious disruptor of Twitter’s cloud, simply by embedding open micromessaging services that do not compete or intermingle with Gmail or Google Reader, whose behavioral data they have promised not to cross-collateralize. Firefox Labs services seem experimental and immature, and Google already has the Jaiku team repositioned on top of AppEngine (just as Identi.ca is deployed on EC2.) We’ll know more about Mesh in a month at the PDC, but try as I do to get someone at Microsoft to argue with my description of Mesh as a pubsub router, not a word is said to the contrary.
But even more threatening than the eventual Borging of micromessaging by either the new or the old Microsoft is a trojan horse attack from within. If messages from and to Twitter were indistinguishable from those of other services, the open source community argues, the economies of scale and free would eventually overwhelm the proprietary play. If Twitter rolls Track out as play for pay, the price creates an incentive to undercut it or do a Google Apps in order to disrupt the incumbent’s relationship with its investors.
This scenario depends on the existence of the OMB, however. Twitter execs appeared to rule out a level playing field, using the example of guarding against a tweet that makes sense in one context appearing in another context where the information might not seem so appropriate. Proffered was the notion that people who tweet in the clear may not realize just how public this is or could be. In all of the commentary in the room and over Twitter and Identi.ca I heard not a word about how the “community” felt about this issue.
If there can be no community understanding about this most sensitive of issues – whether people realize what’s good for them or not – then how will any market standard evolve? Turn the argument around and examine it from the perspective of someone like me who is more concerned about not having my ability to send messages impeded by business model, political correctness, or any of the other gating factors that are used to intercept, control, and tax our networks of communication. What does the email community do to our messages, and can that be applied to micromessaging?
First, there’s spam, which we define not as a civil rights issue but a civic form of garbage collection. Do the political parties get treated as spammers when they use email as a sort of global franking privilege. No, they leave it up to us to define by hitting the spam button when it reaches the intrusion point. Does that spam filtering work point to point within messages from friend to friend. Not likely. But Twitter appears to be using concern about the community as validation for terms of service that validate a business model. Fair enough, if they can get away with it. So where is the so-called open community to offer alternatives?
What is Laconica’s stance on sharing of public content in inappropriate ways? How does Twhirl justify or limit undesirable content flowing over its multi-service transport? If the BearHug has pretensions to work, doesn’t it have to start with unifying standards about what the value and risks of micromessaging are, rather than arguing over the inevitability of one or another business model? And when better than to hash these issues out than at the start, before economic leverage makes real discourse more difficult.
Ironic, isn’t it, that Twitter has raised this issue first. Certainly it can be used to create a rationalization for picking partners, moating data, or any other barrier to entry. But taken at its simplest, micromessaging needs to walk before it runs, and strategically, respecting the rights of the individual at a very early stage will do more to create power than all the market force momentum that can be mustered.
Only when we recognize Twitter’s willingness to engage with the open community will we understand how much that puts the onus on the open source community and standards wonks to come to the table themselves. BearHug Camp should reconvene inside the Trojan Horse, and once aligned around real issues, emerge to inject itself among the users in the city.