Thoughts on Threads and ActivityPub | Hacker News Reader

Thoughts on Threads and ActivityPub

anderegg.ca

66 pointsGavinAnderegg56 comments

> Google has, perhaps, the worst track record for chat clients and for killing their products in general. It’s a real shame how Jabber/XMPP users were effectively “ghosted” by the far more numerous Google Talk userbase, but does this mean that we should ignore any efforts form big companies to support open standards forever now? Weren’t things good while everything did interoperate? Is it absolutely inevitable that Meta is going to do something terrible with ActivityPub?

FB messenger (back when each chat was a pseudo-window at the bottom of the FB screen, and not its own app) also supported XMPP. you can't say (effectively) "i buy the XMPP argument, but this time it's not Google" because the XMPP argument applies equally well to FB.

But I'm both cases XMPP is still around, if you care for it. Neither Google nor Facebook extended it in a way that broke the protocol when they stopped supporting it.

I look forward to being able to follow my friends who are on threads from my Mastodon account. If in a couple years meta stops supporting it, alas, we had two years.

They just used it to hijack XMPP users from the federated Jabber network onto their own platform. Google did at least. And that's what Meta is probably going after with Threads, growth using the fediverse userbase. Just say no and defederate it before they screw everything up again. Your next "friends" are going to be ad serving bots powered by AI trained on your feed.

> And that's what Meta is probably going after with Threads, growth using the fediverse userbase

Facebook has 3 billion users, probably 90% of Mastodon users are on one of their services already. The entire Fediverse userbase is not just less than a percent of their userbase, but less than 1-2% of their annual growth, it doesn't even make a dent in the first derivative

It probably doesn't pay for the Threads servers and salary costs if they vacuumed up every last Fediverse user. At the bottom of all these EEE conspiracies is always a completely inflated sense of self

That only proves they can achieve growth without the fediverse, no need to ruin everything they touch for a mere 1 or 2% and fake it for being cool and supporting open standards to get the regulators off their backs. But then it's all about revenue, infinite growth and user data hoarding.

> that's what Meta is probably going after with Threads, growth using the fediverse userbase

This seems like a terrible play given how small the fediverse userbase is.

They are targeting the enthusiasts . Then the enthusiasts do the rest. It's a good tactic.

Sounds like they forgot to do their market studies.

The whole boondoggle is happening because many Fedi instances are ran by those very same enthusiasts. They chose to settle on Fedi for various reasons, and I’m pretty sure Meta transgresses most of those reasons.

Personally, I don’t care too much whether my server federates or not, but I’m a very casual user.

Two federated protocols co-opted by them have a similar story to tell. Email, besides XMPP. Both still work. But both are impractical if you want to selfhost. XMPP lost a huge portion of its users. And the vast majority of email users are on just two providers who either send mails from outside into the spambox or drop it silently. The logical conclusion is to not let these big players on to the federated network if you care about the federated nature or userbase of these networks.

XMPP never had a strong federated ecosystem of users that existed outside of GChat.

ActivityPub already has a few million users and a thriving federation

It got much much harder to run an email server because all the hoops that had to be jumped through to reduce spam made it a nightmare. I'm not sure I really blame Google/Microsoft for that.

I have a pretty decent experience on Fastmail and had at one point reasonable experience on Protonmail as well (except that they could never manage to import my enormous 20 year archive of email :D). Most every ISP in the world still offers email, friends and family that have those addresses still manage to get their emails through. There are also plenty of newsletter hosts mail services that don't get dropped or put in spam.

Granted those aren't self-hosting, but these other hosting options just reinforce that it's more a matter of spam prevention than it is evil companies being evil (to be clear, I accept they are evil, and they do evil things all the time).

