Isn’t the greatest advantage of Mastodon that it allows you to migrate wholesale to another instance? I was reading Cory’s piece on this the other day, and it’s true that you don’t often see this feature written about much as yet because they’re still trying to compare very different services. Should there be some Muskrat type bad actors running an instance then you can just jump ship, or is this oversimplifying it?
This feature will become more prominent, because there is nothing about running a Mastodon server that means that you are good at running a Mastodon server. Elon Musk isn’t an evil genius – he’s an ordinary mediocrity who lucked into a lot of power and very little accountability. Some Mastodon operators will have Musk-like tendencies that they will unleash on their users, and the difference will be that those users can click two links and move elsewhere. Bye-eee!
Dead on, as always. The critical mass thing isn’t about whether or not Mastodon needs it, as you point out, but instead about whether or not it could replace mass-market social media with something decentralized.
My OP was about it replacing one of these large networks. To do that my parents need to want to move away from say FB for some reason, which would require their friends and the rest of my family to do so. And for my sister to move, her business reach would have to be unimpeded by moving off Twitter.
Those two things are very much not in the goals of most early Mastodon users, but as you point out:
Totally right again, IMHO! the Fediverse needs to be large enough that it’s the place of choice instead of the monolithic entities, but… that’s it. “growth” after that isn’t relevant. If it’s where people who want to network go, then the goal is accomplished. I agree with your insight here.
Mastodon is kinda-sorta making this happen! There are genuine conversations happening about how, and in what form, Governments should get involved in this process.
He totally did; his willful fuckery did way more to drive millions of users to Mastodon than it ever did to attract anyone* to Twitter.
*Excepting those with nefarious intent, of course. The agents of chaos and discord love this shit.
well… discord is an entirely different thing. more like the bbs than mastodon
Continuing the discussion from How to Mastodon:
Hey, Musk dropped $44B without due diligence or understanding how to monetize Twitter.
Move fast, grab control now, break things, figure it out later.
“We’ve added advertising and user data collection, but now people are ‘forking the code’? Is that hacker slang for property theft? Call the lawyers!”
“We have very commited engineers at Twitter. I know because they commit to GitHub after every sprint.”
What my instance currently sees (Mastodon being the big blue part of the ring):
If the Nazis could have their own software so that they stand out, that would be great!
Well, I’ve made the plunge and shut down my Twitter. Be the change you wanna see… decentralization is literally 5 orders of magnitude harder than simple facebook-style centralization, but… we gotta start if we want to eventually get there, right? So I’m on Mastodon and not Twitter any more. Good riddance.
So yeah. You, too, should give Mastodon a shot. Like Discourse, this very software, it is open source so we can collectively make it better together, at least in theory?
Cloudflare has a Mastodon-compatible project Wildebeest for quickly setting up instances.
While it’s open source, I notice at least one prune in the pudding:
Some features like data persistence, access controls, and media storage are handled by other Cloudflare products:
- D1 for the database.
- Workers KV for object caching.
- Zero Trust Access to handle user authentication and SSO on any identity provider.
- Images for media handling.
Most of our products offer a generous free plan that allows our users to try them for personal or hobby projects that aren’t business-critical. However the Images one doesn’t have a free tier, so for setting up your instance you need to activate one of the paid Images plans.
There’s a philosophical problem too. The Fediverse makes it easy for users to pack up and migrate to another instance, not even running Mastodon, if things go sour. This will lock instances into Cloudflare, making it messy to move if (when!) Cloudflare changes its offering or TOS down the line.
I don’t think there’s any lock-in there - most of Cloudflare’s offerings are either inline caches or inline security products. They don’t change access URLs or anything of the sort, so disentangling your instance from them should be more more difficult than changing DNS records.
This is true for Boing Boing as well, btw, as we are supported by Cloudflare’s Project Galileo.
There aren’t a lot of free options out there for protecting sites, so I’d imagine this will be pretty popular, especially if the image handling is optional instead of required.
Nice of them to be open about their project, and to shut it down after negative feedback, but I’m sure that there are others who aren’t so open.
Hm. “WellKnownBot”, that’s new.
I am blown away by this. This is so unheard of on the modern internet that it feels absolutely novel to me.
Read not only the EFF post, but the author’s retrospective. It is fucking brilliant, and the developer basically says:
- I planned out a very specific experiment to solve very specific problems
- I specifically made the opt in but also looked for opt-out signals already in use to piggyback on and respect
- I specifically thought about how bad actors could use this and made sure I wasn’t creating a new vector before I began
- I specifically thought about vulnerable populations and how their data is available in the fediverse today, and made sure I was not introducing a new way this data could be spread
- I used existing APIs and authentication mechanisms that required a user to explicitly give the experiment access before it did anything
Followed by the most important bit to the author:
- I announced what I was doing and explicitly said I would stop if the community disagreed
The author then went on to explain the feedback he received, then shut down and deleted the experiment even though there was no technical reason to do so, just to avoid community discontent.
I’m sorry, but this is fucking amazing. I echo the EFFs take here. What commercial service ever has behaved this way? This is so unusual in this space that it seriously feels novel, and perhaps the best part of this whole thing is the retrospective the author wrote that spelled out that:
- getting this data is so trivial that almost certainly people are already doing this without anyone’s consent
- individual fediverse instances can do this in private and never tell anyone,
- the fediverse is almost certainly going to go down this route sooner than later especially given commercial interests, and perhaps most importantly…
- Mastodon needs to think really hard and quickly about making sure the technical restrictions on sharing data outside an instance match folks expectations that it can be restricted, because today they very much do not match and there is no defence against a bad actor doing these things.
Seriously, this is amazing, kudos to the author for being this deliberate, transparent, and considering the wider community like this.
What a contrast to Cambridge Analytica, Facebook, and really, all the rest of the lock-in social media platforms.
An iOS app.
4 months ago (Received 6 minutes ago)
Huh. Reminds me of the old days of Usenet lag.
Is that because someone on your instance or followers list boosted an old post that previously wasn’t in your orbit?
My instance is just me. I do follow EFF directly, and maybe Friendica is gradually backfilling. I really should update to the current release.
eta: Okay, I think I see it. I follow EFF, and there was a recent reply to that post from someone that I don’t follow that pushed the original post to me. Hm.