Okay, so Cory Doctorow published his take on the #freeourfeeds thingy, all about "billionaire-proofing" the Internet (or more accurately, billionaire-proofing social media), and I have some thoughts...
pluralistic.net/2025/01/14/con…
His main argument is this --> yeah, Bluesky isn't perfect, but it's what's fun and popular right now, and if we can billionaire-proof it, then we should, rather than getting people to adopt the existing, already proven, billionaire-proofed Fediverse ecosystem.
Okay, that's fair I think. I have no objection to this on its face. The "Fediverse" has been around for a while and though it has never been more popular than it is now, it has struggled to really gain mainstream appeal. (Perhaps that will change more drastically this year as fedi platforms like #Pixelfed gain more attention though). So though he admits that Mastodon/Fediverse is ideal, would be perfect for what they're trying to do ("Free our Feeds"), it's just not the clearest solution because the difficulty of getting people to adopt it en-masse seems somewhat insurmountable at this time.
Fine.
He goes on to say that the answer is, as he puts it, "installing our own fire-exits in Bluesky", or in other words, standing up a second Bluesky relay/infrastructure that "federates" or otherwise interoperates with the main Bluesky infrastructure - similar to how distinct Fediverse instances can communicate with each other. This too makes sense. If the main Bluesky enshittifies, or goes down, or is subject to hostile/billionaire takeover, than people have the choice to switch over to the other relay, the "good" one. Now here is where his argument starts to fall apart...
First, "FreeOurFeeds" is asking for 30 million dollars to do this. Wow. That's a lot, and doesn't speak well to how easy it is to stand up and maintain a second instance of the entire bluesky infra. Second, the FreeOurFeeds website speaks nothing of how they will keep that infra alive, i.e. pay for ongoing operating costs. Even the main Bluesky company doesn't have a realistic monetization strategy, they are kept alive by VC money alone pretty much.
Finally, what Cory fails to address, and in my mind, the biggest issue with his whole argument is that having just ONE single other relay/instance of Bsky as a whole is not the robust "fire exit" he thinks it is. Let's say FreeOurFeeds succeeds and they stand up a fully interoperable, yet fully isolated instance of Bluesky, and they manage to get a sizable portion of the existing Bluesky userbase to move over there. Like, 5 million of Blueskys ~30 mil or w/e. Let's say then that Bluesky sells out to Elon Musk or Bezos or whomever. This triggers another mass migration of 5 million, or 10 million or even everyone over to the FreeOurFeeds relay. (Though we all know that they'd be lucky to get half of regular people to figure out how to "Switch" relay providers). Now they are RIGHT BACK to where they were to begin with. A single provider, just as vulnerable to hostile takeover or enshittification as the original Bluesky was.
So you might say, okay? well then stand up ANOTHER relay. Crowdfund another 30 million. Right? ... What?!? Bags of 30 million dollars don't just grow on trees ya know? It's hard enough to get people to pay for these types of services, and I just don't think having one, or two or even 5 other relays is enough redundancy, enough resiliency to combat the billionaires and the governments of the world that want to control what we see, what we talk about, and who's allowed to speak.
That's why the Fediverse works. It's made up of literally thousands of servers, of all different sizes, some Mastodon, some GoToSocial, some Lemmy, some I mean w/e. Theres tons of different platforms that all interoperate, and they don't cost 30 million to run. I'm typing this out on a GoToSocial instance that literally costs $3.75/mo. You heard that right. Less than $4/mo. To put that in perspective, I could run 8 million GoToSocial instances at this price if I was given 30 million dollars. I'd like to see Musk try to buy all those out 😄.
I'll try to put it another way here. I'm in cybersecurity, and one thing we're taught is that attackers, with enough time and resources can literally get into anything. So our jobs as defenders, on top of all the usual stuff (e.g. defense in depth, blah blah), is to just slow them down. Make it impractical (from a time-spent perspective) to achieve a reasonable return on (time) investment. It wouldn't be hard for a billionaire or a hostile government to take down one, two, three or maybe even 10 Bluesky relays. But it seems wholly impractical for a billionaire, or a team of billionaires to snuff out thousands of Fediverse instances. And hell, these are just instances that exist now, in a very fledgling model where Fedi only has like a couple million MAU. Imagine a timeline where Fediverse got REAL world-wide adoption, and we went from a few thousand instances, to hundreds of thousands, or millions of instances. Taking that down would be like taking down the Internet itself, and if that could happen, we might have bigger issues.
Oh and I forgot to mention, that FreeOurFeeds can't even stand up another relay. Bsky devs have already stated that's impossible at the moment (shellsharks.com/notes/2024/11/…). Though theoretically this will be addressed in time...
Alrighty, let's wrap this up. I get it. Bluesky is hot. It's popular. It's got steam. It's got gobs of cursed VC money. It's got some admittedly cool tech behind it. But it just isn't it. Not unless it can find a way to solve for the whole it's-insanely-expensive-to-run-another-relay issue. If FreeOurFeeds succeeds, stands up another instance and figures out a way to run it on the cheap, then I'll be the first to say "congrats" and "I was wrong". Until then, I will remain skeptical, and I'll keep saying that this sort of money would be better thrown at building out the proven solution, the Fediverse.
