2022 Unofficial Mastodon Survey Results

By shlee on 7 August 2026 • ( Leave a comment )

Post Views: 4

Introduction

I’ve spent the last few years sitting in the Mastodon Discord, listening to the developers talk about the project. During this time, the Core Dev team has been involved in a lot of conversations about the ecosystem and its future roadmap. Some of these discussions have been passionate.

So in July 2022 – as an experiment – I decided to send 25 questions as a survey to a small subgroup of Mastodon administrators and see how they felt about several topics to have some data for future discussions. I want to bring data to a discord fight. I hope the information from this report can be useful for the community to articulate their needs, and hopefully shape some discussion into the future roadmap of Mastodon. https://aus.social/web/@shlee/108645723393938414

As noted in my tweet, the immediate response was excellent. There were some confused #Mastoadmin posts wondering if my survey was a phishing attack, and that’s exactly the kind of cautious response you’d want from your local instance administrator. So Kudos to all of those people for calling me out for not adding enough proof of my identity to my initial email – this was an oversight due to my excitement to get the thing out of the door – I should have tooted before sending the email and linked to that.

Thank you to the 75 administrators who answered this survey’s questions! I didn’t expect such as high response on a cold email, and I appreciate that they all responded in good faith to the survey. Their comments to these questions were mostly well considered and well detailed. Due to this response, I’ve decided to put more effort into this results post which is going to result in a VERY long document.

As the target audience of my survey was exclusively administrators, my only wish is that the reader takes into consideration the needs of our community members (end users) on balance with the needs of these instance administrators (as I consider this survey to be full of bias, while also being full of actionable detail). The following will be anonymised, and I will only reveal wholly verbose comments (no edits, unless it’s abuse or identifying).

Mastodon Developers: I highly recommend you complete an official survey with more promotion and better prepared questions (One for Administrators and another for Users). People like the engagement. I will make more of the survey results available (not all of it) to the core mastodon development team if they request access to the information.

Survey Results

As following is a breakdown of insights from the 25 survey questions with a lot of flavour added from the comments that came from each question.

Question 1: “Do you believe your instance has a unique theme/culture? or is the instance just a gateway to the fediverse as a whole?”

22% of administrators consider their instance a generic node, and simply a gateway to the thousands of other federated instances and their users.

I have a big focus on privacy and data security so I get more registrations from privacy aware persons or people from restricted countries like china.

Neither. Our instance is our own community, having a unique theme is not required and but also not nearly enough. See https://runyourown.social/ for more explanation on this. I think there are a lot of misguided admins that think that just because they make a “themed” instance, people will obviously want to join it. Nothing could be further from the truth—instances should be built as self-sustaining communities, not just “themes”. The distinction is very important.

Only a general gateway, with strict moderation relative to French laws

Yes, the members on it are members of an organization.

Yes, my instance is for and by people who make things.

We’re basically just a group of friends gathered around the same instance. (One could argue that friends share a same culture, etc.)

instance is geared toward infosec professionals

It started as an instance mostly focused on tech, programming, automation and Linux. It ended up covering more and more as it grew, and by now I see it mostly as a gateway to the fediverse.

my instance is limit to speaker of Japanese language.

we’re limited to alumni and close friends of a particular study programme.

Yes. While gateway instances are appropriate at times, the instances I run have unique themes and cultures.

Unique theme insofar as it serves to entice Welsh-speaking, Wales-based, or Welsh diaspora members but otherwise general gateway. The culture of the instance is unguided – in other words there is no admin-led desire or goal for a certain type of instance or culture of instance.

Unique theme: Local, regional topics

yes, we have a unique theme, which attracts a subset of the furry culture

I do, yes, since it’s intended as a regional instance. But, it does have its own aspect of being a gateway to the rest of the fediverse.

Though it’s expressing regional relation by intention, it’s not the only one in that region and thus more a common gateway to the fediverse.

It has a unique culture, driven by users who know each other personally and now an expanded community.

find the idea of “themeing” instances somewhat odd in an intentionally federated system. I wouldn’t inherently call our server themed in the way it’s typically used here

50/50: We act, curate, and moderate as if we’re a general gateway to the fediverse, but realistically, most of our user base comes from my group of friends from two or three very longstanding Internet micro-communities.

Yes, our instance has a unique theme and culture, and we curate the servers we’re connected to as a way to prevent harassment so we’re not a gateway to the fediverse as a whole, just our clique within it.

Yes, x.x is geared towards those who identify as queer, or feel an affinity or tie with queerness, though it acts as a gateway to the fediverse as a whole.

our instance has a unique theme and a unique culture. It is a mix of people joining the fediverse because of that interest we catter for and people already on other instances making alts.

Just a gateway. I like the idea of having a unique culture, but it is not the goal or current reality.

Unique. It is the only Hebrew Mastodon instance.

Our instance is primarily aimed at German-speaking roleplayers.

We have a unique culture that is tight to the company culture where we run this instance from.

Our instance exists to allow BDSM and fetish models & creators to share their work with fans. So while it is a gateway to the fediverse, it is specific to that interest.

The instance has a high proportion of UK tech people, but it’s only implicit, not an intentional focus. I’d think of it as a gateway, not a particular place.

A super majority of administrators raised that their instance was built around a theme/culture. Mostly commonly language / Regions / sexuality, but also including friends/family groups, alumni/organisations, hobbies/fandoms, skill/talent and other life styles.

I believe adding community features which strengthen these instances should be prioritised – from allowing features such as “themeing CSS/UX customisation” to “sister/bubble instance timelines” to “group chats” to “federated groups/events”.

ActivityPub has the potential allow millions to communicate together, while also nurturing small subcultures.

Mastodevs: Only 16 of the 72 people stated their instance was purely a gateway. This is not a surprise to me, as it follows the creation of my own instance focused on my local region. Lots of admins borrowed the word unique from the question to answer how they saw their instance. They almost all talked about community/local/culture… (something to think about for Discoverability, Groups and UX)

Question 2: “Do you believe the fediverse should be considered one large ecosystem? (or are some instances islands?)

“66% Yes 33% No”

whynotboth.jpg the fact that it’s technically possible to have both a very large universe of interconnected servers as well as isolated cliques of islands is what makes the fediverse wonderful

One of the great features of a fediverse system is that islands can form but still be connected to the larger ecosystem.

This diversity in purpose makes the fediverse a heterogeneously connected network, and it can actually be one of its greatest strengths.

I like to think of the fediverse as a large connection of islands. I don’t think a large unified ecosystem aligns with the goals of Mastodon, I have blocked entire instances because they do not align with what I think is decent behaviour

I believe instances thrive best when they are expressions of local culture and then interconnected from there. Local culture emerges naturally when tools like local timelines are added.

Some instances are definitely islands, in a way, but I think having “one ecosystem” or “not one ecosystem” is unhelpful black-and-white thinking. The protocols, the software, and the culture around them are one ecosystem, but that’s not the same as the instances’ content and interactions being one ecosystem. It’s complicated, that’s a good thing, and that should be recognized.

As a historical matter niche spaces are important. Doing otherwise usually leads to harassment and creation of entirely isolated places. Unfortunately, most people that think that niche instances aren’t appropriate are feeding the trolls. They likely have never been harassed or targeted in any meaningful way for how they identify.

Instances by-and-large federate with most others, and I feel this is both in the spirit of the ActivityPub/federated web standards, and in the interest of a less commercialised/walled off internet. I see it as one large ecosystem, in which some islands do exist.

I am in both camps. I feel the Fediverse as a whole should be more cohesive and considered a single platform, but at the same time I feel like each instance has the ability to choose whether to be part of it or have it’s own focus.

People use social media for very different purposes, and sometimes they want those purposes to be entirely separate from each other. For example, instances focused on erotic or NSFW content are important, but not all other instances may want to federate with them, and that is OK.

This question was purposely vague to make people get a little creative, but it also is an example of a leading question, so the actual percentage is meaningless in this case. Almost everybody agreed that the fediverse is one large ecosystem, but the they all argued that instance being islands are the best part of that ecosystem.

