Follow

I'm happy that I'll be part of two meetings this month in front of the EU Commission talking about decentralized social networks and the work we've done socialhub.activitypub.rocks/t/

I believe this conversation is an important one to have, and I'd like to expand on why below.

You've probably noticed that governance bodies all across the world (including the US, and in the EU but definitely elsewhere as well) are taking note about the shape of social networks.

Here's some early draft legislation from the EU. It says "gatekeepers" 16 times in 6 pages:

There's good reason to be concerned about gatekeepers to our communication, so 16 times in 6 pages isn't so surprising.

Yet, we've *built* alternatives to gatekeeping social networks across ActivityPub and other fediverse projects... how many times are those mentioned? Not once.

There's another way to look at this situation: as my friend Randy Farmer said to me: "Society is interpreting centralization of communication as a pathogen. What we're seeing right now is an immune response."

Having an immune system is important to keeping a human body, or a society, healthy.

However, immune systems can misfire. What we've seen in the COVID-19 pandemic is that what's killing people *is* the immune system response.

But regarding centralization: we have a cure.

The federated social web community has been at this work for over a decade. I'm proud to have been part of the ActivityPub standard: w3.org/TR/activitypub/

There's also further research happening for the next generation of this work in @spritelyproject.

But...

Centralized social systems are playing a losing game: there's no way to make everyone happy. But the immune response we're seeing could also misfire, resulting in only the creepiest social network possible being allowed to play.

This would be bad. But we have an alternative.

I think for the most part, legislators aren't even aware there's another option.

But recently, @jack@twitter.com did mention an alternative option in front of a congressional hearing, that we can build protocols as an alternative to centralization: twitter.com/jack/status/137512

Whether you're optimistic about an organization like Twitter launching an alternative or not, it's important to start getting that message out there. This is a good first step to counter an immune response misfire.

You can see what was actually said here: youtu.be/u7ijNBTDUcM?t=169

This is a good start, but it's not enough. It should be clear from watching the video that most people in congress had never even heard of such a direction before... certainly didn't know this work has been ongoing for over a decade.

We need to make ourselves visible, show a real effort for a system that empowers and delivers agency to our citizens.

Legislators should know that this is a viable path, otherwise they'll write legislation that assumes gatekeepers-only, because gatekeepers is all they know.

I said before that centralized social networks are currently playing a losing game: there's no way to make all political parties happy.

But a top-down solution which *assumes* gatekeepers will *codify* a losing game for the citizens legislators are *intending* to protect.

There is a path out. Instead we should have governing bodies rally around & *demanding* people-empowering decentralized solutions. Solutions which give agency to all citizens. Solutions without centralized gatekeepers, but which also empower communities to self-govern & moderate.

There is also much to be done in the centralization space; this is why after ActivityPub was released, I decided to turn my work to fill in the remaining pieces with @spritelyproject.

I'm optimistic we can build a better future and get it out there.

Let's do so together.

Well this one was meant to say "much work to be done in the *de*centralization space" but to be honest with you: I think @spritelyproject is going in some powerful directions that even the centralized space hasn't yet explored.

To be clear about this one: I am not suggesting that we should mandate a *specific* technical standard. Sometimes the world changes, and codifying a particular moment can lock us out of a better future.

I think instead we should recognize and support decentralized tech efforts.

@cwebber I am so happy to see you being part of these meetings! Good luck.

@rysiek Thank you! And thanks for all the work you've done to push things in the right direction here. :)

Yes, there's huge potential in what we are building here together. In addition to the inspiring toots of @cwebber a call to action. I'd like to point you to #SocialHub, the community of technologists that is dedicated to evolving the #fediverse and build further on the great foundations we already have.

Your participation and our concerted efforts are vital to go next-level: Towards #SocialMediaReimagined all across the board.

Not necessarily in size. That's irrelevant. In vision and humanity!

Check out the topics that interest you on the SocialHub forum, and join the fun.

socialhub.activitypub.rocks

@cwebber

@cwebber except for states (and therefore the EU) , the centralisation of communication is a goal. centralised services are far more similar in nature than decentralised services, more amendable to surveillance, etc

@cwebber do you have a link to this document? The part in the screenshot suggests that all the smaller platforms are lumped together as irrelevant and unable grow because of the gatekeepers?

@cwebber If I understand correctly, you're making sure the regulators know that things like the Fediverse exist, so they won't accidentally cause us problems like with Article 13 (the "All websites must scan every post for copyrighted media" mandate), alongside potentially supporting non-gatekeeping services?

@cwebber @how Politicians, who are having coffee with GMAFIA lobbyists on a regular basis, are being told not to listen to people of the fediverse, because it is just an amateur club that doesn't contribute to economic growth. As a result we are being reduced to exactly that and hence forced to DIY.

The catch being that infinite economic growth is impossible..

@cwebber Politicians also can not remember that the internet has had a long start period when it was not commercial from 1970 to 1993.

