I'm sorry, but by what metric are musical bands "dead"? I'm asking because I follow a lot of bands that are actively releasing new music and touring across the US and Europe. Not to mention the musical festivals.
Mastodon has changed quite significantly since a few years back. We take product design seriously and spend a sizeable amount of our resources on improving usability and reducing friction. If you could, please try again, and let me know how it goes this time. If you are an Android user, I strongly recommend our official app, as in my (obviously biased) opinion it is the best social media app right now and the user experience I am most proud of.
From cloudflare 2024 stats bluesky traffic shot up during the US election but is now back below (aggregrate across all servers) mastodon traffic [1]
But its true that mastodon did not have a major breakthrough as of late and bluesky will likely surpass it in the near future as some important "high information quality" communities (journalists, scientists etc.) seem to migrate there in preference.
Orientation towards general (mainstream, non-tech) users, easy usability etc is indeed a problem for the fediverse. The reasons are mostly an anti-commercial ideological stance which on the one hand makes funding scarce. Hence brilliant open source products - there are many more than peertube - remain unpolished, not marketed at all etc.). On the other hand this hostile culture keeps mainstream actors from joining the revolution.
But make no mistake this is a revolution. The hyper-concentration in social media is an aberration that does not fit any other pattern in society and the economy. Some more pragmatism from the decentralization pioneers will accelerate the inevitable.
Arguably, Bluesky being spun off from Twitter and having Jack Dorsey as one of the founding members is a somewhat Meta-like advantage, in a sense of immediate legitimacy in the press and networking opportunities/connections in Silicon Valley. Mastodon had to start absolutely from scratch. I had zero connections to anyone important when I launched it. Bluesky also raised over $8M in venture capital funding, while Mastodon was being developed on a $0/mo budget for the first year of its existence, and something like $5000/mo for the next 5. Our current annual budget of around $500K still pales in comparison to the money Bluesky has at their disposal right now to spend on e.g. marketing. They also have the advantage of not really trying to do decentralization. That being said, venture capital money isn't free, while Mastodon's funding comes from the community with no strings attached, so in the long term, I believe in our approach.
To add some color to my comments: I also believe in your approach and I admire the work you and your team have done.
To sum up my entirely unoriginal opinions:
1. Mastodon has far better usability than any other Fediverse software I'm aware of
2. Despite this, usability is still a material coefficient of drag on Mastodon's growth
To be clear, I don't believe Mastodon has to or even should aspire to match the growth of other more centralized networks; only that usability is a drag on what would otherwise be natural growth for Mastodon itself.
I know you and your team spend a considerable amount of time and energy on usability, so I hope I'm not saying anything you don't already know infinitely better than I.
That there is any discussion on "what" instance to join. I understand this is the whole point of being a decentralized system, but your average person doesn't care about that for social media. Maybe I'm wrong and mastodon.social is perceived in that way similarly to how bsky.app is for Bluesky.
I haven't followed mastodon in some time but the account migration thing was a pain as well, which I feel that the AT Protocol addresses much better.
Mastodon.social. These enormous monolithic servers offer the worst experience and degrade the whole network. And now joinmastodon.org points you at mastodon.social first, then a list of the largest servers.
mastodon.social runs nightly releases off of our GitHub. Anyone can run nightlies to get these features. Or wait for the stable. What's hacky about this feature?
Mainly that it only benefits giant instances. That's a convenient feature for people on a tiny instance following users on mastodon.social, but adds friction for users on mastodon.social who want to follow users on a tiny instance.
I haven't looked at the implementation, but I'll assume it's flawless. The issue is that it doesn't address the broader problem. Like the "official" iPhone app that has a giant, colorful "Join mastodon.social" button above a transparent "Pick another server" option, it serves to push people toward m.s and away from a good federated experience.
> I’m not sure I grasp the argument here. No feature should be switched on until all 23,000 servers have updated to the necessary version?
It's not about having the necessary version. The feature has to be cookie driven, right? Otherwise, mastodon.social wouldn't be able to remember that an unauthenticated visitor has an account elsewhere. Such a cookie almost certainly won't be available across servers (thanks, ad trackers for ruining it for everyone). That means a user would have to fill in that form for every single remote instance they visit.
Think of it this way: suppose a user visits a tiny Mastodon server at social.example.com. They click a user's follow button. How will social.example.com know to redirect the user to the home instance they configured when they visited mastodon.social?
> If I’ve downloaded the Mastodon app without having a server in mind what is the app supposed to do, just list them all?