The answers I got to this question were all positive of the fediverse and the federation – The central theme of these quotes are community, culture and safety.

Mastodon Developers: People used the word niche over and over again. I think this supports the need for greater discoverability across instances of these niches. (hashtags are a major driver of communities in Tumblr for example because they’re global)

Question 3: “What do you think the Mastodon project is doing well?”

Launching into the feel good section.

I’m just quite happy about the technical notes associated to releases, which both help understanding what changed and how to upgrade.

Its a good and very modern tool that evolve quickly, without using bad systems that exists on some GAFAM (big 5) tools

Integrations: it’s very easy to build integrations, the APIs are well documented, it’s relatively easy to build against the ActivityPub specifications, and that makes it easy to feed content into the fediverse from multiple sources. – Politeness: people are generally more open and polite here, and the climate much less toxic than Twitter and Facebook.

The mastodon developers do seem very keen on evaluating the effects of features they implement before actually implementing them.

The protocol that connects millions of people and hundreds of different ideas.

Making it easy to join, being the first to implement activitypub properly for a popular usecase, it’s relatively well documented, has a modern UI

Providing a well-regarded, high functioning platform for microblogging. Consistently delivering on stated goals for reducing abuse, toxicity, and generally fostering civil discourse.

I don’t really know how to answer this. It seems like the name branding of “Mastodon” as “fediverse” has gone extremely well, and it has captured a lot of mindshare. That’s good, right?

providing decently reliable software. Ensuring stability and performance. Many issues we encountered over the years have since resolved.

I am really happy with the overall maintenance, promotion, and improvements.

Software quality is generally very good, and I’m personally satisfied with the communications I’ve had with the developers.

On the whole, the userbase is more compassionate and more aware of social issues than other social networks. This should be encouraged and the tools made available to allow for granular blocking, limiting, etc of problem users/instances.

You manage to unite people, nations and their interests, as well as to get away from the control of large companies, and give users the choice of what they really want, not the company

The Mastodon UX for the web interface is above and beyond my expectations for a project like this and I wish all social media platforms were this clean and accessible.

The user experience is constantly improving. The base of developers is growing, and there is also external funding, which is a great achievement.

I think Mastodon is the best developed part of the Fediverse. There are a few things I would change though, but I doubt that Eugen & Co would realize those changes.

Building quality, reliable and secure software.

Growing. Changing the shape of social media, showing that it’s entirely possible to do this without centralisation.

Everybody had a lot of excellent “vibes” about the Mastodon project. The administrators all love their instances and this ActivityPub project in particular.

Mastodon Developers: I know Big G has mentioned plenty of times the stress of being one of the core developers, but I think it’s obvious that all of this hard work has resulted in something worth celebrating.

Question 4: What do you think the Mastodon project could improve?

This section was a list of general features…

Contributions and input from diverse members of the community.

A less weird full text ES implementation, like the stuff for Chinese support, I hate patching source code.

Moderation tools need significant improvement. Pleroma’s MRF allows admins to easily compose policy that Mastodon either does not allow or makes unnecessarily painful due to the way the admin UI works.

When I search for content I wasn’t to be more “open” to the world and other instances. Can you improve the content detection with other instances contents ?

put more emphasis on the local timeline. Mastodon is successful because it has small communities, not because it is a global twitter competitor.

I personally would prefer more frequent releases of Mastodon, even if they are only maintenance releases with a few updated dependencies.

Today Mastodon has very limited support for searching posts. Users who want to search for a particular hobby or culture across multiple instances will find it lacking.

onboarding of new users. It is quite hard to stay interested in mastodon, if you’re not giving it a TON of effort. You must want to make it work. I have tried helping many (also very tech-savvy) people onto mastodon, and most of them gave up after a few weeks. User Retention is the single most important thing they need to fix at the moment.

Spam registration controls. I really hate being listed as “Request an Invite” simply because my instance gets flooded with spam when I open the doors.

Functions like group DM could improve mastodon. It’s actually biggest obstacle to migrate all twitter based own charactor community.

UX/UI Improvements. / Rebuild the UX from scratch

More core developers – A public roadmap that’s adhered to – Better decision making processes – any decision making processes – Less focus on being like Twitter – Less focus on unimportant things like rebranding when there’s other, critical things that need to be addressed

Yes, Mastodon should enable hashtag follow and enable a web interface for old computers, like minimal browser compatibility thinking on non-smartphone devices and developing countries computers.

An edit button with history would be amazing, my users tend to find the delete and redraft option limiting and cumbersome. While we also understand why boosting doesn’t give a commentary option that’s also somewhat cumbersome in some interactions and i think there’s got to be a better middle ground. Also DMs REALLY need to not be present in the main stream. I like how they aren’t treated inherently differently people logically treat them and think about them VASTLY differently and the UX should reflect that they are, to users,very different. Lastly better tooling for admins to share, import, export block lists amongst each other

Better moderation and federation controls would be nice. It’s only a matter of time before the fediverse sees real, technically sophisticated, large scale abuse, and it is definitely not ready for it.

My suspicion is that we need better moderation tools and a set of people making steering decisions with a more diversified understanding of the consequences of technology decisions.

Multiple frontends: I know it is possible to use other frontends, but (as far as I know) the other frontend are different proyects. It would be great if we had serveral themes available, so users could choose one. The current “multiple columns” theme is good, but many people comming comming from twitter find it confusing and leave mastodon because of it

A governance project for the overall Mastodon project would be valuable to ensure that instance admins aren’t just joking/not joking about having no influence over the project, even if they have the capabilities to be core contributors.

External project integration, integrated UI templates for users (personalisation)

The fact Mastodon has no internal modularity (ability to write or add/remove plugins of features desired or undesired) is frankly a crime. Many of the feature focuses have degraded the experience of my userbase. We had to turn off all the interactive features like trending tags because of e-mail spam and endless new administrative tasks we don’t have time for.

because some ActivityPub-based software projects provide extensions, but then do not document them in a way that permits interoperability. I believe someone should take charge of better specifying these extensions. I do not think that Mastodon should act here alone, but I absolutely think it plays a leading role in bringing the development community together here.

Diversity of configuration, acknowledgement that smaller instances have different requirements to large ones.

Better translation support and multi-language support

Add markdown support,

I wish there were ways that an instance could programmatically declare information about its culture, so that other instances could use that information to determine how (or if) to federate with it. For example, a NSFW instance should be able to mark itself as such, and other instances should be able to federate with it in such a way that content warnings are automatically added to posts that do not have them

Mastodon Developers: Roadmap and more communication.

Question 5: “Do you believe instances admins have enough influence on Mastodon?”

70% answered No BUT this question was interesting because a lot of the administrators feel that this is a good thing – and having the project driven by the core team is fine, and big G is doing an excellent, but silo’ed job… and they repeat that forks exist if people want to make major changes which is a nice balance. There is just too much focus on the “big instances”, which is echo’ed lots of times.

At present it feels a lot like decisions are just Made and it’s up to either those running pre-release/bleeding edge installs to notice and complain or offer feedback.

No, and its good I think

Any instance admin could write any code they wanted to run on their server. The only limitation on the code they can run is the labor required to produce that code

Probably influences on his own instance. Not on governance of the main project

I don’t see why they should influence mastodon in the first place, so I think the current amount of influence is fine.

Not in the slightest. I’ve been made to feel on more than one occasion that – that my desires to make an instance that isn’t just a gateway to the fediverse is anathema to participation in the fediverse.

it’s pretty clear that mainline Mastodon is being developed for large, centralized, and well-funded instances, leaving smaller communities behind

Mastodon seems to have had issues ensuring two-way communication for years. There is not, nor has there ever been a valid reason for the hostility within the contributor space. Often admins and the greater community do influence design decisions, but have given up on being heard by the time they’re implemented.

I carry some local patches to change the software for my use, but feel like the overall project direction is driven by the largest instances’ needs.

