Jack Dorsey skipped ActivityPub, built AtProto, lost Twitter, funded Bluesky, watched it become a company with VCs and a board, said it was "repeating all the mistakes," left, and now funds Nostr.
-
@mastodonmigration @baralheia @cwebber in Christine's article (and I've just spoken with her about it), it assumes a network topology that does not exist in the real world.
It assumes that every user is on a different pds, and every user runs a full network relay. The reality is that multiple users are usually on a single PDS, and there's only like 12 relays.
- 2 from bluesky (+ 1 deprecated)
- 2 from hose.cam
- 1 from blacksky
- 1 from upcloud
- 3 from firehose.networkplus a few more from various people.
In the ActivityPub ecosystem for every user to message every other user, you need connections between 30,000 servers.
For the same in AT Protocol, you need connections between N PDS to one or more relays (most use the bluesky relay, which others get their list of PDSes from).
@thisismissem @mastodonmigration @baralheia My analysis assumes a network architecture in which each node is a major participant in the functionality of the network, because as I argue in the piece, from a power distribution perspective of decentralization, it is important. What I describe in the piece is that if you want more than a pantheon of gods-eye view participants, then not having addressed delivery means that the system can't scale down.
And this is true: you can run a gotosocial node that isn't *dependent* on other major players in the network, and it scales down great.
The question is whether or not that matters and is important to people. Maybe it doesn't, I don't know. It matters to me, though.
-
@thisismissem @mastodonmigration @baralheia My analysis assumes a network architecture in which each node is a major participant in the functionality of the network, because as I argue in the piece, from a power distribution perspective of decentralization, it is important. What I describe in the piece is that if you want more than a pantheon of gods-eye view participants, then not having addressed delivery means that the system can't scale down.
And this is true: you can run a gotosocial node that isn't *dependent* on other major players in the network, and it scales down great.
The question is whether or not that matters and is important to people. Maybe it doesn't, I don't know. It matters to me, though.
@thisismissem @mastodonmigration @baralheia I also think that the ATmosphere and the fediverse could both learn a lot from each other. ATproto could adopt directed messaging, and the fediverse could adopt content addressing and portable identity, both of which I believe are important. This isn't a pivot, I have been saying these things *since 2017*, before ActivityPub became a recommendation, and before ATproto even came on the scene, including even in my co-proposal with Jay Graeber about what Bluesky could be, and including in the articles referenced previously.
Unfortunately, I think there is a real risk that the fediverse is going to learn the *wrong* lessons from the ATmosphere, and adopt some of its "shared centralization" components, rather than the decentralization components that ATproto has that the fediverse doesn't yet.
![😕](https://forum.wedistribute.org/assets/plugins/nodebb-plugin-emoji/emoji/android/1f615.png?v=8eaae6ed46b ":\")
-
@thisismissem @mastodonmigration @baralheia I also think that the ATmosphere and the fediverse could both learn a lot from each other. ATproto could adopt directed messaging, and the fediverse could adopt content addressing and portable identity, both of which I believe are important. This isn't a pivot, I have been saying these things *since 2017*, before ActivityPub became a recommendation, and before ATproto even came on the scene, including even in my co-proposal with Jay Graeber about what Bluesky could be, and including in the articles referenced previously.
Unfortunately, I think there is a real risk that the fediverse is going to learn the *wrong* lessons from the ATmosphere, and adopt some of its "shared centralization" components, rather than the decentralization components that ATproto has that the fediverse doesn't yet.
@cwebber
@thisismissem @mastodonmigration @baralheia Content addressing and portable identity is so important and hurts so much everytime a server closes or (like me) had to switch domain name. -
@thisismissem @mastodonmigration @quillmatiq @dansup @evan I appreciate this information, but my main interest in modern decentralized social media protocols is for social media (microblogging, as well as picture & video sharing), and I am not a developer - just a nerd. My primary concern in all of this is I simply don't trust Bluesky PBLLC. That's why complete and total independence from their infrastructure from a microblogging standpoint is important to me before I would ever consider using an ATproto-based service as my primary social media platform. While I'm okay using community resources for now, I *want* the ability to host my own completely independent full stack so I would never need to rely on a company or community to stand up a relay or AppView or something. I can do that easily and *cheaply* today with Mastodon or WAFRN (the ActivityPub side of it, anyway) or Pixelfed or Loops - it's not a big lift because these platforms were already designed with the full stack as a single deployment. This makes the network highly resistant to adversarial takeover, especially as things exist *right now*. It's a far, far, far bigger lift to stand up a similarly independent full stack of Bluesky (for example), and far far far easier for the majority of the network to be compromised if all statuses must be funneled through corporate or community chokepoints.
Tl;dr if I can't easily host a full stack that can run completely on it's own (while still being able to federate with others), that's not independence. *Every* service built on top of ActivityPub can give me that independence *by design*. It's the difference between human-scale networking and corporate-scale networking.
