> However, I disagree with some of the analysis, and have a couple specific points to correct.

[cwebber@social.coop]

People want Bluesky's devs to prevent their content from being replicated and indexed by people they don't like, well, I think it really is that: a *conflict*.

People were encouraged to join a Twitter replacement, they are expecting Twitter-like solutions. Can't blame 'em.

[cwebber@social.coop]

Given that "anyone can replicate and index!" is literally the *entire* design philosophy of ATProto, it's not going to be something easy to solve. I don't have an answer, but hey, I'm working on fairly fundamentally different designs, so it's not my problem to solve.

That said...

[cwebber@social.coop]

Like the present-day fediverse, Bluesky was majorly popularized by a bunch of queer people early on. As a trans person I watched a bunch of my friend join and felt so safe they posted things they never would have in today's environment when the community was small.

[cwebber@social.coop]

The decision about whether or not to boot horrible, well known transphobic people (protip: answer is yes) from the platform seems clear enough to me. I'm not sure the "speech vs reach" approach is working.

And it seems to me people are finding they don't have tools in their hands to do anything.

[cwebber@social.coop]

For all its faults, and there are *many* and I have *railed* against the instance-oriented approach to moderation on the fediverse and have been writing about and working towards alternatives for a while, instance moderation empowers better here.

I think this will be a real test for Bluesky.

[cwebber@social.coop]

But more broadly I think *neither* the present-day fediverse nor Bluesky meet the needs of the future.

The "global town square" is a social media concept invented by centralized social media in the early web 2.0 era.

Social media by millenials, for millenials. What's the future?

[cwebber@social.coop]

So to some degree, I don't have a lot of interest in trying to figure out what the solution to this is, because I think these are the wrong designs. I don't like the context-collapse firehose much at all, I'm interested in "contextual communication", "secure collaboration", and "healthy communities"

[cwebber@social.coop]

That's the kind of direction we're trying to build towards with @spritely, but as said, I'm dropping the values discussion here, that's something we'll talk about later in the week. I would like to talk about that independently, focusing there on what to build, not on a critique.

[luis_in_brief@social.coop]

@cwebber I agree with this as first-principles, but I do note that every contextual/healthy/secure community that I'm a part of spends a pretty solid amount of time sharing and discussing stuff from the context-collapse firehoses (for many appropriate contextual reasons - laughter, learning, safety, etc.) I am not sure how we square those facts.

[cwebber@social.coop]

But I do think there's a big collision course ahead, and I don't know how it'll resolve. Investors and users who want quick resolution to real concerns on one side, a vision for public, highly replicated and indexed by anyone content on the other side.

It'll be a challenge.

[cwebber@social.coop]

There's opportunities for collaboration maybe. I've asserted pretty strongly that Bluesky isn't decentralized and as a system, it isn't. You can't tear the power dynamics out of the analysis. Otherwise what's the point?

But Bluesky uses decentralization techniques, there may be collab space there.

[cwebber@social.coop]

I'm not trying to be a mean, horrible person to Bluesky's devs. I'm really not. I actually think that they've provided something much *better* than X-Twitter to a lot of people.

But Bluesky has speedrun this whole thing so fast, Bluesky is already no longer the underdog. It's Twitter TNG.

[cwebber@social.coop]

And that means we can't pretend that decentralization is something that's some future possibility or goal, that it's gonna happen some day we promise.

I'd love to be proven wrong on everything I laid out.

Though I think the only way to do that without being worse than AP is serious rearchitecting.

[cwebber@social.coop]

Who's empowered and who has agency and how we can increase the agency of everyone is indeed, all I care about. It's what "decentralization" means to me and matters to me as a goal. You can't drop the power dynamics. It *is* about the power dynamics.

I want us to build a better future. A real one.

[cwebber@social.coop]

One thing I am confident about: it's not that Bluesky's engineering team doesn't care.

Actually I only really know two of Bluesky's main people well, Jay Graber and Bryan Newbold.

I know they do both care.

But so did Twitter's early devs. Twitter was supposed to be decentralized too.

[cwebber@social.coop]

It's easy to forget that Blaine Cook led a team at Twitter in early days to make Twitter decentralized and the team there was worried about the effects centralization can have.

Investors killed it anyway.

It has to be more than about caring, the work has to happen and be preserved.

[cwebber@social.coop]

I've said enough. I've said more than enough. I've said more than people probably thought could possibly happen on the subject on a blogpost or social media thread let alone *two*.

And that's with me dropping part of the second blogpost because I fell and hurt my hand.

It's time to wrap up.

[cwebber@social.coop]

I hope I haven't caused emotional strain on anyone. I spent a while walking back from brunch and was pretty depressed and was talking with my girlfriend: was I just *mean* about this whole thing?

She reassured me she didn't think I was, but I still feel like I was mean.

I tried not to be.

[raccoon@techhub.social]

@cwebber
Yeah you've said so much that I don't think I will ever be able to get through the entire thread. :V

I did like some of the points that I saw when I skimmed through it though. :3

[cwebber@social.coop]

But despite there being literally millions of people on both Bluesky and the fediverse, I haven't seen any other analysis that went comprehensively into architecture, terminology, and their implications at the level I did in terms of their *implications* and *impact*.

I think it needed to be done.