Via this Lemmy post I just bumped into this: An Open Letter Calling for the Resignation of Eugen Rochko (Gargron) from Mastodon Development.
Posting here because the broader discussion on the topic is very relevant and this is a brainstorming area where we can think them over. I hope we can - in a very constructive way - find ways to improve on some of the pain points that are brought up, and possible others that are as yet unnamed.
Also: Note that I do not endorse the open letter.
You are right, this is part of a broader discussion (and also a bit similar to the problems of centralization in the Matrix ecosystem).
I think the people that have been raising these and similar concerns over the years are mostly non-technical users of the larger instances that feel insulted if the response is “fork or host your own” as usually done (I am guilty of that as well).
I think they feel insulted because it reminds them of their (to a large extend imagined) tech-illiteracy and lack of capability to take matters into their own hands. It is a kind of self-inflicted wound and has a large overlap with other forms of self-identified victim-hood.
I am not exactly sure how to address this, as in reality it is not technically difficult or very expensive to host your own instance of a Fediverse server (or XMPP server for that matter). Sure it could be made a bit more stream-lined here and there, and not all software is equally suited for it, but the real problem seems to be rather the self-image of the people complaining (so they will not even try to do it).
The only “solution” I can think of is trying to foster small communities that can pool resources and help each other to overcome emotional, technical & financial hurdles. This might for example mean starting a crowd-funding drive to have one more technically inclined person start a self-hosted instance or maybe even a fork.
But as long as people are feeling isolated and seeing the Fediverse as some sort of outlet to their (often only perceived) victim-hood, the same kind of appeals to authority and later complains about said authority (if they don’t react as hoped for) will come up again and again.
I think the people that have been raising these and similar concerns over the years are mostly non-technical users of the larger instances that feel insulted if the response is “fork or host your own” as usually done (I am guilty of that as well).
Yes, this is a very good point. I decided to move this to Fediverse Futures community, because of my interest to take feedback and translate/transfer to technical realms.
For regular non-technical would-be fedizens its very likely they encounter Mastodon first and also to choose a large instance. Mastodon is relatively well-productized and the big instances feature prominently. With little understanding how fedi works it is easy to become exposed to its bad aspects. Some fedizens then leave again, while others move on to different instances. But rarely they move on to a different microblogging app Pleroma is tiny in representation, and most people probably haven’t heard of e.g. Hometown. Luckily more and different apps are joining (e.g. GoToSocial), so gradually there’ll be more choice.
Btw, I am not worried about people leaving fedi and don’t share Eugen’s vision of offering an alternative to the traditional social media platforms for the masses. I feel for fedi it is way more important to retain and further improve its culture and lively communities. And that works best on a path of slow growth, while the technology matures. On the other hand I am worried about the fediverse not evolving fast enough, and becoming irrelevant and slowly languishing. Our Spiral Island must be stronger to weather future storms.
The only “solution” I can think of is trying to foster small communities
Communities are absolutely key and crucial to fediverse. I advocate we take community way beyond mere instances.
communities that can pool resources and help each other to overcome emotional, technical & financial hurdles
A “Tragedy of the Commons” for the Fediverse is imho huge problem that indeed needs to be tackled. There are few people willing to do the community work, advocacy and the boring chores (also a problem on SocialHub). On technical side (and maybe funded too) I’ve been working on Fedi Foundation, but if that picks up steam I don’t know.
On a non-technical side there’s more interest in offering bundled services to fedizens, which is very positive development.
As with all things Mastodon is not perfect, but I do think that it brings a lot of value. It’s one of the biggest open social media platforms, and it’s done more to help wean people off commercial platforms than any other project I can think of in recent history. ActivityPub and the whole Fediverse are direct byproducts of the effort speared by Mastodon, and probably wouldn’t exit today otherwise.
When somebody starts a project that they put time and effort into because they have a particular vision for it then it is their right to run the project the way that makes sense for them. If their vision diverges from what a significant number of people want, then people should put in the effort to make their alternative vision a reality. They can even leverage all the hard work that was put into making the original project since Mastodon is open source software.
This sort of thing regularly happens in open source world. GNOME forks like Cinnamon and MATE are great examples of this. The original project started getting too bloated for a lot of people, and they got together to fork it and move things in the direction they wanted.
Another option is to make a separate project entirely. Pleroma is an example of an alternative to Mastodon that was made because people wanted to do things differently. In fact, Mastodon is itself is based on OStatus protocol extending the work done by GNU Social project.
Personally, I agree with the reasoning in the reply to the issue on GitHub and I do think it’s valuable for the official client to prioritize the needs for new users. The underlying functionality for supporting local and global timelines is not being removed, and it’s possible to make an alternative client that leverages it. Tusky for Android is the client I’m using, and it supports this feature.