If there's an easy, lightweight, fully independent way for me to participate in Bluesky-compatible microblogging on ATproto with an experience that is comparable to what I'd get from using Bluesky directly, I'm interested to learn more - because I'm not currently aware of anything like that. I can get that from multiple platforms on AP *today*.
A cool project you may like is appviewlite: https://github.com/alnkesq/AppViewLite
You can run an appview entirely locally--i was even able to run this on my phone.
Its possible to directly crawl PDSes, meaning there's no reliance on a relay.
There's also https://reddwarf.app, which runs entirely in your browser, without an appview or relay.
And there's wafrn (which I'm using right now), which natively supports atproto and activitypub. It has its own appview, but it currently uses an external appview for notifications and fetching possible missing posts iirc.
-
@thisismissem @mastodonmigration @baralheia I also think that the ATmosphere and the fediverse could both learn a lot from each other. ATproto could adopt directed messaging, and the fediverse could adopt content addressing and portable identity, both of which I believe are important. This isn't a pivot, I have been saying these things *since 2017*, before ActivityPub became a recommendation, and before ATproto even came on the scene, including even in my co-proposal with Jay Graeber about what Bluesky could be, and including in the articles referenced previously.
Unfortunately, I think there is a real risk that the fediverse is going to learn the *wrong* lessons from the ATmosphere, and adopt some of its "shared centralization" components, rather than the decentralization components that ATproto has that the fediverse doesn't yet.
Goodness, yes - it puzzles me greatly that portable identity isn't a part of the ActivityPub standard. It's an obvious gap, in hindsight.
-
@quillmatiq @dansup I do know how toxic the response was to Bridgy Fed when Ryan first announced the Bluesky bridge, and I think his ability to weather that storm will go down as one of the most heroic efforts in the history of the social web.
To be fair, most of what I saw against Bridgy Fed wasn't against _bridging_. It was against (a) the initial decision to make it opt-out, (b) the justification "if we make it opt-in, people might not opt in, so making it opt-out will make it more useful". After Ryan listened to people explaining what's wrong with that, and switched to opt-in (respect to him for listening), I don't remember seeing further opposition, though of course there could've been some I didn't see.
-
@thisismissem @stefan the reason for doing this multiplier is not so much for getting MAU right in absolute terms, but because mastodon/fedi MAU data also includes lurkers in their data. So you need it to get a fair comparison between fedi and the atmosphere
@laurenshof So your estimate is based on one "skeet" from a CTO of a company with a multi-million dollar VC "loan" that *needs* their numbers to look good?
Fair enough!
-
To be fair, most of what I saw against Bridgy Fed wasn't against _bridging_. It was against (a) the initial decision to make it opt-out, (b) the justification "if we make it opt-in, people might not opt in, so making it opt-out will make it more useful". After Ryan listened to people explaining what's wrong with that, and switched to opt-in (respect to him for listening), I don't remember seeing further opposition, though of course there could've been some I didn't see.
@unchartedworlds@scicomm.xyz I don't see it either, but I can certainly believe that @quillmatiq@mastodon.social gets a ton of hate simply for being associated with BridgyFed, which is BlueSky adjacent, which is honestly enough for some people to instantly judge you as a bad person.
-
To be fair, most of what I saw against Bridgy Fed wasn't against _bridging_. It was against (a) the initial decision to make it opt-out, (b) the justification "if we make it opt-in, people might not opt in, so making it opt-out will make it more useful". After Ryan listened to people explaining what's wrong with that, and switched to opt-in (respect to him for listening), I don't remember seeing further opposition, though of course there could've been some I didn't see.
@unchartedworlds @quillmatiq @dansup he got brigaded here and on GitHub. It was toxic.
-
@mastodonmigration @baralheia decentralized *where* and *how*
Is ActivityPub really decentralized when everyone builds for compatibility with Mastodon (apart from Lemmy) or is it only decentralized in operations? Where mastodon.social accounts for a significant portion of the network? What about Pixelfed? How much decentralization there? Loops? I think there's only really one maybe two loops servers of any size?
Decentralization doesn't mean "run absolutely everything myself", I mean, sure, you *could* but that's expensive, complicated, and time consuming. Moderation? Most servers just import some blocklist snapshot at a given point in time.
Thing is, decentralization isn't the goal, the goal is better social apps.
Decentralization focuses on technology, not people. It's the "how" not the "why" and "for who"
"Thing is, decentralization isn't the goal, the goal is better social apps." And that's where I have to disagree. Decentralisation is the goal of the Fediverse and ActivityPub. It isn't for ATProto.
ATProto is designed to replicate a sort of "digital town square" where people meet and talk. ActivityPub in it's design is in a sort like the postal service where a instance can be a city, a neighbourhood or a single street/house.
-
@laurenshof So your estimate is based on one "skeet" from a CTO of a company with a multi-million dollar VC "loan" that *needs* their numbers to look good?
Fair enough!
-
@laurenshof Right. From the same person.
Well, again, fair enough, this all answers my question.
Appreciate it, both of you!
-
@julian @dansup @thisismissem @evan @quillmatiq
I need to call this one out, sorry.