I think so, but I don’t really manage my instance that much, I just let it run and I update it. For me it works,

no, I’ve been parts of too many “won’t happen here” discussions to believe it.

What influence? When we comment on feature we like or dislike, we get shut down (despite popular fork having those features)

I don’t personally tend to be too vocal around mastodon in dev channels, but I know other admins that are, and I can’t say a lot of attention is given to them. Many have legitimate concerns regarding their instance and the direction of development, and they should at least get a good listening to

Honestly, I don’t care that much about the governance of the Mastodon project as a whole. I want to focus on my community.

Multiple people have raised they felt shut down when they raised questions in GitHub/discord.

Mastodon Developers: I think the admin community is feeling a little lost. Same as above, no roadmap, Not enough engagement or community.

Question 6: “Would you support establishing a steering committee? (Building community roadmaps)”

60% Answered No. This is an interested one that has flowed in one direction – steering committees are fine as long as they don’t make demands or place additional oversight or even derailing the original project with demands. This is a similar answer to the others. Communication is key, and I think having a formal method for admin/users to create mocks and requests. Similar to the PEP style “enhancements”, but instead of at a protocol level, it’s features. No harm, if they’re not introduced and let’s people act in good faith.

Free and open source software projects are a do-ocracy. They’re run by the people who contribute labor. Many “better governed” forks have been tried in the past

A steering committee does not feel necessary but I think it would be nice to have community driven working groups focused on building feature requests.

It’s quite evident to me that the Benevolent Dictator For Life gets a lot of people pissed and it’s detrimental to transparency and participation, while a purely community-driven approach gets to a point where nothing is delivered because everyone feels entitled to feel offended by everybody else’s decision. A steering committee (made by instance and relay admins, project contributors and community representatives on rotation) could definitely be a reasonable trade-off between these two extremes.

Telling open source devs what they should do with their projects and (for the most part) their spare time will certainly be a recipe for disaster.

I would support it once the project lead asserts the need for one, but I am signed on to a vision that does not currently want it, so I’m not pushing for it. To force this issue is to impede or alter the experiment. mastodon will grow, or not. It will be replaced by something, or not. It might be the best answer for the next ten years of civil discourse, or not. Until I’m ready to stand up and take on a whole lot more responsibility I will happily consume what Eugen et al are putting out and be guided by their goals and intent.

Given the amount of traction Mastodon has gotten, and the lack of a steering committee or any sort of board of governance is, frankly, quite appalling. Other major FLOSS projects, even ones with dictators-for-life, have these structures in place because otherwise the load on the founder is immense and will directly lead to burnout.

now that Mastodon is largely feature-complete and stable, having the guidance and creativity of a committee to keep things moving forward is essential

As the project grows it’s governance is also need to grow. I get mastodon’s reluctance here though

Any steering committee should maintain their own branch of Mastodon that reflects their preferred version of the code too. This helps to guarantee that any suggestions that are expensive technically have some implementation.

The ones who best understand the needs of a platform are its users. And the ones who best understand the needs of the instances are the instance admins. I dont think it is a good idea to take over the project, as the people who have been working on it for years have created this as “their child”/”their project”. They habe created something amazing, there is no need to take it away from them. So I think having a group of people suggesting ideas and talking about their needs is good, as long as it is done in a peaceful way.

I’d be hesitant about any formal organisation given the joy of Mastodon comes from its ad-hoc nature, but some level of community involvement would be good. What Mastodon project members want/imagine may not line up with what instance admins want, which again may not line up with what the users of Mastodon instances want.

Another tough one. My experience with committees are that there is a lot of discussion, but little actual useful output. Happy to be pleasantly surprised though.

If not a steering committee then at least an admin advocacy group

they already rejected it, though: https://github.com/mastodon/mastodon/issues/17107#issuecomment-1027200402

feel like a group of people who can aggregate feedback and present it to the developers in a single, unified message is a good idea in most cases, but I feel like that goes against the spirit of what Mastodon is and instead giving everybody their own voice is the right path. Maybe a group acting as a proxy for instance admins who don’t want to be involved or don’t have the time or experience to participate would be a better option.

Too much process kills motivation. I’m happy for the Mastodon developers to continue to build Mastodon as they have been doing; I’m not complaining about receiving their work for free!

Mastodon Developers: I think the short answer to this is, I think administrators would like a formal method to request features, and if that’s possible, then some kind of draft/submission/response flow could be considered. If possible?

Question 7: “There is no public roadmap for Mastodon, are you aware of the planned “Groups” functionality?”

85% said no. This is a key example of communication failing. Nobody knows what the future holds, nor what the outcome or plan is for Mastodon. Apparently it was mentioned in a blog post, but I couldn’t find that post?

Not aware, but I hope that it is compatible to other software like Friendica that already offer groups or forums.

They should ideally be cross-instance, and basically behave like subsets of instances – with their own moderation chain, rules, option to choose public/private/restricted visibility and option to choose their registration policy.

Groups should behave like a private room of toots. Akin to a private messaging group on Whatsapp. With the option to make them public, allowing anyone to follow them, or keep them private, making the tweets in the group only visible to those who have been approved. It would also be nice if they could alternatively function as a shared news outlet for communities or similar. Allowing one handle for a changing rotation of community leaders without requiring the sharing of account passwords.

It seems that Mastodon has been planning this feature for a while. And for me, I think ‘Group’ should do sth that gup.pe had been doing, to allow people to direct toots to a group of people who share the same interests as his/hers.

Groups, to be useful, should be able to be marked as instance-only or fediverse-wide – otherwise I imagine we will have knitting@one.domain, knitting@anotherdomain, knitting@many domains, which I don’t believe is helpful (because we will then immediately need “groups of groups…”. Groups would be a good way to surface content without having to allow whole-text search, and can serve to help people find the folks who are not known to their newly-onboarded instance account. What I don’t’ want is two public warring knitting groups based on cliques. If you need a private group, do it on your own instance, otherwise join in the fediverse and enjoy the chaos.

Like twitter’s group dm function. (or circle function that implemented recently)

I don’t know how it should behave. I have a great many concerns about how it’s going to behave, since there’s no published documentation or roadmap or design concepts on how it could work, and no ability for instance admins such as myself to provide feedback. For instance, how is “groups” going to interact with defederations? How will group names be policed for hate speech? There’s a lot of unknowns, and the lack of a steering committee or any sort of systemic oversight makes the entire feature both nebulous and troubling.

Like any other group, only interacting with selected contacts and no leaking conversation to outside.

I’ve heard word of groups. I have no idea how they should work, the concept seems like it wouldn’t fit well in the Mastodon UI design without major changes, which is bound to be unpopular. Mastodon’s UI has never been it’s strongest point

I was unaware of this functionality, but feel it might be good for users – intuitively I’d expect it to work similarly to twitter circles, which is nice.

Mastodon Developers: Nobody seems to know what groups are, or what the plan is. I am one of them. Nobody having the same idea of what groups SHOULD be is fascinating.

Question 8: “Does this paragraph match your expectations of how search should work in Mastodon?”

65% said no… the funny thing is. Most of the people who marked yes, gave no comment. most of the people who marked no offered indepth responses.

yes, Search should work to minimize the harms caused by harassment and abuse.

No, I think this is the role of the moderation to protect contents

I didn’t know it wouldn’t look for strings in particular. That makes search less useful to me as I’m often looking for specific mentions of a word. Would be nice to be able to configure this at an instance level.

As I mentioned earlier, search features are now basically absent. A user should AT LEAST be able to search in their own toots, favourites, bookmarks, mentions, and even profiles that they follow. I’m appalled that the community is even still discussing such a feature: if I’m searching for something that I have already written, saved, followed or been notified about, there’s literally ZERO risk of abuse. A more complete hashtag search (ideally across the whole federated timeline) should also be implemented: if I don’t want my content to be searchable, then I just avoid using a hashtag, or I don’t publish my toot as public. By using a hashtag on a public toot, I imply that I want my content to be indexed and searchable by anyone. So not providing a proper hashtag search defeats the whole purpose of having hashtags.

