This is a response to Manton Reece’s “Bluesky Downtime”. No hate to the guy, he just happens to be wrong and misinformed. Due to it’s nature, this post will have a lot of direct quotes from his article. I still recommend reading his first, and then coming back to this post.

Reece starts his post by posing the question “What would happen if mastodon.social went down?”. It’s a pretty significant instance, apparently hosting 15% of the entire fediverse (and probably more if you only include microblogging instances).

Even so, if a single very large Mastodon server went down, no one would use it to question whether Mastodon is decentralized, other than as a reminder that smaller communities are often better and make the network more resilient. Likewise, it’s not really fair to frame bsky.social as merely a large server; it’s effectively the only thing right now, which isn’t true for mastodon.social.

Are smaller communities really better? I don’t like how this is presented as a given here. They’re fine if you want a more insular experience, but I don’t think that’s what the majority of people are on social media for.

Bluesky specifically got hit with a DDoS attack. If the top 5 mastodon instances got hit with the same attack it would have roughly the same effect.

A running theme throughout his post is a misunderstanding of atproto’s architecture. This is understandable, it’s certainly a lot different than other models and I’m not even 100% familiar with it. But here he speaks on it so authoritatively. If bsky.social disappeared tommorrow (which, by the way, it won’t. They’re well funded enough to continue building the protocol to maturity), the community would take over their efforts and have something runnning before the week was over. And, as a bonus, everyone gets to keep their data. If mastodon.social disappears, a whole 15% of the network has to relocate, and start rebuilding their follower base, without their data or posts to build off of.

When you call the Bluesky API via bsky.social, it actually proxies your requests to the appropriate backend data server if it’s hosted on Bluesky. This makes the API feel very centralized, but all of the data could still be accessed directly in a more distributed way, like accessing individual websites.

I’m having a hard time understanding what this means. Yes, you can proxy your requests through bsky.social. The API “feels centralized”, but he’s saying it isn’t? This is purely vibes based, you can easily directly query the PDS if you want to. There’s a pretty cool example at pds.witchcraft.systems

Bluesky is not federated like Mastodon, but the open architecture that Bluesky was designed around is valuable. It is distributed sort of like the web is distributed even though Google exists and dominates search. Eventually, there will be other AppViews that work like Bluesky but run in parallel with their own timeline. In that case, if Bluesky went down, some other apps might still function, and could utilize the same data and social graph.

It is absurdly difficult to divest from Google. If you use all their services, you have to change your email address on every account you have across the internet, you need to move your documents, your videos, you need to convince everyone else to move their videos, and you have to put up with a worse experience because without a Google account you can’t interact with Google services. This is fundamentally not how atproto works. If you need to divest from Bluesky, all you do is download your repo and move to a new PDS, with all your data intact. Bluesky turns evil, or bsky.app dies? Just use deer.social, it has all the Bluesky posts because the protocol is decentralized. This is a fundamentally uninformed criticism.