> But I'm both cases XMPP is still around, if you care for it.

great: then DM me at xmpp:colin@uninsane.org and lets talk more.

if this were 2008, your browser would render that as a link and clicking it would put us in direct contact, because likely than not you'd have an account somewhere. all i want is for that experience again.

but chances are you don't have an XMPP setup anymore, in which case my invitation above is useless. the specific protocol (XMPP, Matrix, Activity Pub ..) is less relevant to this: it's the social experience (and norms) which matters.

today, Activity Pub is growing back into those shoes which XMPP used to occupy. i can drop my handle at the end of a tech-related blog and people do take up the invitation. this year, (ex-)coworkers and IRL friends are actually using it. give it a decade, and maybe that invitation would be familiar even to my parents. 10 years sounds long, until you consider that's about as long as we've been without such a thing already.

and so it's like this: 1) how much credence do you put on that hypothetical future? 2) how much of a risk does FB pose to it? because if you believe we're on a good path, and that FB has even a 20% risk of derailing us from that path, then why risk it?

Around 2008 I was indeed running my own XMPP server for a while. I didn’t give up on it because Google was temporarily-xmpp-compatible-and-then-not though - I gave up on it because all of the apps were terrible :P

Conversations, Cheogram and plenty more have solid user experiences, video calling, and services like jmp.chat provide gateways to other communication platforms.

The Jabber ecosystem has not stood still, it is competitive with other over the top messages and calling options.

Messenger supported XMPP clients but didn't federate with other servers.

This is correct. Messenger never really was XMPP. It just had a client API based off of XMPP. There were a bunch of quriks and no federation. I suspect the idea was just that it provided a starting point for clients.

my memory was just of running Pidgin and being able to reach everyone through it (google, FB, weirdo self-hosters). guess it must have had good multi-accounting support and i was relying on that. i assumed it was using the same code paths for all my contacts, but if not then i guess the "FB used XMPP" point would be almost entirely moot.

I used FB Messenger over XMPP for the last handful of years before it was shut down using Pidgin and I'm pretty sure that's how it worked. You could talk to people on different networks but you needed your own FB account to talk on Messenger and you couldn't do group chats cross-network.

GTalk and self-hosted XMPP could federate and you only needed one account and conversations could span servers.

I'm convinced that federating Threads offers two major benefits to Meta. And that this is why they have no incentive to EEE.

First, federating is a clear signal to antitrust-bodies that, no, there is no monopoly. With the EU shifting gear in this, and the US also pushing back at big-tech, there is a real "danger" that these companies will otherwise be forced to build interoperable social-media. Now, Meta is the first mover, and can do it their own way.

Secondly, by federating, you give yourself leeway to block and ban people. If you want to block, say, the POTUS, you can now say "sure, but you can just set up your own instance". Where, in a siloed social media, you're practically blocking people's speech, with federation, you're saying: 'sure, speech whatever you want, just not on our instance'.

Call it elitism, snobbery, parochialism or whatever you like, but I maintain that as a space on the internet becomes more and more mainstream, it gets worse. Eternal September writ large. I remember the old internet, when it was for nerds, and there were barriers of entry. It was a much better place - it was rough around the edges, sure, but it was not some corporate walled garden monetised monoculture, so the good was absolutely worth the bad.

There are some small vestiges of this - HN being one, and the fediverse being another. Reddit circa 10 years ago was like this, and, well - look at it now.

The Facebookification of the Fediverse will be the beginning of the end of that old internet holdout. It will bring the normies and the advertisers, who will file down the sharp edges and narrow the range of acceptable viewpoints until it's just another part of the commercial borg.

It was nice while it lasted I guess.

That internet is still there. Harder to find, but still alive.

There’re still countless personal websites and blogs, small communities, old school forums.

Like the author, I never fully understood the EEE argument.

The assumption is that if Meta convinces a large section of Mastodon users to use Threads, and then makes Threads incompatible with Mastodon, Mastodon, as a network, will suffer harm.

But Meta clearly doesn't need Mastodon users. They've built a network something like 5x larger than Mastodon in a handful of months. If anything, Mastodon needs Meta.

