The quirks and oddities of a social network affect the community that grows up around it. Instagram’s lack of a repost feature pushed users to rely on hashtags to spread their pictures across the network, sowing the seeds of the heavily interest-based communities that still live there today. The anonymity offered by 4chan lead, perversely, to a uniformity of tone, as users conform to the zeitgeist of the site, unable to build a name for themselves as an individual. TikTok – built by people who knew what they were doing – carefully sculpted its quirks to nudge users in its preferred direction, boosting harmless dance trends and discouraging political rants of the sort that litter competitor YouTube.
Over the years, some of the largest social networks have filed those quirks off, pushing for a homogeneity that is more accessible to all, even at the expense of what makes them unique. Twitter’s strict text-only, reverse-chronological, 140-character timeline is now algorithmically curated, offering 280 characters plus a range of multimedia; Instagram posts can be reshared in Snapchat-style stories, which can also contain TikTok-style videos; TikTok pivoted towards political content and now plays a leading role in the culture wars.
Tumblr, however, has stayed weird. The social network, which launched in 2007 – a few months after Twitter – has always had its quirks. Rather than a content model built around posts, images or videos, for instance, Tumblr’s key unit is a thread: a post, its reblogs, their tags and all the content that rides with them. It leads to a freewheeling community, where the best posts are enhanced, elaborated and riffed on even as they are shared further.
And users can’t simply reply to a post – you either reblog it (sharing it and your comment with friends), send a private message about it, or you send a semi-public “ask”, a reply that the recipient can then, if they want, publish, creating a whole new chain of conversation. Even the commercial model is odd: as well as targeted adverts, the site sells an “ad-free” experience (for $40 a year), offers a Patreon-style subscription services (called Post+), and recently launched Tumblr Blaze, a bizarre promoted-post service that lets users spend $10 to show their content to a completely untargeted selection of 2,500 users, and has been gleefully adopted by the community for a sort of esoteric trolling.
Free-loving fandom
The rough edges have stuck around for so long, in part, because Tumblr has spent years as the barely tolerated subsidiary of some of the world’s largest companies. In 2013, it was acquired by Yahoo for $1.1bn, then promptly ignored. The company’s CEO remained unchanged, “advertising goals weren’t met”, Yahoo wrote down more than $700m of the acquisition, and then the whole company was acquired by Verizon in 2017. Only two years later it was spun off and sold to Automattic, the makers of the popular WordPress platform. But in its years at Verizon, one change stuck: the “porn ban”.
From its early days, Tumblr had an unusually open attitude to porn. As well as allowing explicit blogs, a significant chunk of the service’s community enjoyed the freedom of floating between fandom, erotica, pornography and conversation – until, in 2018, Verizon implemented a content policy with strict restrictions on adult content (including an infamous ban on “female-presenting nipples”).
The ban caused a mild exodus from the site, as users skipped to other services with looser policies – primarily Twitter, which still allows adult content with few restrictions. But Tumblr stuck around and – since the Automattic purchase – users have been pushing for the policy to be reversed.
The problem with credit
Which takes us to today, and the news that, well, that’s not going to happen. Matt Mullenweg, Automattic’s co-founder, posted on Tumblr explaining why the old policy – “Go nuts, show nuts. Whatever” – couldn’t work in 2022. It’s about more than porn, really, and functions as a concise explanation of what it’s like being a mid-sized social network in the shadow of the major platforms.
Mullenweg writes:
“No modern internet service in 2022 can have the rules that Tumblr did in 2007. I am personally extremely libertarian in terms of what consenting adults should be able to share, and I agree with ‘go nuts, show nuts’ in principle, but the casually porn-friendly era of the early internet is currently impossible.”
Mullenweg breaks the issues out into four areas: credit card processing, App Store policies, verification of consent and age, and service provider limitations.
Of those, three are sort of the same thing. If you work in porn, a lot of other companies don’t want to work with you. “Credit card companies are anti-porn,” he writes. “Whatever crypto-utopia might come in the coming decades, today if you are blocked from banks, credit card processing, and financial services, you’re blocked from the modern economy.
“In addition to a company primarily serving adult content not having access to normal financial services and being blocked by app stores, they also need specialised service providers – for example, for their bandwidth and network connections.”
Even if you take payments without involving Mastercard, and switch the entire business to a porn-friendly set of service providers, you still have to deal with the twin gatekeepers of the mobile ecosystem, Google and Apple. “If Apple permanently banned Tumblr from the App Store, we’d probably have to shut the service down,” wrote Mullenweg (pictured above). “If you want apps to allow more adult content, please lobby Apple.”
But those three reasons are also, in their own ways, reflections of the last: hosting user-generated porn is increasingly seen as an unacceptable risk. Mullenweg phrases the problem as one of “lots of new rules around verifying consent and age in adult content”:
Non-consensual sharing has grown exponentially and has been a huge problem on dedicated porn sites like Pornhub – and governments have rightly been expanding laws and regulations to make sure everyone being shown in online adult content is of legal age and has consented to the material being shared.
That problem gets reflected up and down the stack. Even if you are convinced your restrictions on child abuse imagery are sufficient, you also have to convince your web host, and your payment processor, and likely regulators around the world.
“If you wanted to start an adult social network in 2022,” Mullenweg says, “you’d need to be web-only on iOS and sideload on Android, take payment in crypto, have a way to convert crypto to fiat for business operations without being blocked, do a ton of work in age and identity verification and compliance so you don’t go to jail, protect all of that identity information so you don’t dox your users, and make a ton of money.”
One rule for Tumblr, another for Twitter?
One big question is how, if Tumblr can’t make adult content work, Reddit and Twitter get away with it. Mullenweg’s best guess could be paraphrased as ¯\(ツ)/¯:
Ask Apple, because I don’t know. My guess is that Twitter and Reddit are too big for Apple to block so they decided to make an example out of Tumblr, which has “only” 102 million monthly visitors. Maybe Twitter gets blocked by Apple sometimes too but can’t talk about it because they’re a public company and it would scare investors.
There are two stories here, really. One is the broader one of gatekeepers. Apple’s anti-porn stance isn’t about child-abuse imagery: you simply cannot publish porn to the App Store at all. But the App Store isn’t just a store, it’s also the only way of installing software on the most popular smartphone in the world, and the company’s decisions about what businesses it wants to support have a way of warping the entire culture.
The other story is of society looking at a formerly ungovernable space, and deciding that it should follow the same rules as the rest of the world: that even unknowingly hosting child abuse imagery or nonconsensual pornography, for instance, should be seen as a very, very bad thing, and not a simple cost of doing business. Allowing user uploads of porn while policing what, precisely, is uploaded, is really hard – and something that a lot of companies simply don’t want to bother with.
If you want to read the complete version of the newsletter please subscribe to receive TechScape in your inbox every Wednesday.
