In the current ongoing debate of federation with Meta Threads

whiskers@lemmy.world to Fediverse@lemmy.world – 22 points –

Everyone is talking about how Meta is trying to Embrace, Extend, and Extinguish the Fediverse. Meta won't be alone for long in this goal, there will be a lot of capitalist actors that would try to do the same in the long run.

Defederation with them will be a shot in the leg, and handicap the Fediverse movement itself. There will be users/instances in the current Fediverse that would want to federate with them, and banning such instances would create silos and echo chambers.

The way out of this is to focus on the 2nd E - "Extend". I think we can all agree that UX of Threads app will always probably be the best out of all the federated instances. But that is something that people can still live without. Before long, Meta will tout shortcomings like lack of E2E encryption in the private messages and some other core features, that will create a bigger divide amongst ourselves. The Fediverse developers and community have to keep abreast of Meta on such core features, so that they can never extend the core of the Fediverse.

Let me know of your thoughts!

27

A for-profit and a non-profit silos existing along each other might suffice. The resources and goals of the for-profit one shouldn't dictate what the non-profit community of developers, admins and users, a lot of which are volunteers do. We can't match either and we probably shouldn't.

It seems likely that some instances will federate and others will not. I think eventually, many people who hate Meta might move to the instances that federated with Meta, just so that they can find people on their social graph.

I'm afraid, that's when silo of the defederated instances will become echo chambers, especially if they do second order blocking as some have threatened to do with mastodon.social

I'm not sure I totally agree with that assessment. If we compare the sheer amount of resources and talent behind Meta vs. behind the Fediverse, it's clear that there's no way the Fediverse can ever compete in terms on innovation and features. Not to say that there aren't talented devs on the Fediverse - they're all pretty talented if you ask me - but Meta can just hire hundreds of similarly talented devs without even feeling the cost.

Really, the only way to prevent getting washed away by Meta (in development, in users, in culture, in basically just about everything) is to not associate with Meta at all. Co-existence doesn't mean we have to cooperate, after all. If the Fediverse shows that it has the userbase and staying power to withstand a full EEE "assualt" and remain standing, then I would be more inclined to take the "wait and see" approach. But I think as it stands now, Meta is very clearly trying to overwhelm the Fediverse before it has a chance to get going on its own merits

I honestly don't think Meta even cares about the Fediverse as we know it. Mastodon barely cracked 2.5m active monthly users in late 2022/early 2023, and the Fediverse user counts had been dropping until the Reddit debacle. There's no way they are agile enough to push this out the door in response to what happened in June, and this is a lot of effort to kill something already in decline with only 0.01% your active monthly user base.

Even so, what do they gain by killing Mastodon? Paying for the dev work and the server cost to host their own instance? All the data on the Fediverse that they'd want to sell is publicly accessible anyway; they could grab/sell it all without going through this effort. I think they see long term value in the ActivityPub protocol, and they are positioning themselves to get ahead of any other major corporate ActivityPub-based platform.

I suspect the intention is that if another major platform appears (BlueSky, anyone?) that it will pressure them to also implement ActivityPub, thereby increasing the pool of data on the Fediverse and letting Meta analyze and sell it too.

Letting them dictate the pace for technological development is actually the shortest path to be extinguished.

They have already 30 millions of users, which is approximately 2.5x the whole fediverse. Shortly they will easily reach 100/200 millions, probably, which means the whole fediverse will be <5%.

Now, in this condition, with Meta turning >100 billions of profit in a year, Mastodon (and Lemmy, and Pixelfed) etc., should compete by aiming for feature parity with an organization that can throw hundreds of full-time developers at the code? Sorry, no.

The whole idea in my opinion is framed poorly. For me the fediverse is a technical implementation of an idea. The technology comes after the idea, and the idea is simple: decentralization, non-monetization, no ads, and no-profit. It is a corner of the cyberspace which is and should remain out of reach for the big companies. We cannot, and should not, compete in their game.

This means that our tech should be poor out of principle? No, obviously. But we need to be realistic that fedi software will fail to keep the pace in terms of features with Threads. Aiming to do that seems already saying that Threads will take decisions, the rest will need to catch up, and it's just a matter of time before one of their feature is a change in ActivityPub, or requires an extension of it, or breaks compliance with it.

I think that the way forward is simply acknowledging that while there are technical similarities, Threads and the fedi software are wildly different things, and they should be considered as such. Some will federate, some will not, but we should keep that distinction.

Yes, I agree that we can't let them dictate the pace. Let me try to express my thoughts more clearly. Couldn't we have something like The Linux Foundation for Fediverse? It has many corporate members including Google, Microsoft.

Some corporations can be partners to a ActivityPub foundation and contribute to the codebase of the protocol. The foundation itself needs to be independent to steer the features and technical direction according to the Fediverse principles.

As you said just because this is mostly run by volunteers, it need not be an inferior product. But some thought on what I said above might make it something that is adopted by the general populace as well.

I think there is a fundamental difference between a tool like the kernel and a protocol which is then implemented by others. Google is part of those who standardizes the web, and it killed any browser competition exactly because it pushed so much stuff, that if you start a browser from 0 today, you will need millions and years to work with most websites.

The Linux kernel instead is one, centralized and that gets distributed, and Linus and other maintainers are gatekeepers as well.

I honestly think there is simply no way to avoid a complete takeover when there is this much asymmetry. Or well, the way is to keep things separated, maybe.

What need is there to Federate with Meta? They're a corporation where growth matters. Lemmy and Mastodon are private, networked instances where funding can and often does come from donations. Provided instance operators can subsist off those donations, the need to grow for the sake of growth/profit is a non-issue.

I don’t think it’s inevitable Threads will have a better user experience than non-corporate apps. Meta has different goals and will optimize the user experience to drive profits. Look at how they continuously make Facebook and Instagram worse to drive engagement.

Also, see Reddit. Apollo was one dude’s project that blows the official app out of the water. And all the other apps scrambling to support Lemmy benefit from apps like Apollo and Reddit is Fun lessons in UX. It’s early days but we’re already seeing Lemmy apps with a lot of potential.

Linux is another example. Technical issues for new users aside, there’s a lot of beautiful, easy to use desktops that blow windows 10/11 out of the water. And other less new user friendly options for power users are available as well.

The community driven approach has been demonstrated to work because it’s only goal is to give the community a good experience.

I think a promising way to defeat a giant for-profit corporation like Meta is by investing capital in many smaller non-profit corporations which can each focus more tightly on the fediverse while Meta spins off to focus on the next hype train a year from now.

Little non-profit companies can carve off small portions of the market for federated social media content and, if enough exist to carve it in small enough pieces, then giant corporations like Meta will focus on other markets that are less fragmented.

Sure, Meta might try to buy out the little guys and consolidate, but as long as the little guys keep investing in free and open source IP(not trademarked/copyrighted systems) there won't be much room for Meta to technically differentiate themselves in a meaningful way to most consumers.

Defederation with them will be a shot in the leg, and handicap the Fediverse movement itself. There will be users/instances in the current Fediverse that would want to federate with them, and banning such instances would create silos and echo chambers.

I don't quite follow this line of thinking. Instance A can still be federated with Instance B even if Instance B is federated with Threads.

The greater danger, IMO, is the open source Fediverse losing users to Thread because of the slicker interface and greater engagement.