I do not think people should freely search tweets like that. If someone is determined they probably could, but I’d rather not hand tools like that to a malicious actor.

I don’t have a preference over it. As other search engines may act as a internal search engine, and different posts are scattered across the fedivers, it is not easy to implement a comprehensive search engine

I would like for search to be comprehensive (i.e. the whole database). I am not a fan of intentionally limiting features just because “they could be abused”.

Users who want to search for a particular hobby or culture across multiple instances will find it lacking.

Art, comedy, fandoms and other internet cultures need to searchable.

yes, because I am an admin and have pondered whether to activate Full Text Search. As a user, this kind of fundamental difference in how instances might work (one instance might support it, another one might not) is Very puzzling. It is not obvious to a regular user why this would be this way. I think full text search should be always active, AND I think it should include other people’s posts. The dogpiling could be mitigated with other things (mainly: who is allowed to reply; maybe an individual setting if people want to be on public searches … etc. pp.)

What happens when somebody writes a server that just collects all the toots and makes them searchable? Then this feature will be rendered useless. I do think however that the option should be given to the user: should be toots be full-text searchable. The default should be “no” and it should be easy to change.

no, I am sympathetic to the motivation of restricting searching, but I also agree that it’s not the way one would expect it to work initially.

yes, It meets my expectations because that’s what Eugen said it will do. Until Mastodon gets so big he has to cede control or let it die, I’m fine with it as is. It’s simply not my call. If I disagreed I can make the “better”: thing and see if people jump ship. That said- this, and the need to use hashtags to be discoverable, could be elevated for users in the UI so they know to make this choice. Saying it once doesn’t help. The whole notion that one can choose to publish sort-of-quietly is subtle genius in my opinion. Some people want a small soapbox, not a stadium speaker system. Avoiding hashtags allows folks to publish to a narrower audience.

no, I expected search to be federated across instances, useful for a solo instance user to find other people by topic.

no, I’ve deployed ES for my instance, and honestly it’s usually just expensive and useless. It doesn’t let me search for things I’m interested in finding out, it doesn’t let me track anything, it’s just there chewing up a server.

yes, omitting un-related posts is the one thing saving us from the hell that results from mentioning particular words on Twitter

no, I get a why this choice is made but it makes trying to search for things dramatically harder and is particularly cumbersome for anyone who needs search yo be able to find things. I have no good answers here but this particular choices causes pain either way, and one i think a sufficiently motivated actor can get around anyway.

yes, Yes or No isn’t enough nuance for me 😀 — I understand that’s how search should work, but I also feel it’s extremely counterintuitive for users. Honestly no search at all would make more sense (and save a few Gb here and there).

no, I appreciate this design decision, but I would prefer it to be optional.

no, I thought it searched in the emitre database. I domt think it is useful not not be able to search things that are public anyway. I expect to not be able to search for deleted messages/toots. But for public existing toots that can be found by google, I expect to be able to search those with the built in search field.

yes, This is how search should work. If it were enabled further, it should be based on a per-user opt-in choice. Mastodon users value their privacy and we would be wise to heed that if we want to keep our userbase.

No, I understand wanting to exclude results, but this should be a flag provided to authors: exclude this toot/all toots from searches

Yes, I both like and dislike how search works on Mastodon. I see the benefit in being unable to namesearch but I’ve also heard a lot of complaints about how the behaviour of search is non-obvious. Note however that my instance does not offer full-text search due to capacity constraints, so my experience of search is perhaps not adequate.

yes, I welcome a search feature. Elastic search is super heavy on servers. Being able to find your own toot should be a minimum. being able to opt out of other people search should be a opt out by default that people can change. It should also be combined with what toot can be seen, eg. only show public toots, or toots with hashtags etc.

yes, Being a long term user, I’ve come to terms with how search works here. However, it isn’t necessarily how a lot of people think it works. They generally expect searching to search the fediverse much like Google searches the www. I don’t think I’d change it to a FULL search, if given the chance, for the reasons stated in the current description, but also from a server/network load perspective.

No, I’ve done quite some research into the search function. It’s one of the big UX weaknesses of Mastodon in the current state, in my opinion. Good search would help users discover other users with certain interest (or at least with certain content in their posts), to then follow them. User discovery is the even more central UX weakness of Mastodon (and the Fediverse in general). I would expect the search function to at least cover the instance’s entire index.

Yes, Again no strong opinion on this. I don’t frequently use search feature in Mastodon. The image text shows a drawback of doing full-text search, but on the other hand sometimes indeed to hope to search certain text string. Only again that it is not a must-have for me.

No, Everything had abuse potential, from search to quoting to simply tooting. The decisions what is implemented as how seems random and bizarre. The end result is a social network where search is broken, embedding is broken and you can’t see the replies. I reiterate – different communities have different needs and things need to be configurable. Also – don’t make me install Java – the dependency tree of Mastodon is ludicrous as it is.

No, I can understand why the search was created this way, but it also makes it difficult to find good posts.

no, I would expect public content to be searchable. If users are concerned about their content showing up in a search, maybe they should use a different privacy setting? Search is not just a vector for abuse, it’s also a way to discover other people with similar interests!

no, Search is for discovering people and posts you’ve NOT already encountered. So this completely defeats that discoverability aim.

no, I want full search to be available. Controversial words can be banned manually by administrators if needed.

Mastodon Developers: This one is pretty clear. Search needs to be reconsidered.

1) Consensus is that search should be for full text, as global/universal as possible. 2) Use of hashtags are considered opt-in for search by default.
3) Maintaining user faith and trust is ultimately important.. so people need to be able to tag their posts/profile searchable or not (I touch on this in later questions).

Question 9: What do you feel about the act of moderation? Do you feel the tools provided (as of July 2022) are adequate? What could be done to improve moderation?

Tools are adequate, but please make it possible to remove the default activated “mail user about ban”, I don’t want to let spambots know i block them.

LOL

Didn’t have to do any kind of moderation other than blocking full nazi instances, so all good for me

Moderation is fine. The trending tools could use a bit of an upgrade to make them less … intensive to maintain for small instance owners.

No. The admin UI is still too simple, e.g. the long-waited group action feature has been added since the 3.5.0 release, but we cannot customize the option(for example, setting custom warning when suspending accounts.).

It’s fairly OK.

Moderation is slippery and tools for it could always use improvement. The developers are working in the right direction by providing appeals. I hope we see more user and moderator empowerment in future releases.

The ability to mark all posts from an instance as sensitive or unlisted would be greatly appreciated.

The lack of group management of instance suspensions (defederation) has been a huge issue for me. It means I have to pay attention to FediBlock and quite often look at hate speech and objectionable content in order to make judgements on if I need to institute an instance block. There’s no way for me to go through the many instance blocks I have to see which ones are still up, which ones are still awful, and which ones may have actually resolved their issues and can be unsuspended. Basic user-level moderation has been fine. As I run a small instance, there’s very few reports.

In the beginning, I was struggling to understand the different options for moderating single posts and contributors. I haven’t stopped struggling to understand, but having to moderate people and their post has vanished as an issue. In addition, trend moderation is often working well, but sometimes getting information to decide whether some hashtag may trend requires avoidable extra clicks. However, it’s FOSS, thus I’m fine with that extra fuss from time to time (like twice a day)

Moderating is frustrating, and adding granularity makes things worse. Explanations are poor… for example, I would like to be able to silence a local user while deciding whether their account should be suspended and have the silence also prevent their posts from federating outward, such as when a remote user has filed a report against them. Does it? I have no idea. Most instances I know are small enough that there’s usually only a couple mods, who often need time to agree on things like suspensions, and who rely heavily on mass blocking but do most local moderation through DM or group DM. I can see some attempts to honor these practices, but the moderation system seems designed for a huge platform…aren’t Most instances actually quite small? The system tries to streamline in strange ways, sometimes. I just want to silence or suspend people and decide whether or not to notify them, and to maintain a large instance blocklist I can see most entries for on one page.

