Categories
The Longform

The Paid Social Network Nobody Wanted11 min read

In the early 2010s there were two mighty giants in the social media landscape: Facebook and Twitter. One behemoth was scrapping its way in an attempt to be a third. This company was backed by lots of money, a thoughtful design, and a single focus to topple the others. It wanted to be king.

But this is not about Google with their Plus endeavor or even the weirdly-limited Path wherein you were only allowed 50 friends. This is the story of a scrappy startup who set their sights on Twitter. A company who felt that enough people were willing to pay for a social network. It was a company who learned the hard way that the opposite is true.

This is the story of App.net.

Problems Abound

It’s no secret that social networks have a lot of problems. While Facebook Meta gets its deservedly-fair share of flack recently, the early 2010’s were a place where Twitter was changing. Twitter was a tech darling from its initial launch wherein it was strictly used via SMS. In the following years with astronomical growth, Twitter became a place that drew in the bad along with all the good. Every few months would be another dust-up of people calling for Twitter to do something about it. It didn’t help that Twitter had been adding lots of advertising into the service, which annoyed people as well. They also turned their backs on developers making apps for the service. Twitter, like many tech companies, focused on growth and generating money via advertising. Interestingly one idea that kept gaining traction among Twitter users was a paid version of the service.

The thinking around a paid tier of Twitter went that if people had to actually shell out money to use it, the advertising would be gone. Also the majority of the trolls and hate-mongers would go elsewhere. Of course today Twitter remains free, but they recently rolled out their premium Blue platform. That indeed costs money.

During this time people loved how Twitter worked between the ease and restrictiveness of posts, using hashtags to follow topics you wanted, and curating ones feed by only having people you follow show up. It was different from what Facebook was doing with their news feed. People also wanted the ads to go away. People wants great apps. One person was listening to these repeated calls for a “better” social network. In 2012, he hit the ground running.

Proving Grounds

Dalton Caldwell posted an “audacious proposal” to his website on July 13, 2012. In his seven page entry he writes about the difficulties and failings of modern web companies. He saw the status quo as a path one we should not go further down. In this post he proposes his solution: a new social network called App.net. It would be developer-friendly, allow the free flow of information (similar to Twitter), but most-important it would be a paid service. App.net would have no advertising. He even posted a series of videos outlining his vision.

Caldwell believed in his vision so much, he was willing to put the time and effort into building this network from scratch.

I believe so deeply in the importance of having a financially sustainable realtime feed API & service that I am going to refocus App.net to become exactly that. I have the experience, vision, infrastructure and team to do it. Additionally, we already have much of this built: a polished native iOS app, a robust technical infrastructure currently capable of handing ~200MM API calls per day with no code changes, and a developer-facing API provisioning, documentation and analytics system. This isn’t vaporware.

Dalton Caldwell

The blog post is long, but Caldwell ends it with a call to action. He needed 10,000 backers totaling half a million dollars in order to launch the service. He felt this metric was a “minimum critical mass” wherein the service could have a chance of success. The project far exceeded his goal as only a month later he raised approximately $750,000, with over 11,000 backers.

A Different Approach

App.net (ADN as it would be abbreviated) focused on core values from the beginning. I already mentioned the lack of advertising, which went hand-in-hand with being a paid service. Other ideas that would guide App.net would be a revenue sharing program with developers. Because App.net was taking in money from users, Caldwell felt it was fair to incentivize great apps by sharing a piece of that revenue. The concept made sense. ADN wanted great apps to hook into its service. They wanted to pay those developers. Even better was the app could be free to users but the developer still make money because of that revenue sharing. It was a win-win-win.

Caldwell was around Twitter in the early days and saw the many examples of Twitter shutting out developers. Frankly, they became hostile toward developers. He felt this put Twitter behind the 8-ball when it came to courting developers to build great Twitter apps.

3rd-party developers added all of the value to the platform in the early days, and instead of being shut down they should get paid. The reason the Twitter developer ecosystem has failed is that Twitter itself has a flawed business model.

Dalton Caldwell

He isn’t wrong here. Nearly every feature of Twitter was invented not by Twitter themselves but by app developers. Retweets, mentions, pull to refresh, hashtags, the list goes on and on. Twitter eventually made all these native to the platform, but no developer ever saw a dime for their contributions from Twitter themselves.

App.net Has Liftoff

In August of 2012 ADN launched the alpha version of the service. By the time October hit the revenue sharing idea had come to fruition with a $20,000 pool of money up for grabs to participating developers. ADN would pay developers based on app usage and customer feedback.

The company TapBots took their existing Tweetbot app and repurposed it into a new version called Netbot. If anyone was a big Twitter user at the time, Tweetbot was one of the most popular apps. Netbot looked nearly-identical, but was for ADN. Netbot received a glowing review on MacStories. This set the stage for Netbot to be the app to beat for everyone making ADN apps.

