I'm befuddled that Automattic is still investing in adding features (e.g. federation) to Tumblr. There's just about no chance of them actually making money from it, because its users are just about the most monetization-hostile users on the Internet.
It doesn't help that the CEO actively antagonized a large portion of the site not too long ago.
Their play is not to make money from Tumblr users, but to grow the total number of users connected to ActivityPub. Then they can go to BigCos and say "here it is, you can drop Twitter/Threads/Instagram/TikTok and control your social media presence directly."
It's not that different from what Facebook is doing with Threads. They are not interested in making money from social media, they are just hoping they can become the main infrastructure provider.
basically bring your own domain like Google Workspace or O365. I'd love to see the Mastodon project (or any other AP implementation) adopt that managed services model.
I've been trying that with Communick for over 5 years already, and (thanks for Trump and VP Elon) this was the first month where my revenue (barely) passed my operating costs.
Companies/agencies/media institutions are not interested in being trailblazers, so they will just go where they see their audience going, meaning Bluesky nowadays. On the other side, the absolute majority of end users still believe that (free) social media is not something worth paying for, and the most you'll see is people that contribute a few dollars per year "to cover hardware costs".
It’s absurdly expensive to offer, because none of the major implementations support serving multiple domains in a single process, so you end up with a lot of duplication.
I wonder if Meta's Threads is being built with that capability. I would expect an incumbent player to make an entrance into this market and they seem like the furthest along in capacity and capabilities.
I mean specifically, I'd like to see the actual Mastodon organization (Eugen Rochko's team now that they're a legit nonprofit) take that role. Or any other fedi software project tbh, doesn't need to be Mastodon themselves.
Very few social platforms have figured out that the right way to get money is from the readers supporting the posters, not from the posters themselves.
Reddit gold is a pale version of this. Neither Twitter nor Bluesky show any signs of getting it.
Hard to say that isn't the "right way" when they are making money with it.
If we're talking hypothetical right ways, I think what "social" media gets wrong is to envision an anarchy for its users. People are never equal in a society. A teenager isn't the same thing as an elder, nor is a doctor the same thing as a nurse.
I think I've read before that anarchy doesn't work because eventually people find someone who is reliable and naturally that person becomes their authority on the subject, creating an hierarchy.
Social media tries to put all users in the same bucket even when it's well-known that most users belong in the "never posts anything" bucket while very few users belong in the "power user" bucket. There are passive users and active users.
While hierarchies can become ugly fast, I don't think resigning on implementing them in any form is the right answer, specially now that everyone is on the Internet.
> I'm befuddled that Automattic is still investing in adding features (e.g. federation) to Tumblr.
The article is saying the opposite though: they are not adding federation directly to Tumblr's own codebase. Instead, federation will come about from some planned future migration of Tumblr onto WordPress.com infrastructure, which already has that feature.
That doesn't make sense. The primary use-case (and traffic driver) for Tumblr is a social network for logged-in users, similar to Twitter but with some differences in the feature set.
The public blog network is a minority of traffic. If you've never actually been a Tumblr user, you won't have an understanding of the product at all.
If Matt wasn't so fixated on WPEngine, he could actually do something important and radical by pushing Tumblr as a true user-controlled (i.e. each user has their own TLD) alternative to Twitter, BlueSky, Instagram, Threads, etc.
It gives control to the node operator. You join a node on the basis that you like the operator and trust them to federate responsibly. The end user is along for the ride with the node operator.
That's not how it works in the case of WordPress though, in this case it's not posting to some account on someone else's instance, each blog is it's own ActivityPub instance (as they just add support for the ActivityPub endpoints to the existing hosting)
Mastodon suffers from the same problem Reddit does. Moderators have too much power.
A truly P2P system where the end user has 100% control over what is blocked, and furthermore where they can't be shut out in the cold by capricious mods, would be the ideal social media vehicle.
You can already control who and what you see by running your own server. If you choose to use someone else's infrastructure, don't act surprised when you're subject to their rules.
You sound much better informed on the technical/plumbing details than I. Mainly I was trying to emphasize that the lock-in and high-switching cost from social networks comes from centralized control of the urls. If each content creator controlled their own domain, the discovery mechanism/feed could be separate.