Moderation is critically important; I actively moderate, and control federation. The tools are fairly good for the current state of the fediverse’s use, but they are not ready for the next stages of larger scale, more sophisticated abuse.

I believe that the tools have grown tremendously in a positive way in the past few releases of Mastodon. While they are not perfect, no system is. The main thing that I would want is a way to let users know that a report that they sent has been acted on, even if the outcome is not listed.

Moderation is a challenge and the tools provided are far from adequate. There’s no way to communicate back to the submitter of a report anonymously when a report is submitted anonymously, for example. There’s no link to an instance’s administration or moderation staff, when that list is trivial to pull through an API. Multiple reports about a single user must be closed independently. There are now email notifications sent to users when moderation actions are taken, but we can’t see what those messages are or what triggers them. The whole thing is a mess and the lack of documentation isn’t helping.

Reasonably okay. Shared block-lists one can subscribe to akin to pi.hole block-lists might help.

Moderating on Mastodon is not enjoyable, and I feel lucky that I have relatively few reports to deal with on a weekly basis. I have no concrete ideas about what could improve the moderation experience, but something I would like is to know if the remote administrators have done anything about a report a local user has made about a federated post. Currently, the only information is “has the user allowed this report to be passed to the federated instance’s admins” and “if you go to the reported person’s profile, have they been banned/deleted the toots/etc”. Some bidirectional mechanism for federated moderation would be nice.

Only one issue I’ve come across so far. By default, reports are sent to the reporting user’s instance admin AND as far as I can tell any actions by that admin are also forwarded on to the instance admin of the reported user. Where the problem user is a single user instance (or the admin on their own instance), it may sometimes be beneficial to action blocks/limits locally on the reporting users instance without alerting the reported user’s instance (to avoid stupid retaliations). This has only really popped up once, so it is a bit of an edge case, but the option to not forward actions on, might be useful.

I think that moderation works quite well (if the administration of the instance well structured everything)

Yes, they are adequate for now; but I feel they would not be very effective against a motivated attacker with technical understanding of ActivityPub

It’s fairly ok. I haven’t had serious problems with the current tools. A minor nitpick I see: If anyone reports an account, it’s hard to establish direct communication to the reporter in order to clarify anything or give feedback regarding what has been decided as a result of the report.

Don’t make me manually login every hour 24 hours a day to manually approve hashtags, toots and links on a 100 active user site would be a start. And the explore tab is still mostly empty. As for site/user moderation the ui/ux is terrible and i18n is deeply broken, but luckily we don’t need it much, so it’s mostly workable.

The moderation is so far quite good, but the federation should be improved, there is currently no way to respond to messages that come from other instances.

Federated moderation is missing. Mods and admins from different instances should be able to work together to e.g. create a common blocklist. I was involved with Fediblock and got harassed for it all the time. When people on “the good side” started picking on me it was time for me to stop.

Moderation tools are pretty good.

Mastodon Developers: I think this hits on the common request for federated moderation. Communication between instances and administrators. I think having edit functionality would enable moderators to edit/flag to add content warnings or mark toots as sensitive if they will. Additional options than delete or not.

Question 10: Do you have a problem with spam on your instance?

85% said No.. but then almost all of them said they were too small or forced manual approval in response to increases in spam.

I get like backlink profiles and spamtooters which I delete. I added 2 moderators after some years of soloing which makes it easier to get rid of the spam.

A little, a captcha can help I think

Approval mode and blocking Tor exit nodes do the magic to avoid spam

We block spammers. For those who are not obvious, we DM them to see if we get an intelligible response. If not, they get banned. We don’t spend much time dealing with this anymore since we restricted sign ups to approved accounts.

Spam registrations are a problem and we alternate between being open and friendly to newcomers, or restrictively approving every application. This is a shame as it slows down onboarding and we of course get drop off that never returns. I’d prefer to see new members offered some kind of quiet mode where they can build their profile, their #introduction post, etc., while waiting for moderator approval to go loud. Generally I take two 30 minute periods per day as a baseline to admin/moderate the instance, plus whatever else extra pops up from time to time. The mod team are less active and likely for them it’s a total of one hour per week or so.

I check account list every day, and delete account when they started to send spam toots.

we changed the “why do you want to join” text on signups to “tell us something interesting about yourself” and you bet your bippy that new accounts from spammers stick out

I was having lots of spam in the beginning, but it’s mostly gone by now. I used to setup certain regexps to prevent escort spam from the US or shops offering smoking stuff there etc. But when local user base was growing recently, those spam posts mostly vanished

I have a list of malicious IP and y manually block them. Only allow accounts that says something about the reasons to join. There should be the ability to customize the message in the “reasons for joining” so emphasize that it is an absolutely must to fill that blank.

No because I turned off open signups. Before then: very much. If I wanted to run a big instance this might be a problem, but I want to run a small one so it’s okay.

I very reluctantly require approval for new signups, and reject about 75% of them for being obvious spam. Captcha support (ideally including a captchaless WebAuthn privacy-pass sort of thing) and IP reputation tools would be really helpful here.

BOTS!! We closed off our sign up to approvial only to deal with the number of bots.

Only because I closed registrations. I have about 20 spam account creations for every genuine account that comes through. Seems any instance that doesn’t close registrations ends up with a spammer issue.

In the past, spam was dealt with by modifying the signup process code to add CAPTCHA functionality to prevent automated signups, but this led to accessibility issues for some users, and some spammy signups still got through. Currently, signups are simply limited to requiring approval, and I’m considering moving to an invite-only basis – allowing existing users to invite new users.

I dont have a spam problem since I block early and block often, users or instances. The instance is not open to registration because that can only lead to spam. The capabilities around that are also abysimal.

I have approvals turned on for signups, thus reducing instances of spambots greatly.

the spammers I get are pretty bad at spamming, so it only takes a few minutes a week

Since unlisting from “join mastodon” site, very few spam accounts have signed up.

Problem might be an overstatement, but I do get spam every now and then.

There had been some spam waves in the past. Currently there is none.

We have had a problem in the past, so we have changed to manual approval of new accounts.

Mastodon Developers: As a surprise to none of us, Spam is a problem, but most of the instance admins turned to the only thing under their control – switching to manual approval.. or taking your site off “join mastodon”… both options I refuse to do because I want to maintain growth.

Question 11: Do you have a problem with abuse on your instance, or the fediverse in general?

88% said No, and mostly left no comment. Those who said yes had a lot to say. I think this is similar to real life, targeted people get seriously abused and the rest don’t.

Yes, The fediverse is full of abusive people, and blocking doesn’t work all that well with dogpiles.

No, I spend at least 1 h cleaning up zombie accounts.

yes, If by abuse you mean harassment, we have always had to deal with it: both from people on our instance and those outside of it. Most of the time we talk to the people directly first, and if that doesn’t resolve it, we don’t hesitate to block them. Abuse shouldn’t be tolerated, but sometimes people genuinely don’t understand what they are doing. We don’t spend much time dealing with this, thankfully. Most of the people we know are nice.

no, I assume you mean abusive behaviour from one user to another? >how do you feel about abuse in the fediverse? I feel, in general, the underlying code and its presentation have done a great job of vastly diminishing the opportunity for abuse, and for fostering a community that is quick to recognise and excise it. Nothing is perfect, and any ecosystem is going to have a spectrum of ideology, but the larger Mastodon universe tends to be one of the more generally-rowing-in-the-same-direction efforts I’ve seen in quite some time.

yes, Nearly all of our moderation involves investigating and suspending remote users or remote instances, and we spend several hours a week reading threads and often communicating with other admins.

no, Not many. Only a handful of them.

No, We haven’t seen much, but we are intentionally a very small targetted instance. Our users do fall into vulnerable groups but being on a smaller instance seems to limit our exposure there

No, I spend a little time on it and have other moderators too, but I find that what counts as abuse on fedi is often “I don’t like this person so I am reporting them”, mixed in with a small amount of genuine rule-breaking.

