Rob Ricci
banner
ricci.io
Rob Ricci
@ricci.io
Research Professor in the Kahlert School of Computing at the University of Utah

Not planning to be here much, my main account is on the fediverse at https://discuss.systems/@ricci/ - you can follow it directly here as
@ricci.discuss.systems.ap.brid.gy‬
You've seen me sat this in other threads, but this action perpetuates the status quo and makes it less likely that this other world comes to be.
January 25, 2026 at 3:33 AM
Which is fine on some level because decentralization is not an a priori good, it's one means to other ends. But I want people to go into these kinds of conversations with data in addition to anecdotes and ideology.
January 24, 2026 at 10:06 PM
I'm not giving up on Bluesky PBC, it may still happen, but we seem very much to be in a mode where the company itself is, on an organizational level, the "enemy" of a decentralized network.
January 24, 2026 at 10:06 PM
Let me put it this way: it is extremely uncommon for something to start out centralized and become decentralized. The reverse direction is much more common (see: search engine and email markets). Just because you build the protocol to allow it doesn't guarantee that it happens.
January 24, 2026 at 10:06 PM
But instead of actually moving toward the decentralized network that they have, on a technical level, made progress on, they seem intent to just keep making organizational decisions to stay on the centralized path. If anything, they seem to be moving *more* in that direction.
January 24, 2026 at 10:01 PM
They did not of these things. I'm not saying that their reasons for doing what they did were bad. I get why, in a network that is centralized for practical purposes, the felt like they had to do it, and had to do it this way.
January 24, 2026 at 10:01 PM
3) Governments, if you want to have verified handles you need to 2 and/or we will set you up as a trusted verifier and you can verify your own handles.
January 24, 2026 at 10:01 PM
2) Okay, but you need to use a handle on a government domain.
January 24, 2026 at 10:01 PM
1) We're not hosting our content on our servers, set up your own PDS if you want to join (this is not particularly meaningful in practice since it's invisible to most people, but it's technically-possible stance)
January 24, 2026 at 10:01 PM
With this ICE thing, they could have said:
January 24, 2026 at 10:01 PM
it's going to have to give up on the 'everybody sees the whole world' plan because that is not web-scale. But I'm sure their engineers know this and it's probably doable with evolution.

They problem is that they are actually building things that give them more choice and are not using them.
January 24, 2026 at 10:01 PM
Oh no, I'm not taking it as difficult at all, I'm quite interested in having these conversations with people!

My point is actually that on the *technical* side, I think they're doing the right things. It's a really interesting protocol, there is cool tooling for it, etc. At some point...
January 24, 2026 at 10:01 PM
And maybe it was the right choice. But it's worth keeping an eye on.
January 24, 2026 at 6:08 PM
I disagree that's is a neutral act, in particular, Bluesky has built a system for distributing trust with respect to verification, for example enabling governments to do it for their own accounts, doing it themselves was a choice.
January 24, 2026 at 6:08 PM
No, I am not missing your point. I know you're glad they verified ICE. I'm not. The Panopticon is about performance of surveillance, and I'm not happy that Bluesky helped in this performance. Yes, I agree that it's a good thing that people know to keep them out of their feed.
January 24, 2026 at 6:04 PM
That's fine as far as it goes, if that's what they and their users want. But I want to watch it.
January 24, 2026 at 5:59 PM
Sure - I assume this is probably the same assumption that the Bluesky verification team was working under. (a) I'm not sure that it being clear is actually better, see my Panopticon comment, and (b) Bluesky has made a lot of noise about building decentralized trust, but is acting like Twitter 2.0
January 24, 2026 at 5:59 PM
I see your point, but I can't really get behind blaming people who told us they did the thing instead of the people who did the thing
January 24, 2026 at 5:54 PM
The don't *have* to reach them, this is Panopticon stuff
January 24, 2026 at 5:53 PM
The harm in verifying ICE is that it's doing the fascists' work for them, bringing attention and legitimacy to an account that's there to intimidate people, without them having to actually do anything at all.
January 24, 2026 at 5:45 PM
and

> The intimidation of ‘we are here, you cannot escape us’ is the point, and the accounts by the regime are deliberately trying to provoke an outrage. ... Fascists intuitively understand this difference, and are skilled at exploiting it.
January 24, 2026 at 5:28 PM
As for ICE, I agree with the points made in this article: connectedplaces.online/reports/fr15...

chiefly

> Bluesky built a verification system designed to distribute trust, and then didn't use it when it mattered.
FR#150 – On ICE, Verification, and Presence As Harm
Bluesky built a verification system designed to distribute trust, and then didn't use it when it mattered.
connectedplaces.online
January 24, 2026 at 5:28 PM
I'm glad you are. Not everybody is. And I'm well aware that numbers and charts are nowhere near the whole story, but they are one way into starting to understand it.
January 24, 2026 at 5:28 PM
Basically, they need to get *off* the Twitter 2.0 path if they don't want to just, you know, be Twitter 2.0, but they have had multiple chances to turn off the path and are not taking them.
January 24, 2026 at 4:59 PM
I just think the the "future company is the enemy" model, the company is cruising pretty fast down the "enemy" path
January 24, 2026 at 4:56 PM