ONE can still see the remnants of the housing boom across America: developments in which streets were platted, concrete poured for the foundations of homes and pipes laid all over for sewer, water and gas. One might even see a partially erected wall silhouetted against the horizon.
The same phenomenon recurs regularly online, where services build like mad, but the population growth slows or never arrives. When the bust comes for a given offering, the end sometimes comes quickly. It staggers from burning funds ahead of revenue (if it had any), and then shutters, leaving not even a virtual footprint.
A promising service awkwardly named App.net, which combined social-networking features with robust and unconditional support for third-party software developers to build on top of it, announced this week it will lay off all its full-time staff, yet not close its doors. It is in the peculiar situation of being almost, but not quite, popular enough to thrive.
App.net’s founding intent was to create the user registration, network and server plumbing that would allow third-party developers to build robust communities on top, much like developers buying plots of land in a town without running the utilities and the post office. Rather than be a clone of Twitter or Facebook, App.net set itself in opposition: it would never sell advertising to its developers’ customers, would rely primarily on fees paid by users and wouldn’t interfere with software firms’ apps that required its infrastructure. (This Babbage explained App.net in September 2012 in an extended metaphor that swept in the American prairie and homesteading.)
Developers turn to social networks for identifying a user, transmitting messages between users or among members of a group, broadcasting information widely and sharing rich data, like photos and video. Different aspects get combined in an app, such as games that allow one to play among members on one’s social graph (the interconnected web of one’s contacts); or photo-sharing programs that authenticate a user’s identity on one or more services and post on his behalf.
App.net was created in mid-2012 in opposition to shifting sands at Facebook and Twitter, in which third-party developers were (and still remain) at the mercy of those firms’ business models. While App.net released a Web app to allow microblogging like Twitter, and hundreds of software products appeared (and stalled in various states of completion) to offer that interaction on desktop and mobile platforms, its real potential was in areas in which developers needed reliable services that they would otherwise have had to build themselves and which would have benefited from a network effect. These included offering storage, a la Dropbox or Google Drive; and a notification service released a few months ago that had much in common with single-platform options in Apple’s iOS and Google’s Android. Favd was an App.net-centric app your correspondent particularly enjoyed. It both showed a stream of images from those one followed on App.net, while also allowing cross-posting images to App.net, Facebook and Twitter in a single operation.
The trouble was that for App.net to thrive, it needed a critical mass of users that would make the infrastructure platform attractive to software firms, which would in turn market it to users, and so on. Often, a killer app is needed in which the utility and desirability is so high that it overcomes the friction of registering for yet another online offering. Despite adding a free tier of account with significant limits, paying developers from an incentive pool based on a combination of usage and user feedback and the regular addition of new features, no such app materialised. App.net says over 250,000 users ever registered; only a subset were paying members.
After the renewal date had passed in February 2014, when original users (including your correspondent) were prompted to pay up for the following year, many waited nervously to hear the verdict on App.net’s future. The founder, Dalton Caldwell, wrote in a blog on May 6th that the “good news” was that renewals in February would fund all the operations of App.net, such as hosting and operations; the bad, that all its staff was being cashiered. Mr Caldwell and a partner, who have stakes in the company alongside venture-capital firms, will continue to provide service gratis and hire contractors to keep it ticking along. It will end its incentive program.
But it is hard to see how App.net, even in a maintenance-plus mode, in which some improvements and extensions are promised, can possibly persist beyond current subscribers’ expiry. The problem that the offering sought to solve remains. Internet companies would love having a modest utility company as a partner that sought consistent returns rather than a flash behemoth that might snap its fingers and destroy them. The stars didn’t align. Many of the people Babbage first met there have migrated to Twitter; tumbleweeds now seem to have drifted into town.
Correction: The article originally stated that the total number of registered users at App.net was 55,000; that is, rather, a third-party estimate for the total number of active users. This article has been corrected to reflect App.net’s statement that it has 250,000 registered users, both free and paid.
Powered by Facebook Comments