On the other hand, I am ideologically opposed to partnering up with entities like EUnomia, but this being a social media platform the posts are already public and I don’t think anybody should be using such a platform for anything they want to keep private in the first place. Ultimately, you’re trusting server admins for any instance and you have no idea how they use the data collected by the instance.
Finally, as an open source maintainer, I can tell you that it’s hard and thankless work. Funding open source projects sustainably is incredibly difficult, and I’m really glad to see that Mastodon managed to secure a decent amount of funding for itself.
Yes, this is exactly how I think about it too (see below). And I can subscribe to that notion of ‘thankless work’ too, but more from the perspective of doing community work and fedi advocacy in general.
I feel the ball is in the court of the critics to come up with solutions and alternative ways to mitigate the issues they now have. There are valid reasons to cast a critical eye to some of the developments of Mastodon vs. Fediverse at large, and this is a good opportunity to start to address them.
Yeah completely agree with that. It’s also worth noting that even people who are non technical can still create a vision for a platform and then try to convince technical community why it’s worth implementing. Alternatively, it’s also possible to get together and crowd fund a project which lets the funders exercise direct control over its direction.
As you’ve noted in your comment, there are broader issues that need to be discussed here. What makes Mastodon successful compared to other similar projects, what are the problems that can’t be addressed by simply running an instance that’s moderated in a particular way, or developing another client with a different feature set. If these solutions address the concern, then I don’t really see what the issue is.
non technical can still create a vision for a platform and then try to convince technical community why it’s worth implementing
Yesss! This brainstorming space is exactly that, and an input to the more technical oriented SocialHub community. SocialHub is about technology direction, evolving open standards and the ecosystem that rests upon it. But a lot of feedback for this must come from the fedizens directly. Anyone can help.
There are many interesting topics on the SocialHub forum relating to cool innovations, that would greatly benefit the Fediverse.
Another point where quite a few federated app developers are critical on Mastodon project, is that most of the time they go their own way in their own community, oblivious or ignoring or not participating in developments elsewhere (like SocialHub, which is THE place for this). It is an “our way or the highway” stance, probably resulting from their dominant position, plus the fact that in broader community consensus is harder, and things may move more slowly.
My opinion on this is: No problem. This is also a FOSS project choice. But SocialHub should be more in the lead to set the overall technological direction. Then not following the open standards & best practices would automatically mean that a project would side-track itself. A strong SocialHub is needed for that. Unfortunately many app developers don’t realize that their participation there constitutes a win-win for their project, and are too absorbed in their own work.
Alternatively, it’s also possible to get together and crowd fund a project which lets the funders exercise direct control over its direction.
There are two threads that more or less relate to this:
- Fedi-wide governance body (Hamish Campbell): What would a fediverse “governance” body look like?
- A foundation (might be funded) to SocialHub (Me): Organizing for SocialHub Community Empowerment and Presenting Fedi Foundation
Slight word of caution here… one of the grand-daddies of federated communication, the Jabber / XMPP federation has developed over its 20 years existence such governance structures in the form of an officially recognized standards body and so on, yet many of the issues you describe for the ActivityPub Fediverse remain today. Sometimes even causing people to unnecessarily break off and create a new protocol all together, but in the end replicating most of the mistakes (and making some new ones) that XMPP already tried to learn from years ago with varying success.
Yes, important observation. I am very wary of funded efforts, as it is somehow controversial and leads to a lot of complexities to organize it well in any grassroots movement where “herding cats” is the reality and infighting and power plays, conflicting interests and strong opinions are the talk of the day.
Your mention of XMPP led me to write about, what I see as: The Fediverse Challenge. And it rises well beyond this mastodon discussion.
Yeah, SocialHub seems to be exactly the way to go about doing this. I completely agree that we want to have some baseline in terms of protocols like ActivityPub that projects that participate in the Fediverse use, and these need to be flexible enough to accommodate different use cases. Beyond that projects are free to run any way the authors want, but we can create social pressure to encourage project maintainers to not stray too far from what’s considered acceptable by the larger community.
Indeed. IMHO especially stressing the importance of being part of such a community as active participant should be highlighted. To many FOSS developers going at it in their own way individualistically means missing out on a win-win of broader collaboration, that is a requirement for their own project’s future.
Btw, specifically related to the topic of Mastodon vs. Fediverse, there’s talk about creating a community-driven fork of Mastodon based on Hometown. See: Discussion: Mastodon and the Fediverse (comment) by @wakest.
Agreed, I think the main advantage Fediverse has over commercial platforms is being open. Commercial platforms want to keep users on their site and make it difficult to share content between them because they’re in a zero sum competition for the users. On the other hand, Fediverse creates a positive sum scenario where everyone benefits from having more content on the network. So the focus absolutely has to be on growing Fediverse as a whole as opposed to individual sites.
Yes, but there’s a big problem in this. It is already hard to find contributors to a FOSS project. It is even harder to find people to volunteer on all the community work and chores that need to happen to foster good collaboration between different FOSS projects.