Yes, I follow #fediblock and review peoples’ recommendations, making my own decision, usually agreeing with #fediblock contributors and placing instance blocks about 75% of the time. There is definitely a “mirror fediverse” of abusive instances, and they largely seem to be one interconnected cultural group, and then just a few small individual outliers. It would be nice to be able to analyze the social graph more fully in order to make defederation decisions more efficiently.

No, I’ve been happy that I fly under the radar.

yes, Our instance has not been a problem, but abuse and harassment is rampant across the fediverse. We’ve suspended federation with over 200 instances for harassment and bigotry, sexism, and outright racism, white supremacy and nazi imagery. All this content is easy to find if you go looking for it on an unmoderated instance. Now that the right wing has adopted the fediverse, that problem has gotten substantially worse.

No, I have luckily not had much experience with abuse on my instance, nor on the fediverse. I am, however, aware of the negative experiences that others have had, and I find it a daunting prospect to deal with.

yes, There is plenty of abuse in the fediverse and all of it is fixable. We just need tools to be able to trust instances dynamically, all the while we can share and receive feedback from other instances so we dont have to jump everytime a nazi create a new instance

No, We have a low bar for blocking problematic insurance, so we get very little abuse

yes, a lot. it’s very easy for chuds to start a new instance and harass people, and always requires user time and moderator time

No, Most of what is reported isn’t actual abuse, just very different opinions from what is locally acceptable. I block in some cases, not in others, but I feel the main problems can’t be technically solved, at least not with the current paradigm.

Mastodon Developers: ?

Question 12: As a Mastodon end-user, how do you rate the UX?

This question didn’t get a lot of responses.

Mobile app need features like all instances tab

Mobile app is missing alot of features. Biggest problem was as of musk told that he buys twitter, that alot of people installed the app and didn’t see the federated timeline. And give up because, „there are no users“ and „i am alone“

Needs draft toots.

Lack of good UX for the existing DMs feature continues to be a constant irritation, not just for me but my users. The “oh but future e2ee DMs!” is, well, not here yet, and forcing me and my users to deal with poor implementations now and for an indeterminate (no roadmap, after all) future is frustrating at best. While the official Mobile app on iOS is the best Mastodon app around, it’s still pretty sub-par. Lots of slowdowns, lots of quirks, LACK OF A DMS TAB

DMs really need to be treated, in the ux, as more separate than they are to keep them logically distinct and prevent confusion from the main stream

Just the other day I wanted to create different timelines with groups of users and I couldn’t

I didn’t even know a “Advanced View” existed.

I think the UI is ok, I think the UX is not as good as it can be. Eg. the data is there, we just cannot access it or interact with it in a meaningful way

I think the Standard view and the official app, both miss the point of local instances somewhat. In that the Local Timeline is very important and shouldn’t be hidden away in a way that is non-obvious to users.

I think the lack of the local timeline was a bad decision. (App)

A lot of improvement has to be done

We need a whole new Front end. Shame sandbox is written by a creep.

Mastodon Developers: More general vibes. People like the federated/local timelines. People want real DMs. Drafts.

Question 13: Do you or your users experience problems with the accessibility of the UX?

This question got two answers.

In terms of web content accessibility? no-one’s complaining, but having worked in WCAG/Section 508 etc I’m sure there’s a never-ending list of accessibility issues, especially with the software still evolving quite quickly. My experience is to expect accessibility, like most things, to be forever a little bit behind, I’d be more concerned about the project’s approach to rectifying major accessibility concerns than the pick-a-day-and-point-to-whats-wrong deal. Without a larger dev team, I find it unlikely the increasingly-complex software can maintain high accessibility scores for very long, but as I said, no-one’s complained directly to me.

I have and do, especially around the difficulty of telling if a notification is a boost or a favorite in the UI, an utter lack of visual indicator if something is @-only privacy (a DM), etc.

Mastodon Developers: I can’t get much information from the limited responses. As I’ve personally said many times, being colour-blind makes my experience on mastodon pretty bad, and I’ve applied a bunch of patches to help me on my instance…

It’s interesting, but I’d like an auto accept on trending stuff.

Yes, but need more content in there. I think “big” instance have sufficient content, but little instance lake of contents

our instance is too small and doesn’t see much message traffic, so the usefulness of this tab is quite limited imo

There’s literally nothing showing up in it, unless I’m on a large instance – see my remarks about discoverability.

Yes there is already value there, but it is tiny. As mentioned earlier, I think user retention is the single most important focus right now. Explore (or similar tools) might play a big role in that. I have no idea how to fix this, though. Personally, all I can think of is curated “follow friday” style posts from our admin user.

Everything except the “news” section (or whatever its called where the news articles show up) is fine. It’It’d be nice to see the actual toots discussing the articles in addition to the articles themselves. Otherwise it’s just a poor man’s RSS feed.

I hate the trending functionality, honestly. I have to deal with numerous emails a day to vet and approve things, and it pushes the entire system into the gamified Twitter-esque attention suck that I really did not want it to be.

IMHO still a bit too complex to grasp. But not too bad.

We disabled all we could after numerous user complaints. Our community mostly left Twitter and Facebook intentionally and resist items which remind them of these features… they want to meet and converse with folks by way of introductions and mutual connections. It would be nice for admins to be able to feature a few such links in blogroll style, though, for when our community is holding a discussion.

It’s OK but the News and For You functionality has not been useful.

Manual approval is killing me.

I love this idea in theory but it’s not very compelling in its current state and I’m not entirely sure why. Maybe if our pinned toots showed up below profiles so you get more context about users?

That’s how I find content.

yes, I have to moderate most of it manually and to often.

This is bullshit nobody wants. A better UX around pining hashtag would have been useful.

yes, it makes it easy to attract attention, but the features are still raw

yes, I find value in it, but it is very limited in my experience.

yes, I do see some value in the explore UI. It seems targeted for non authenticated visitors who want to find out who’s active on an instance. It may help discover some users to follow (which I said is a weakness of the Fediverse in general). Since the explore UI is practically not accessible for the authenticated user, I am not using it often.

yes, Make it work automatically without approving every item be a full time job. Give me an interface to change thresholds so it’s not empty most of the time.

It’s a fun gimmick sometimes, but I have not seen much value in it so far.

Yes it’s great, but I’d enable the “allow xx to trend without prior review” for posts and links too.

Mastodon Developers: While most admins responded no, the major gripe seems to be the manual approval. I’ve raised before that all trending posts should auto approve and I’ll spend my moderation time manually removing trash that’s reported to me.

also, allow us to import/export trending approvals/block lists (I’ll share my news domains/hashtags lists for new admins to import) and save them time.

55 no/45 yes.. Welcome to the beginning of the weird questions. Most of these are supportive of the concept, but fear the complexity. My aim with this question was to try and question the idea of adding metadata and control to posts. One of the responses to the search question was “Some people want a small soapbox, not a stadium speaker system.” and I think giving people the option to set details per toot is a good one to discuss.

Basically neat idea, but in the end even more work that maybe needs to be done before posting. Depends on the implementation maybe. Not strictly against it, but for now a no.

This would be great for Internet culture on the whole.

This would be weird.

sounds like a good idea for creators, but in general could be confusing to the majority of users who wouldn’t need it, likely? Could be a good way to promote open licenses. I’m mixed 🙂

I have neither the time nor inclination to police that sort of nonsense for every post.

I don’t mind it existing but I doubt it would be used constructively and on a sufficiently large scale.

If we need to license each microblog post, it’s ridiculous. We should be able to set the referenceability of posts on a per-account basis.

No, This thing needs to be simpler and certainly not more complicated! Add more settings like that at your own peril.

You’re in the public domain, if you don’t want stuff to be copied, do it in private. Posting stuff that’s already got a CC license should abide by it and attribute the person or whatever.

That would be an interesting addition, but it’d have to be done in a way that doesn’t add dozens of options to the toot composition modal.