If someone clicks the "pick another server" option, they're taken to a perfectly serviceable chooser. That should have been the default. It works for all the other apps that don't default to mastodon.social.
> That means a user would have to fill in that form for every single remote instance they visit
It looks like you're correct; I went looking and found that https://universeodon.com had the new feature as well, and I had to type in my home server URL again. I actually thought this was a great solution before I realized you'd have to do it for every server; hopefully they find a better solution for this.
If it worked that way, I'd support it completely! I think the idea is great. I just don't think it can be implemented in a cross-instance way, short of having some sort of central server.
With the feature the way it is today, it makes it a little easier to follow users on mastodon.social, and a little harder to follow users anywhere else. Suppose that hypothetical social.example.com instance has 1 user. If that feature's enabled on the server, anyone wanting to follow that 1 person has to complete that extra step with zero benefit. It's only useful for larger instances, benefitting the few centralized servers without helping the federation as a whole.
> But the thing is, I just hate the Mastodon UI. It's so clunky.
I hope this situation improves once we finish hiring a product designer. Personally I don't think the web client is as bad as some people make it sound, but I know it can be improved. In the interim you can try the 3rd party web client Elk [1], some people seem to really like it.
Mastodon has the same UI problem as Matrix (Element specifically). The UI/UX isn't _bad_, it's just a teeni tiny bit more worse and confusing in everything than the competition.
If someone would just make a 1:1 clone of Discord and slap it on Matrix, it'd be perfect. The underlying technology is fine, the interfaces are the clunky ones.
Mastodon has a bit of the same, just following someone from another instance is like 6 steps. Basically you need to copy their URL, go to your own instance, search for them using the same URL and click follow if it's the only search result. Compared to Twitter's one click that's a good 500% increase in complexity.
This PR is just one part of item MAS-86 on our public roadmap [1] "Markdown formatting for posts" which is meant to enable users to use code spans/blocks and bold and italic text as championed by the Elk third-party app [2]. I believe "rich text" in the PR title is a little not nuanced enough because like before I'm of the position that Mastodon is the wrong medium for long-form articles and the various block-level formatting structures that come with it, but allowing users to use bold text without resorting to mathematical Unicode characters is a win for accessibility.
Gargron, I know that you are adverse to quote-tweeting, but I have a lot of lefty friends who want to stay on Facebook despite all the privacy concerns because it lets them share memes/news stories with their IRL friends, sometimes adding context, without having to worry about their boss / landlord / XYZ seeing what they're sharing.
I think that the fact Mastodon allows per-post privacy is a great feature, but the lack of private retweets and/or retweets with context (eg quote tweets) does scare some people away from the platform.
I've seen how quote tweets can be abused, but still believe there is some potential good from the feature in Masto. Any thoughts?
(Yes, it would be trivial for a third party frontend to add quote tweets, but because that's trivial I'm not asking about that.)
There are forks who use URL unfurling to simulate quotes, so I think this can be a good approach if you can convince your instance to migrate... but, of course, every other instance won't see them nor will the desktop/mobile clients that don't implement this feature.
Other fedi implementations (Pleroma, Misskey) seems to support this approach too.
For now, easiest way seems to be a manual quote, as we did retweets before twitter had RTs.
I don't understand why quote-boosting can't be a user-selected option.
• Allow users to opt-in/out the ability to perform a quote-boost.
• Allow users to opt-in/out the ability to see others' quote-boosts.
• Allow server admins to opt-in/out the quote-boost feature.
Wouldn't this solve everyone's issue with quote-resharing?
It effectively can't have that. Quote statuses are practically just a special way of formatting a link to another status in the message text at the moment. There's a proposal iirc to try and create a more genericized version by extending AP, but it's stalled if memory serves me right.
Even so, you're going to have implementations that will happily ignore those choices; as long as the body of an external post is obtainable in some way (right now most info is just in the embed meta tags), someone can make a version of it that ignores permissions.
Any permissions system is, of course a polite suggestion since ActivityPub pushes content to a large number of other peoples' computers in the typical case. The whole network depends on most servers being relatively well-behaved so I don't think that's a barrier.
Perhaps. My comment wasn't based on technical feasibility, but specifying the particular concern which is why Mastodon and several other ActivityPub projects do not support quote-boosts.
Get on a server running the glitch or treehouse forks, 5000 characters and quote-toots. Mainline Mastodon is very much in catching-up mode with its own forks.