Yet this is an essential fact.

@cwebber

Yeah, the work we've done.
Got rejection 10 and 11 by NGI Zero and DAPSI meanwhile.
Sad to see ActivityPub software becoming closed source due to the lack of EU funding.
America first ;)

@ColinTheMathmo

Your chart is ready, and can be found here:

solipsys.co.uk/Chartodon/10601

Things may have changed since I started compiling that, and some things may have been inaccessible.

The chart will eventually be deleted, so if you'd like to keep it, make sure you download a copy.

@dredmorbius I haven't set up the cron job yet, but when I do it will run automatically. Currently I run the script by hand, then it checks for mentions and creates the chart(s).

So mostly ... yes.

I need to worry about running it while it's already processing previous mentions ... it's not re-entrant.

So I need to set a flag while running to inhibit a re-run, then it's ready to go.

@dredmorbius You can invoke it by replying to a toot you want charted, and mentioning the 'bot by name.

Would you like to test it? We can delete the toots and charts afterwards.

@ColinTheMathmo This collides with my other Mastodon frustration: lack of search.

(I'm going to see if I can't dig up a thread ... somewhere.)

@dredmorbius Search is an interesting question ... some people are very, very insistent that they don't want all their old toots searchable. But I'd like to be able to search *my* old toots, but then I'd be able to trace other people's contributions, even if they don't want them searchable.

There are vile people who dig up old posts and twist them, so the "right to be forgotten" is a real issue ... having everything searchable makes that hard.

Tricky.

@dredmorbius There are a lot of toots in that discussion ... this iteration is 146, then it need to see what toots they reference, etc.

This will be some time ... I have a delay to avoid triggered the surge detector.

CC: @SearchMe

@dredmorbius <grin>

Current estimate for this iteration is 11 minutes remaining.

Would you like to check out a "Gather" space I've created? We can have a quick video call. If so, you need to be on a laptop/desktop with FF or Chrome. If not, no bother.

@dredmorbius ETA for this iteration is 20:38 BST - six minutes ...

I expect there will be at least one more phase.

@ColinTheMathmo First impression is that my previous "compose storms as a set of first-generation replies to a common parent" really blows up your graphing engine. That's why this particular graph is so "wide".

Though it also makes the discussion structure really clear. And you can see which specific items/threads do take off.

@dredmorbius Yup.

Would you want that first-generation replies thing to be "fixed"?

@ColinTheMathmo I'm ... not sure.

My own usage is mostly historical, though there are a few other threads that will resemble this one or worse. My "widest" ran about 80--90 toots, by memory.

"#tootstorm" from my profile page should turn up a few, you might need to go to mastodon.cloud, my previous home, to find the long ones.

I'm not sure what an improved flow would be. I'm thinking "helix", for absolutely no valid reason whatsover.

(See previous Gattaca toot.)

@dredmorbius No problem.

Your chart is done ... hope it helps and is interesting. Your large number of toots that were actually a thread, but not reply-sequenced, really did stress the system and makes it hard to read.

I can "fix" that in this case, if that would be helpful ... I might look at trying to automate a fix.

Hmmm ...

@ColinTheMathmo Right, I just commented on that.

I'm now using glitch.soc's built-in "threaded" reply most often, which creates chains of toots. A reason for the wide mode was to make editing-without-orphaning easier (something I still do with some frequency).

(And this was immediately obvious to me on seeing the output, not sure you realised it.)

@dredmorbius I didn't realise about the edit thing ... noted.

My DiscDAG has a feature ready to push that lets you edit nodes "in place" and the edit appears as an overlay to make the history clear without losing the threading/connections.

Sadly, Mastodon doesn't do that, although if an "edit" were a leaf comment with a correction, Chartodon could detect and render that.

@ColinTheMathmo Also FYI this was a substantive discussion (at least for me), and if you can keep this around, I would appreciate it.

If you need to nuke it, I'd appreciate a ping, though of course, this is regret at losing what I didn't have or know could exist a half hour ago.

(You've passed directly from Potential through Luxury to Sentimental and well into Essential, within a half hour.)

Oh, and thanks a ton!

@dredmorbius You can download the SVG, and/or email you the DOT file and command.

@ColinTheMathmo SVG has been got.

DOT file: so this is graphviz under the hood?

@dredmorbius Exactly so. I have a bash script that recursively decides what to pull, I use "toot" to pull threads, then find references in them, etc.

Once enough toots have been pull I have a python script to generate a DOT file, invoke DOT, then upload the SVG.

I use "toot" to reply to the invocation.

All pretty simple, really. And messy.

@dredmorbius I'm thinking of changing the "make-tree" script to try to autodetect tootstorms that aren't threaded and insert invisible links to semi-thread them. I don't think it's worth it in general, and I might be being nerd-sniped by you discussion.

Sign in to participate in the conversation
Octodon

The social network of the future: No ads, no corporate surveillance, ethical design, and decentralization! Own your data with Mastodon!