Some people don’t see this as a problem, with the argument that grassroots movements just go their own way completely organically and anarchistically. While this can be (and often is) a strength, this is less true for evolving the common technology foundation on which all of the ecosystem has to stand. (I created the Spiral Island analogy for that… there is a hurricane of bad tech trends to withstand),
I use Mastodon and chose an instance that is small and aligned with my interests. As such the local timeline is very useful, so I would not want to lose visibility on it. However, I use the mobile web and I still don’t see any point using an app to access Mastodon (the same goes for Lemmy). This new iphone App drama about a design choice does not make sense. Asking for resignation because you don’t like some design choice? Seems a bit extreme when that app does not have to be used at all. There are many other apps to choose from, using your favorite browser works too.
Yes, I favor web interfaces too, instead of apps (that often come with all kinds of crap and trackers in it). For Mastodon however, I use FediLab installed from F-Droid.
(Copying my reaction from SocialHub below…)
Yes, @heluecht , this is also how I feel. Eugen started the project, and brought years of dedication to it. He has practiced BDFL from the start (he’s even mentioned on Wikipedia for that), and people joined on that premise. In true FOSS fashion there’s the choice to ‘take it or leave it’. Or fork it, and he’s open about that too.
So, I won’t sign up to any open call for him to step down. Instead I’ll make a call to anyone disagreeing to Eugen to consider their options instead. And this includes myself, as - at the time I created the Humane Tech Community account (which has since become more personally affiliated to me) - mastodon.social seemed the most appropriate choice (hint: not anymore for me).
Still the open call has one Good Thing™ going for it: The discussions it triggers.
Some broader issues that could be addressed:
- Has Mastodon become too dominant on the Fediverse (in a ‘mail vs. Gmail’ kinda way)?
- Is this dominance helping or hurting the evolution of the Fediverse and its technology foundation?
- Is there a risk of the Fediverse ‘forking’, one part on a Mastodon track and another on the ‘standards’ track? Can we avoid that?
- If Mastodon is indeed so dominant, then what are the reasons for that? What can we learn from that?
- Why until now didn’t SocialHub/SocialCG manage to position firmly at the center of fedi evolution?
- How can we encourage Mastodon (and Eugen) to address some issues we may have with his projecct?
- If more forks of Mastodon arise (some already exist), how can we help set them up in most desirable way?
And there are many smaller issues that relate to things still missing in the standards & practices or lack of broad adoption. Many of these open-ended discussions in this forum.
(Like I would ideally move to a different microblogging app altogether, maybe Pleroma. But I don’t want to lose all followers when doing so, and my toot history. There’s network effects + FOMO that kept me stuck till now)
This happened a couple years ago with #trending, and my response is the same. The whole point of the GPL, is that you don’t have to agree on everything to cooperate. Fork the code, spin up your own instance, and thank the BDFL for building the majority of your platform.
I agree. But that reaction is mostly from a tech perspective. I think the outrage here comes from non-technical fedizens that don’t have the option create a fork all that easily. But at least all the recent discussion has led people to think of creating a community-driven Mastodon fork.
Yeah, non-technical ppls reaction should be, “who wants to make a fork with me?”
I don’t like that the app is the official app. I have just started a geographic instance with the purpose of ensuring the local content is blooming. Now I need to make sure that people who register doesn’t use the official mastodon app. If they do, they may get the impression that my site is about what is happening on mastodon globally.
I think this is going to make it harder for new users to stay on my instance.
Yes, this is a good point. Though the issue you raise is a more universal one. In a future fediverse there’ll be more and more different application types that are interoperating in all kinds of ways. The question of ‘which client supports what?’ becomes much more prominent then. Potentially very confusing for fedizens, unless specifically addressed.
On the one hand this can be done via standardization or strong consensus on what ActivityPub extensions to use, and Capability Negotiation, and on the other hand - on the client side - a move to more “universal clients” (supporting the Client-to-Server part of the AP spec, and NOT the Mastodon API which can never keep up with all this) is a way forward.
Offering an intuitive UX for that will be challenging. You might say that with all its different apps, the fediverse represents a kind of ‘cloud-based-appstore’, but that doesn’t cut it. I think at this time the fedi will become more service-oriented, while the client devices may become more task-oriented. See also: From silo-first to task-oriented app design.
I never quite understood the obsession many people seem to have with apps. Especially nowadays when many apps are more or less websites wrapped in react-native or such anyways.
Alternative front-ends like Pleroma is doing, or maybe a multi-account PWA client like https://pinafore.social/ seem the better way forward.
So I kind of agree that server fediverse software should probably more cleanly seperate the front-end from the back-end and that will also help with apps, but the possibility for communities to add or customize their own web front-end is probably the more important aspect of it.
There’s more developments that go towards this, like GoToSocial, which is backend-only server platform.