In my opinion, the worst case scenario (Threads becomes the defect Mastadon client/author and then drops support for Mastodon) just means Mastodon will end up back where they are today. A niche social graph for a specific set of users.

EEE works as an argument when the target is larger than the malevolent force, but falls apart when the target is so much smaller.

I don't think the assumption is that many peoe will migrate from mastodon to threads, but rather that people migrating from other platforms will migrate to threads over "really federated" services, stifling growth of mastodon and preventing it from becoming a serious threat. Which is exactly what happened with xmpp iirc.

And yeah, my money is on threads cutting loose from AP at some point in the not so distant future, at which point the whole thing becomes indistinguishable from "mainland" Facebook.

If people move to Threads over Mastodon in this way it means that Mastodon isn't useful to people. The answer would be simply to improve Mastodon.

The network might be larger, but who is generating more engagement & content?

Threads is absolutely generating more content, at least in terms of volume. Quality wise, I think it depends on what type of content you're interested in.

I agree. Having Threads federated will make it much easier for individuals to pick Mastodon as they won't miss out on the network effects of a larger network. They can still follow their friends and be followed. I find it very unlikely that the number of people who sign up for Threads rather than Mastodon is is higher than the number of people that will be "able to" use other services because they can follow and be followed by the large community on Threads. It seems that if Threads is federated than every Threads user is a win for the Fediverse and every non-threads user is a big win for the Fediverse. The only "lose" case is someone joining threads that would have otherwise joined something else.

The biggest strength of the fediverse is that every instance operator can make their own decisions on this matter.

If they want to keep their instance up to date, Threads is influencing instance operators whether they like it or not.

https://gleasonator.com/objects/95399039-9e07-497d-8414-f218...

Diffs?

> The biggest strength of the fediverse is that every instance operator can make their own decisions on this matter.

If you look at how Spotify applied this strategy to the once-open medium of podcasting and successfully redefined "podcasts" to mean shows on their closed and proprietary platform, this seems unlikely to matter.

Exactly. I just don't see this as an either/or. I do worry about embrace/extend/extinguish, and I don't think "well it's an open standard" is an answer. See gmail.

That said, it's better for them to be on it than not be on it, and it's great as a vote of confidence in favor of Activitypub over Bluesky's go-it-alone alternative that I find counterproductive.

A little of column A, a little of column B.

I have personally yet to see something that got better by fb/meta being "on it" (for an of course also very personal definition of "better")

I have my foot in both the ActivityPub and atproto camps as nothing more than an interested observer/developer, but I want to defend Bluesky a bit here.

What Bluesky are really building is atproto. If bsky.app and atproto were in a house fire and the team could only rescue one, they'd rescue atproto. Throughout this year there was such demand for a Twitter alternative that wasn't Mastodon, people jumped aboard bsky.app despite it only being a testbed/PoC for the development of atproto.

The Bluesky team reviewed the existing protocols for distributed public conversation and none of them checked all the boxes. So they built their own. Their goals were [1]:

- Account portability. Being able move your entire social graph (identity, posts, follows, likes, etc) without the previous server needing to cooperate or even be online [2].

- Scale. A "big-world" view of the entire network to enable global conversation.

- Trust. Letting users build custom feeds so you can control what your timeline looks like and being open by default.

One complaint against atproto is: why not work instead to improve ActivityPub? As someone who has worked with both protocols, I agree with the Bluesky team that it would have been too difficult to retrofit these features into ActivityPub (portability and scale, especially). Plus, a lot of Mastodon power users have philosophical disagreements with the things required for "big-world" global conversation so it's unlikely Bluesky would have received a warm welcome anyway.

So, on one hand, "go-it-alone alternative" is not flagrantly wrong. Bluesky is the only app using the protocol and it does fragment the space. But it's important to remember there are reasons for why they're going alone on this.

(Sorry, I've had these ideas swimming in my head for the past few days and your comment spurred me to organize my thinking by writing it down!)