"running Tumblr’s back end on WordPress would allow for greater efficiencies, while not changing the interface and experience that Tumblr’s user base has grown to love"
For people who don’t know the backstory, back in the wild west days of the internet, Tumblr’s official content policy (and I’m not making this up) was “Go nuts! Show nuts. Whatever.”
Threads was preemptively shitlisted just for thinking of connecting to Fediverse, because a lot of instance admins are ActivityPub or Mastodon maximalists[0] and believed that everyone would just flock over to Threads immediately.
Then it was shitlisted again because Mark Zuckerberg started zucking on Trump, and decided to take their already threadbare moderation team and tie their hands further. This is a more practical concern as most Fedi admins do not have the time or ability to deal with one mega-instance that happens to be both indispensable and willing to flood everyone with garbage.
I could see Tumblr getting shitlisted purely on the grounds of "fuck Matt Mullenweg", because Fedi admins are also hilariously petty[1]. But in practice, "just link Tumblr up to Fedi" is going to be just as much of a problem as "just link Threads up to Fedi".
There's an underlying tension between the Fediverse's technical underpinnings - ActivityPub - and the community of Fediverse servers that use it. The technology wants to be widely adopted, but the community wants to be small enough to avoid harassment and attacks[2], and these are in conflict. The Fediverse's structure is already a lot more centralized than anyone would like to admit, and scaling the network makes this problem worse.
[0] As opposed to general enthusiasm for federated communications technologies. See also the pushback Cory Doctorow got for backing Free our Feeds, a plan to make good on BlueSky's currently vacuous federation promises.
[1] This is why I self-host Mastodon even though it's a total pain in the ass.
[2] Like that one time a bunch of Japanese skiddies decided to spam literally every server from whatever accounts they could register
There is no tension. ActivityPub and its creator are well-aligned with the community. What you you are looking for is nostr, and if you find it more appealing you are free to choose it, that’s the beauty of open source.
Regarding [1], do you foresee this changing at some point? I'm interested in self hosting, bit have heard consistently that it is painful, so I've been waiting in the hopes that it will eventually improve.
Hosting Mastodon is as difficult as deploying a traditional web app: if you are familiar with docker, it's just a matter of choosing where you are going to put your database, redis and where to store media.
The problem is that it is expensive. Even if you host on a cheap VPS and put your media on some object storage like Storj (~$4/TB/month), you are problably looking at a minimum of $20/month for the server. If you get for yourself, maybe your friends and stay under 50 users, fine.
If you get more than that then you'll need a beefier server, and if any of users follows lots of media-heavy accounts and does not set it up to delete old posts, your object storage will be full very quick.
Pleroma was recommended to me as a less resource-intensive alternative to Mastodon and with less anti-features. It runs fine with one user on Hetzner's cheapest VPS.
I delete posts mirrored from other servers older than 90 days with a cron job. It has a command to do this.
Pleroma and GoToSocial are indeed better if you are looking for something more efficient.
But I am still not happy with the idea that we should be deleting data from the server just to keep it manageable. It reminds of the before Gmail where people would get 5MB quota of Yahoo Mail and would have to go around deleting stuff from their mailboxes every other month.
There are extreme ideological distinctions between Facebook and Tumblr that you're ignoring. Suggesting ideology only came into it with Trump and that the Threads pushback was about the spectre of a competitor's popularity is very strange. As is the "Fedi admins are petty, but not me, a Fedi admin" commentary. People had well-founded and widely varied concerns about Meta joining the fediverse. It had nothing to do with wanting to remain in obscurity; just about every fedi user I've ever encountered is absolutely begging everyone on social media to join.
> The Fediverse's structure is already a lot more centralized than anyone would like to admit
Hi, I'm "anyone", and: no it isn't? Technically and philosophically, no it isn't. Large instances are not a failure of decentralisation, because as you evidently know, their existence does not preclude you or anyone else setting up a server and federating. There's no universe in which "scaling the network" reduces decentralisation, in fact it's the solution to the "fuck Matt Mullenweg" situation you speculate about. The less homogenous fedi admins become, the more the network is sustained by people who don't even know who he is. Your comment is rife with generalisations about the motivations of people on the fediverse, but it's far less monolithic than you think.
I'd gamble that it'll be less so than the response to Threads. Tumblr users, on average, are generally a better culture fit than Facebook/Instagram ones.
It doesn't help that the CEO actively antagonized a large portion of the site not too long ago.