Why do People Choose BlueSky Over Mastodon?

prototype_g2@lemmy.ml to Asklemmy@lemmy.ml – 231 points –

I's heard news that BlueSky has been growing a lot as Xitter becomes worse and worse, but why do people seem to prefer BlueSky? This confuses me because BlueSky does not have any federalization technologies built into it, meaning it's just another centralized platform, and thus vulnerable to the same things that make modern social media so horrible.

And so, in the hopes of having a better understanding, I've come here to ask what problems Mastodon has that keep people from migrating to it and what is BlueSky doing so right that it attracts so many people.

This question is directed to those who have used all three platforms, although others are free to put out their own thoughts.

(To be clear, I've never used Xitter, BlueSky or Mastodon. I'm asking specifically so that I don't have to make an account on each to find out by myself.)


Edit:

Edit2: (changed the wording a bit on the last part of point 1 to make my point clearer.)

::: spoiler From reading the comments, here are what seems to be the main reasons:

  1. Federation is hard

The concept of federation seems to be harder to grasp than tech people expected. As one user pointed out, tech literacy is much less prevalent than tech folk might expect.

On Mastodon, you must pick an instance, for some weird "federation" tech reason, whatever that means; and thanks to that "federation" there are some post you cannot see (due to defederalization). To someone who barely understands what a server is, the complex network of federalization is to much to bare.

BlueSky, on the other hand, is simple: just go to this website, creating an account and Ta Da! Done! No need to understand anything else.

The federalized nature of Mastodon seems to be its biggest flaw.

The unfamiliar and more complex nature of Mastodon's federalization technology seems to be its biggest obstacle towards achieving mass adoption.

  1. No Algorithm

Mastodon has no algorithm to surface relevant posts, it is just a chronological timeline. Although some prefer this, others don't and would rather have an algorithm serving them good quality post instead of spending 10h+ curating a subscription feed.

  1. UI and UX

People say that Mastodon (and Lemmy) have HORRIBLE UX, which will surely drive many away from Mastodon. Also, some pointed out that BlueSky's overall design more closely follows that of Twitter, so BlueSky quite literally looks more like pre-Musk Xitter. :::

226

You are viewing a single comment

Perhaps... But how exactly?

i wish i had that answer

its usually how corpos and ux people seem solve these issues

Initial log in in the apps should default to mastodon.social with other servers buried under a menu

Defeats the whole purpose tbh. Federation means decentralisation, single point of failure architecture in that is asking for trouble.

Techies who are comfortable with federation can use the menu, no? The vast, vast majority of people don't and I do believe things should be as frictionless for them as possible. Even a big fediverse server is better than yet another walled garden they can't easily migrate off of.

Thing is (me personally speaking) i have an ideological preference towards decentralisation and I'd prefer if people more got used to having decentralised infrastructure rather than sticking to the old model (in form, not function).

Not a solution. Defeats the point of decentralisation, putting most (like 90%+) users in one instance. Big instance is sold to Venture Capital Firm because a bunch of amateur moderators call moderate the whole of twitter... and just like that enshitification shall commence.

How so? Folks who care about decentralization can use the menu, no? A common theme in the comments is that most users do not care about decentralization and don't want to have to pick a server. All that scares them away to centralized platforms like Bluesky and Threads. Even a big centralized fediverse server is better than yet another walled garden they can’t easily migrate off of.

Even a big centralized fediverse server is better than yet another walled garden they can’t easily migrate off of.

No it's not. If a single server holds a critical amount of the fediverse's content, they can enshitify.

The reason why the fediverse is resilient to enshitification is due to the fact that it makes migration less painful: If you want to abandon Xitter, which is centralized, you will be unable to access Xitter's content, which is why it took so long for people to abandon it; but if you want to abandon... let's say... mastodon.world, you can just make an account on another instance and still access the same content. For enshitification to occur, user's must be locked in, the federation stops that.

However, this system has one major vulnerability which can completely subvert the fediverse's ability to resist enshitification: centralization of content. If one instance holds a critical amount of content, they can pull up the drawbridge, that is, de-federate from all other instances. You might think this would upset the users, but it wouldn't. Most wouldn't know what federation is, all of mainstream is on the default instance, only the computer nerds are on other instances, so if suddenly, the default instance de-federated from everyone else, and thus becomeing a walled garden just like Xitter, few would notice and fewer would care. And now the default instance is centralized just like Xitter and the enshitification cycle repeats.

If you want an example of this look no further than Gmail. More or less 95% all emails are Gmail. If Gmail de-federates from your instance, you are removed; that means Google can basically dictate what other instances are and aren't allowed to do. If you do something Gmail doesn't like, they can de-federate and you instance is now basically useless, since you can't email 95% of people. Gmail could easily kill Proton Mail by de-federating.

Let's say I was on a giant Mastodon instance. And they defederated. At that point, would I be able to easily migrate to a smaller one? Or would I have to start up from scratch on the smaller instance?