No, Giving users control and autonomy over their own privacy and promotion of their content. Content creators can already choose to publish their content wherever they want. If they choose Mastodon – as a platform mostly engaged in the freely distributing posts business – I don’t see how this would be helpful I don’t believe, and I believe few to none of my domain instance members believe, that Mastodon is not a content creation platform. It’s a microblogging platform. Yes, all words are content, but there’s person on the street content (mastodon) and content content (DeviantArt). I believe the express intent of Mastodon is freedom to deliver those posts server-to-server. I would actively not support the above, it’s already confusing enough for most lay users. The UX needs to be simpler, not more complex, and there’s plenty of room for a content-focussed implementation to be created. If the world needs a platform where content creators need the above licensing metatadata, I imagine someone should make one, even an ActivityPub-powered one, but I don’t think it’s reasonable to hack in every possible use case to the one ActivityPub platform that’s already doing so well, if anything it should inspire the creation of the Contentadon or whatever that can allow the described functionality while federating some lesser form of the content or something. This seems more a question for https://www.w3.org/TR/activitypub/ than any particular implementation of ActivtyPub.

the ux for this feature would be complicated but if that can be managed I’d be in favor

Oh yeah that’s neat, I like that. I’d want some design work and thought put into how to avoid it being weaponised by harassment, obviously, but I really like that.

Yes! always a fan of more posting options.

The technology instances have consistently had a problem with art theft and are often hostile in their beliefs that anything available on the internet must be free for them to use. Any hint of something smacking of creative commons is usually misinterpreted as public domain, and I think this could even become an abuse and harassment issue. I would, however, like to see boost disabling and trending disabling, for other reasons.

I find that idea interesting, the problem is changing that metdata after the fact as those attributes likely shouldn’t be immutable

No, I’m not against it, but I wouldn’t use it.

No, I don’t personally like the idea as presented above. The main reason is that I feel that with less ownership of the instances (for general users), the harder it would be to present and enforce anything like this. I can’t think of a good UX for presenting the amount of control that people want over their toots without some interface for it. As an example, people put rules for interactions in their bio and pinned toots, but many people do not read those at all before interacting with a toot that shows up on their timeline.

No, Honestly I can go either way with this but in the global scheme I don’t think it will do much.

No, Who is liable for enforcing these licenses? Individual users? Who asked for this feature and why is it necessary? I’ve never heard anyone mention a need for licensing info on their social media posts, but I’ve heard lots of people complaining about the lack of polish on the moderation tools, or how we can only customize our home pages with CSS hacks. We still can’t delete Emoji group names and that feature was released 3 years ago. Come on.

No, While I see the value in having such information attached to posts, part of the draw in microblogging is being able to fire off a post without much thought, and I personally would feel stressed about tagging all my posts with what i’d like to have happen to them.

Yes, Funny how AP is full of meta data, meta data optimised to be searched and indexed yet this is a question that needs to be asked.

yes, Maybe. If implemented, it’d need some decent in-UI explanation of what things do. There should also be instance-wide defaults easily set by the instance admin. Users may change their personal metadata but the admin should be able to define the default for the instance (form the admin web ui, not in some esoteric config file somewhere).

no, it would be too hard to control in the fediverse, plus I wouldn’t consider it the right place to post this kind of content

No, I do not like this idea of putting restrictions on public toots. Users should have more control over their toots visibility and limit who can see it rather than rely on a licensing type.

No, License, discoverability and searchability to me are very different problems and need separate solutions,

No, Adds too much complexity. Either this is a full CMS or it’s a microblogging platform. Pick a lane. This is barely applicable to short form content, most of which would never qualify for protection under any copyright laws anyway.

No, Toots are short, so there is no need to include licensing info, maybe for attachments.

Yes, I like this idea, but it should be developed as a common standard outside Mastodon, probably in conjunction with creative commons.

No, I don’t really see the benefit.

No, Oh my gosh, this sounds awful. The interface for composing toots on the web is already complicated; most people don’t even know or understand what licensing is. Asking them to pick a license just to say “hello world” will make no one want to use it. I would only support this if it were hidden by default, and power users could reveal this feature if they want.

Mastodon Developers: This is a food for thought question. I think it can be done cleanly in an opt-in kind of way… at a basic level, this is an more extreme version of the “replyControl” FEP.

Question 16: As a Mastodon end-user, What front-end features would you like to see added (and prioritized) to Mastodon’s roadmap?

Groups (10 times)

Group DMs (10 more times)

getting extended notifications, e.g. when someone favorites/retoots a toot in which i’m mentioned.

post length should be longer, markdown should be supported

Better search and discoverability in general – More granular instance discoverability and toot discoverability/license settings (can definitely help create a better search where content that authors want to be searchable can actually be found) – Better support for notifications on the apps, ideally with UnifiedPush as a guiding priority – Markdown support (PLEASE!!), and ideally LaTeX, at least as an optional officially supported package – Support for quotes, and ability to boost toots with attached comments.

Better comment threading implementation.

emoji reaction. #4106

in the advanced view, having the possibility to have a column that EXCLUDES users or toots from lists. As in, I create a list with bots I follow, I should be able to create a new column and exclude that list from the column. And as mentioned earlier, the ability to bookmark toots with a single click, instead of of clicking on the burger menu of the toot card to then select “Bookmark”.

Drafts

Fresher UI look and feel

a DMs UX rework, so that they stop being awful, since e2ee is indeterminately far away – cross-device notification clearing

A instance-only chat window or shoutbox would be a god tier feature for our instance (and likely many other small ones).

Enabling a text-based interface, lower the browser requirements for mobile versions, maybe adding a chat like pleroma, adding groups and hashtag follow.

hahtags in private mode, answering toots to a public toot must always be published as public too, black theme, a new interface like pinafore

DMs removed from the main feed and only shown in the DM category. A wish list is an edit w/history (i.e. original toots are not removed the way delete and redraft works), a boost with commentary would be nice for context about a subject.

Ability to use Markdown. Ability to use inline images (using Markdown). Ability to add spoilers so we don’t have to abuse CW.

client functionality was unbundled from the server, like — a lot of my users use Pinafore, so the web UI isn’t their primary interface.

List of integrated UI templates each user can. Choose from.

More text allowed under content warnings, better emoji selection and management features, (or at least an emoji selector that doesn’t break half the time you use it), more customization options for client view (column width on multi-column? font and emoji size, etc. How about adding a text-to-speech engine as an option? That’s both accessible and fun!

I would personally appreciate more control over the appearance of the web interface, including mindfulness controls. User-customisable theming would be nice, as well as the ability to turn off numbers.

boost with comment #14597

Multi-language support in tweets

Make everything accessible from the front end be available through API so others can develop their own front ends.

Some users would like to see timed posts. But I personally do not miss anything.

better ui for liking/sharing posts from other instances, I don’t like typing my username all the time nor having a popup open…

Mastodon Developers: ?

Question 17 As a Mastodon Administrator, What back-end features would you like to see added (and prioritized) to Mastodon’s roadmap?

in-place upgrade via GUI or by a simple script that takes care of the upgrade – multiple auth backends like internal and LDAP at the same time, so registration can be open but users from additional LDAP backend are also able to login (not easy, though)

A captcha solution on registration.

I want a full admin API that lets me do everything the admin UI can do. I want to manage instance policy with Terraform for transparency.

Easy way to display an official “support” link (libera pay, kofi, patreon, paypal, …) after someone creates a new account on an instance, and on the instance public page + admin profile. Bonus: display a progress bar with financial monthly or annual goal (for example, running my instance cost me 128.45€ / year)

Better relays – more stable, better maintained, easier to search and apply to for instance admins, and with more sense of curation rather than randomly aggregating instances – Cross-instance groups – “Real” federated login for the instances that want it – meaning that users from other instances can log in to my instance through their account on their original instance, and participate on the instance just like registered users. This can obviously be disabled by default – Integration with Matrix: as a user authentication source, and as an alternative back-end to deliver messages. Mastodon messages could automatically be bridged to Matrix, and the other way around

Federated search.