- [1] https://blueskyweb.xyz/blog/3-6-2022-a-self-authenticating-s...

- [2] https://atproto.com/guides/faq#why-not-use-activitypub

> every instance operator can make their own decisions on this matter.

Not entirely. "The fediverse" has this concept of "fediblocking". If a large enough group of admins decides they are very much against this, they will not only block Threads, but block anyone who does not block Threads.

Basically strong-arming every admin who does not follow suit, to do as "the hivemind wants it to".

This has some positive sides to it, but it's also a weak point in the idea of decentralization.

> If a large enough group of admins decides they are very much against this, they will not only block Threads, but block anyone who does not block Threads.

I don't think this is true. The instance my group is using isn't participating in any of the fediblocking/blocklists stuff, and hasn't been blocked by it.

They will be if one of the fediblock proponents decides to report you. This is not an automated process at this time. It's decided by a cabal on a private Discord, really.

Lot of consternation over being blocked by a bunch of losers you don't want to talk to anyway.

It's really about the non-losers who you may want to talk to that get swept up as collateral damage

What the hell? I don't use the fediverse but that sounds extremely bad.

I feel like it's also its greatest weakness.

Threads will be a boon to mastodon.

Discovering the various instances or the idea of federation via threads certainly isn't going to hurt it. I am curious what kind of stress tests will happen with that many more people added to the network. EEE isn't meta's goal here though. The good does not wash out the bad, nor the bad the good, as stanis baratheon would say.

Meta is a horrible company for how is weaponizing data and manipulating consumers to make more money. I remember seeing papers published by them on how to foster “micro-depressions” on young consumers to make them more pervasive to ads. Nothing against them on engineering from me, just consider them immoral and cancerigenous.

Microsoft under Ballmer was the evil of EEE. Now under Nadella is slowly doing this but more subtly and elegantly. Let’s see the marriage with OpenAI involve. Their engineering mentality was quite nice until recently, let’s keep tabs on GitHub to see them evolve.

Google is trying to be evil, but their own internal struggle from going to Internet God to 80s Corporation (with greed and politics overshadowing the real business) makes them go around in circles.

Long story short: happy FB adopted ActivityPub, not going to touch them anyhow as I consider them immoral. Afraid they will try to influence the standard in the future to cripple it.

I don't think Meta will intentionally try to snuff out Mastodon but I don't think Threads breaking from Mastodon will have it just return to its pre Threads momentum.

Look it just makes sense to start out connecting Threads users to a pretty active user base which creates content for Threads users. Then as Threads grows, maintaining federation with Pubiverse will have a high marginal cost relative to the shrinking piece of pie users outside of Threads is. Compatibility will get worse and worse as the Threads devs allow their federation code to rot until eventually they make a clean break to focus their resources elsewhere.

Meanwhile if you happen to be a poster on the Mastodon side of the fence, you're going to have to get on Threads and start posting there to keep up with the audience you've gathered, it just makes sense. That network effect of posters needing to follow their audiences would hollow out a post-break Mastodon.

All this being said, I don't think Mastodon instances should resist a Threads federation. Focus on a superior user experience by really trying to empathize with the average user. Don't let yourself be relegated to power users or enthusiasts because, Threads federation or not, when push comes to shove people want to talk to people and they will go where people are.

> Look it just makes sense to start out connecting Threads users to a pretty active user base which creates content for Threads users.

Are there any decent analytics of mastodon's network? Threads has many more users than mastodon, and so many keep posting on there about having better engagement than on xitter. Mastodon interop being axed on the threads side would waste so much of their work, unless the mastodon networks themselves start to empty out.

If another protocol becomes dominant, threads might pivot, but I think the current devs I've spoken to are genuine in regards to thinking of threads as just another good part of the fediverse. Mosseri and Zuck have control in the end, but the 18 devs or so that make threads right now seem to have been pretty thrilled by the rollout.