"The smaller devs doing interesting things with fedi get absolutely fucked over for simply exposing the fact that... public stuff is public"
More and more, people joining this space are entirely unaware that this is a technical foundation of the space, and more and more people come to me as a service provider asking me why their content is on Web sites they did not sign up for.
-
@julian @dansup @thisismissem @evan @quillmatiq
I need to call this one out, sorry.
"The smaller devs doing interesting things with fedi get absolutely fucked over for simply exposing the fact that... public stuff is public"
More and more, people joining this space are entirely unaware that this is a technical foundation of the space, and more and more people come to me as a service provider asking me why their content is on Web sites they did not sign up for.
@julian @dansup @thisismissem @evan @quillmatiq
I understand why.
You understand why.
Thousands of people do not, and tens and hundreds of thousands being invited in will also not understand this.
We need to do better than "public is public".
-
@julian @dansup @thisismissem @evan @quillmatiq
I understand why.
You understand why.
Thousands of people do not, and tens and hundreds of thousands being invited in will also not understand this.
We need to do better than "public is public".
@jaz@toot.wales fair, and I'm certainly open to user education, but I refuse to engage with individuals who have demonized me from step 1.
I'm not planning to put in the effort to try to win back the trust of someone when I didn't do a damn thing to lose it.
-
@mastodonmigration @baralheia @cwebber on activitypub, if I have 30,000 followers (1 follower per server), and I want to post a message, my server has to send out 30,000 messages.
In AT Protocol, if I want to do the same write operation, I send one http request to my PDS, the PDS then publishes that message to N connected relays (where N =< 12)
@thisismissem @mastodonmigration @baralheia @cwebber
I would argue that neither the AP nor atproto are built directly for patterns of human communication that exist in the real world.
The "everyone messages everyone else and must have access to any message from anyone at any time" pattern embodied by the present Bluesky is not a real human way of communicating and forming communities, it is a figment of tech companies' imaginations and represents a massive amount of over-indexing that relies on, and therefore tends towards, centralized platforms.
The "everyone preferentially messages people in their nearby vicinity but sometimes people further away" view embodied by most present Fediverse software assumes a flat social network in which "non-local" is functionally the same in all cases and does not model human social networks very well.
AP assumes you are building bunch of villages with a flat road network between them. atproto assumes you are building Saudi Arabia's The Line.
-
A cool project you may like is appviewlite: https://github.com/alnkesq/AppViewLite
You can run an appview entirely locally--i was even able to run this on my phone.
Its possible to directly crawl PDSes, meaning there's no reliance on a relay.
There's also https://reddwarf.app, which runs entirely in your browser, without an appview or relay.
And there's wafrn (which I'm using right now), which natively supports atproto and activitypub. It has its own appview, but it currently uses an external appview for notifications and fetching possible missing posts iirc.
@irelephant yup, and this weird protocol tribalism hurts projects like wafrn and bridgyfed, because it drives a wedge between two communities that roughly care about the same thing: better social media/networking.
-
@cwebber
@thisismissem @mastodonmigration @baralheia Content addressing and portable identity is so important and hurts so much everytime a server closes or (like me) had to switch domain name.@maswan @cwebber @thisismissem @mastodonmigration @baralheia ActivityPub is mostly fine with decoupling domain name and identity. From my understanding, you could have an activitypub account on a self-hosted server and have multiple domain name have webfinger software point to it and that account would have multiple handle.
-
@maswan @cwebber @thisismissem @mastodonmigration @baralheia ActivityPub is mostly fine with decoupling domain name and identity. From my understanding, you could have an activitypub account on a self-hosted server and have multiple domain name have webfinger software point to it and that account would have multiple handle.
@gkrnours @maswan @thisismissem @mastodonmigration @baralheia ActivityPub is mostly oblivious to the concept of "instances". It's more actor model, and more like sending mail, and less like villages. Heck, ActivityPub doesn't even have webfinger. Technically, sharedInbox doesn't even need to be on the same domain!
Both instances-as-important and webfinger are activitypub-in-practice rather than activitypub-as-written. But maybe that's immaterial. "The purpose of the fediverse is what it does"???
-
@skarnio @mastodonmigration @baralheia as someone *actively* developing on AT Protocol, I can tell you that Bluesky PBC could disappear tomorrow, and we'd just work around it. There's complete mirrors of the did:plc directory, and we'd just pick one to replace the existing directory. Sure, it'd be hugely disruptive, but life would go on. We would work around it.
There's alternative relays, hostile migration of PDSes is possible, and changing the plc directory is possible. Blacksky probably couldn't handle all of Bluesky's users suddenly all using it, because they're still new, but *shrug* life would go on.
@thisismissem @mastodonmigration @baralheia When I talk about governance, the "disappearance" of a corporation is the least of the problems. The problem is what that for-profit corporation can do to the network while it retains total control. As long as the code doesn't officially transfer its governance to the community in the form of a non-profit organization or something similar, the technology will continue to be controlled by a corporation. Even with the best intentions of the contributors (which I believe are truly very good), that's the reality.