For me, App.net is, right now, a Good Guy Twitter. On the outside, the service is essentially the same, but its creators have bigger (and different) plans for its API and third-party support. These new APIs still have to reveal their full potential, so, like I said, for now I’m using App.net as “another Twitter”. I think we all are.

As more people keep switching from Twitter to ADN and the service gets more exclusive functionality, Caldwell’s vision will mature into a more “complete” and unique product.

Federico Viticci on MacStories

The quick hits kept coming as ADN opened a free trial membership at the end of November. Another app, Riposte, landed on the various app stores in January. While it looked similar to Netbot, it also had enough of a unique look to make it a popular application to put on one’s phone. The App.net service itself passed 100,000 users later that year in May.

Sophomore Jinx

There is a phrase in sports called the “sophomore jinx”. The gist is a rookie player does fantastic in their season but has a slump in their second season. We’ll consider 2013 the first year of App.net since it launched so late in 2012. Given that, its second year in 2014 gave way to the rocky waters of running a paid social network.

In the end of January they launched “Backer” wherein ADN themselves or developers could crowdfund additional features. ADN presented it as a way to gain insight into what people wanted, but asking for more money was the first tell something was amiss. The entire point of ADN was to pay for a full service. By asking a customer to pay more for developing features, there was an inherent negativity with that idea.

It is difficult to find out what happened to many of these projects mainly because they were posted to ADN. As you can probably guess, the site and everything is long gone from the internet and so it is near-impossible to get enough information to paint the full picture. However I do know that when it was time for customers to renew their subscriptions, ADN fell short of their needs. Caldwell was upfront about this on the ADN blog.

There were people who did renew their memberships and it was enough to cover the operational costs. However they did not have enough money to keep on any full-time employees. Caldwell and his co-founder would no longer draw any pay from ADN either. In order to keep improving things, he open-sourced parts of ADN’s code and asked for community contributors. The service was officially running in “Maintenance Mode”. By this time the writing was already on the wall for ADN.

Be Happy It Happened

My research into this stage of ADN’s life has turned up nothing. Unfortunately when a company is treading water and fighting off users abandoning the platform for Facebook & Twitter, you don’t normally hear much noise. Posts on ADN itself (long gone) were likely the main place to hear about this. The May 6th blog post about going into Maintenance Mode was the last anyone would hear until the calendar turned to 2017.

Three years of silence finally broke. In January of 2017, there was another (and last) post to the ADN blog. The three paragraph entry was succinct: App.net was shutting down. They saw a constant decline in revenue for the company and there was no way to resume building it. Caldwell likened it to the chicken-and-egg problem. You need users on a platform in order to draw in developers to build great apps. You need developers building great apps to draw in users. The initial efforts of Caldwell to get developers building for ADN worked. The aforementioned Riposte and Netbot apps were stellar. Even Buffer was part of it from day one.

However, the users never came at a sustainable pace. Yes ADN hit the 100k user mark, but a closed-off service needs an “it” thing to constantly draw in users. Asking people to pay for what amounted to a cool Twitter clone was not resonating.

Caldwell announced an immediate halting of new signups and renewing subscriptions. The service would remain online for two months to run out the clock. All code for ADN would become open source and everyone had this lead time to export all their data off the platform.

And so it went. ADN made good on their promises. The service went offline on March 14, 2017. ADN erased all user data at that time. They had baked in ability to export everything out. It was simple to do prior to the deletion. The GitHub page for ADN’s code remains online and fully viewable to this day. The only thing lost are the connections and the hopes they were able to continue those relationships on another platform. The Internet Archive has some of the final posts saved if you want to see what the remaining community was posting at the end.

Room For Two

App.net was a fantastic idea looking to forge its own path. Dalton Caldwell saw the inherent flaws with Twitter and how it conducted business. But even the biggest companies encountered problems trying to muscle their way onto the scene. Google’s Google+ social network had 200 million users and Google-level of money backing it. They still failed to make it stick. In comparison, it’s impressive that ADN was able to get anything to market and make a two year run at keeping the lights on.

In the end, we are still left with the same social networks we started with: Twitter and Facebook Meta. Both have their plusses and minuses. Both have gone in very different directions on how they function and both have fended off competitors. Meta has been in the business of acquiring apps (Instagram & WhatsApp) or straight-up ripping others off (Snapchat, Craigslist). Twitter has not solved its hate problem but keeps implementing features to limit replies to tweets, it recommitted to developers, and launched a paid version of the service. Neither of these networks will ever sick the landing. They can’t due to the sheer size and diversity of their users. Lucky for them they are big enough where they don’t need to.

App.net was a flash in the pan. However it was driven by a single focus that is rare for a startup to embody. One where users are customers, advertisers are not welcome, and where everyone can win if goals are aligned. ADN had a spark of that rarity and it’s one that should be commended.

This has been another article for curious readers like you.

TimeMachiner helps you find that next great article, like this one, by sharing the best in tech, culture, and nostalgia.

Subscribe below. It fits great in your inbox.