Though it's not finished by any means I think the AT Protocol has a really great future potential vs ActivityPub.

ActivityPub is cool, but I think long term it will be a stepping stone to something a little more like the end goals of atproto.

what makes ATProto better?

All the users in a single Postgres database that a bunch of project members appear to have full access to based on the screenshots they've been posting to brag about the user count /s

> The Embrace, Extend, Extinguish argument also falls down at the last point for me: Extinguish. This is a tremendous amount of effort that Meta is undertaking to try and… what? Stop the fediverse from growing? Defeat Mastodon, Meta’s mighty competitor?

Well, if the author had actually cited the people whose arguments they're dismissing, the answer to these questions would be clear.

Here's one.

https://emacs.ch/@ramin_hal9001/111579818136072605

-----

In case of the article being edited after I leave this comment, here's an archive snapshot of what I'm responding to: https://web.archive.org/web/20231215205458/https://anderegg....

> after attracting a critical mass of users large enough to decimate the user base of the competing Mastodon network

Threads already has 10-100x as many active users as Mastodon. How many more does it need to "decimate"? How much growth can you seriously expect Threads to gain from federation?

> At this point, people begin to wonder what the point of Mastodon even is.

This is already the case for many, many people. It's true that more people would end up thinking that if Threads successfully federates with Mastodon, but that's because the number of people who even know what Mastodon is would increase. A non-user is a non-user.

You're welcome to disagree with that poster's argument.

I just shared it because I don't like when people don't cite the arguments they're refuting. It seems dishonest, even if no malice is intended. Thus, I sought to correct it.

I don't have a strong opinion on Threads' existential risks to Mastodon, personally. I want nothing to do with Meta as a company, ever, and would simply block them if I ran an instance (regardless of whether they pursue 4-E or remained good neighbors).

I do not, and never will, trust the company behind Threads. I don't need to spin more cycles on it than that.

I think the OP and most of the commenters here are missing the point. Meta threat is not about killing AP or Mastodon. It's about mining your data and spying on you. That AP is an open standard doesn't change anything. It's exactly like SMTP and gmail. Sure, gmail did not kill the emails, they are still here, it's still decentralized. But eh, when you're sending an email with vacations pictures to your friend, it's enough that a single one of them is using gmail for Google to have everything. The content, the addresses, build a social graph of everyone, and mine, mine, mine the data to spread more ads or share it with the governments or... Threads is the exact same threat. If only one of my friend is using it, everything that I write will be received and processed by Meta. And I don't want that. Until now, I was able to tell to my non tech friends, come on my mastodon server, and I knew we would have privacy. Now I will have to explain to everyone that if they don't want their data to be analyzed, they should not add a contact from Thread. The fediverse was a safe space. Now we have to constantly be on guard again, and this exhausting.

Btw if a lawyer is around, I would be curious to know how legally they can store my messages on their server (because technically that's how Activity pub is working) while I did not accept any terms and conditions... I guess there is something like "while connecting with a Thread user you accept that we store your data", but because I don't have an account I did not accept those conditions... In the end, Solid is a much better approach than AP or diaspora, kudos to Tim Berners Lee

If you choose to have a public account on the fediverse or let someone from another server follow you, you are granting the right to duplicate your data and store it on the other server.

How that data can be processed for a private account is a gray area, but you have authority to determine whether your content is public, and if it is not public who is allowed to see it.

Unless you have a private profile they can scoop up all of that data whether or not you have a Threads follower. If you do have a provide profile it is pretty easy to deny follow requests to people subscribing from Threads.

I don't think this is a major issue, it can be dealt with on a case-by-case basis by the owners of private profiles.

> Google then used this for its chat service, but then stopped interoperating with standard Jabber/XMPP clients. This left a lot of people in the lurch, and that’s terrible.

I mean, yeah, that's basically the thing we're afraid of!

Reply on news.ycombinator.com