Block Together instance-block features, so that admins can collectively make decisions on the shape of the fediverse, – Exporting my block-list for other admins to be able to use, – Importing block-lists, – Ban evasion detection, like, email addresses that’ve been used elsewhere and banned, to prevent bad actors from service-hopping,

Moderation tools should include a communications suite for communicating to instance admins/moderators and reporting user(s). Please let us delete custom Emoji groups, I’ve been asking for that feature for 3 years. Give us customization options to change the look and feel of our homepages. We’re running websites first and apps second here and not every one of us has the time to manage merging all Mastodon development in our own fork. MAKE MASTODON MODULAR so we can customize things without having to manage all of Mastodon development integration on top of our tiny little feature changes.

Post character limit to be configurable on a per instance basis, from the Web Admin UI. (also, since the text field is missing on the following question: YES Local and Federated Timelines are crucially important, especially the Local one. And NO, I would not disable them if possible).

the ability to add a theme more easily,

Work with Debian packagers to include Mastodon in Debian. https://github.com/mastodon/mastodon/issues/3576

Synchronized or shared block lists. 11510

The return of automatically followed account list for new users.

federated moderation: This contact other administrators via the moderation front-end

Easier configuration of the length of toots (5 times)

Some way of adding location data to a user profile, so that users can declare their general geographic location. And a way to find other users in a particular geographic location!

Mastodon Developers: The separating the FE and BE is an interesting one. Apart from that, these speak for themselves.

Question 18 Do you find value in the Local and Federated Timelines?

70% Yes – I apparently didn’t enable comments on this question.

Mastodon Developers: The federated timelines rule.

Question 19 Would you like to see rich-text supported in Mastodon?

80% yes. I think this is a pretty clear consensus. The question is, where is the limit of rich text?

I run Glitch for this reason.

YES PLEASE!!! Instances with longer character limits could then really turn into self-contained well-formatted blog-like platforms, rather than pointing to content properly formatted on other platforms.

yes, but it depends on the AP protocol features or it’ll be incompatible with everyone else

It can be an advanced feature but not necessary

yes, Since ActivityPub is based on HTML, I think that’s where we end up. especially the problem is losing “strike” and “quote” when importing posts from external servrices. it raises we can’t read the meaning of the sentence accurately.

Yes, But I imagine that complicates counting characters A LOT.

no, best way to read a post is with a plain text

Do you mean allowing more source mediatypes? Per instance/per implementation choice. I think if an instance wants to turn on HTML, markdown good for them, and if other clients don’t read it, good for them. (My -personal- choice is plain text)

I use glitch-soc which supports this natively

it’s supported by nearly everything other than mainline Mastodon, it’s more than due

I think I already have Markdown formatting with glitch. That and increased toot size was why I went with glitch.

I used formated text a lot in Telegram. If mastodon provided it, I would use it there too.

I don’t think that this is a feature that needs to be prioritized when some instances wouldn’t be able to use it, but at the same time glitch-soc does support this. If Mastodon has that feature parity, then there will be less confusion overall for toots that have rich text enabled where others do not.

No, This is neat but we don’t need it and it doesn’t do anything but make the timeline more obnoxious.

Toots should have a content type, only one example of which should be markdown, and folks should be able to select the content type in which they create. Clients can either render that content type raw, formatted, or filter it out.

it makes content consumption more convenient

No, I’m pretty sure that rich formatting would produce more negative effects than positive ones.

Sure, whatever, wasn’t there full LaTeX support at one time?

Yes, but only the basics

Mastodon Developers: Markdown please. Personal opinion is close to the github markdown standard… blur/spoiler is important. Font style/colour and background colour is a bonus. Animation is a stretch.

Question 20 Would you like to see twitter style “Quote Toots” in Mastodon?

55% Yes / 45% No – this is a real split question.

I know about the line of thinking that this yields passive-aggressive boosts. While I think that this does happen on twitter, I also think that this style of quoting something is still valuable more often than not.

Quotes are abused heavily at twitter to bullying other users. toots should not rendered at all.

Yes god yes I miss this feature so much, it’s SO annoying to not have it, like, CONSTANTLY so annoying.

I’m torn on whether I want the ability to create QRTs, but we should at least render them properly

No, I wouldn’t mind seeing these display correctly but think mastodon is better off without the feature. CW boosts and public posts within threads get the message across just fine without the harassment risk.

Quote toots offer a better commenting experience than boosting and replying. What people do instead is copy & paste links, which is really weirdly backwards – but the only thing we have.

quote tooting should not exist at all. I do not like that Misskey and Pleroma support quote tooting, which is probably not helping my perception of them as platforms for troll instances.

I can see why Mastodon doesn’t support it, but I feel like the benefits outweigh any negatives.

Being able to add a few comments of your own to a boost would be quite handy to add context.

this will allow a more accurate description of a piece of text, users now use the quote symbol instead >

This was discussed many times, and I agree with Eugen’s assessment they would be harmful https://github.com/mastodon/mastodon/issues/309#issuecomment-469260847 Additionally, they make it incredibly hard to follow discussions on Twitter when people passive-agressively quote-tweet each other repeatedly

Not as an explicit thing I don’t think (like twitter’s quote reply button), but rendering a social preview of a linked toot would be good, I think?

Mastodon Developers: This is Split down the middle, but once again. *I* think it’s worth it. and if people can disable quote tooting THEIR toots. opting out would be fine?

Question 21 Would you want to see Tumblr style “toot remixing” in Mastodon?

Mastodon Developers: ?

Question 22 Would you like to see a more interactive (Audience participation) status types like the question/answer format?

Mastodon Developers: ?

Question 23 As both an user and admin of Mastodon, What are your wildest dreams feature requests?

Mastodon Developers: ?

Question 24 Is there anything you would explicitly would like raised that was missing from this survey?

Mastodon Developers: ?

Question 25 Final Question, What’s the best part of being a Mastodon Administrator?

and ending on some sweetness and joy.

Helping people connect with each other 🙂

I like being part of creating a better social network that isn’t selling user info

the gratitude of the users

The freedom of shaping the social media experience exactly the way me and my users like.

I get to send people to horny jail.

It connects me with my community, and refines my server administrating skills.

Give the Internet something back! 🙂

That sensation you have when your users thank you about how are you doing the administrator tasks

Being a small piece of a large puzzle. Being participant in a project and community I can align with. Contributing to the kind of Internet I’d like my kids to be on.

I love being able to say I’ve made a community exist, that I’m part of something bigger, that I can serve people like this.

it was traveling around and meeting cool folks from the community, but COVID stole that from us. so I’ll go with: providing the infrastructure for hundreds of people to learn more about each other and the world 🙂

Getting to know our users well, while still staying connected to other instances and making friends there. It’s consistently the best of both worlds.

Being part of a community and help other people to connect.

The folks on my local are lovely!

I like having my own space that’s not controlled by Facebook or any of the others. I like putting some money to caring about free software.

Being able to give to the community. Being part of a big project, even if in a small way.

Meet good people

Seeing your community thrive and enjoy each other, or build off one another to do something cool together.

When users say thank you, honestly. It’s really nice when people just say thanks.

Not having to worry about corporate control/influence over my content and data.

It’s great to offer a legit, self-hosted, open source based, privacy friendly alternative to Twitter.

the community and feeling involved in maintaining it, for the best and the worst

Support people who need something better than commercial platforms.

Giving other people a home instance that they enjoy.

Knowing I’m helping people escape the walled gardens!

Mastodon Developers: ?

Summary:

Mastodon Developers: Thank you for your hard work. Thanks for spending your time to read this.

<TODO>

Features Admins want:

Admin control to turn off features (trending etc)

Global Full Text Search

Spam Protection

Categories: Mastodon

Tagged as: 14597 4106 fediblock introduction Mastoadmin

Shlee

Lover and Sickle

2022 Unofficial Mastodon Survey Results

Introduction

Survey Results

Question 1: “Do you believe your instance has a unique theme/culture? or is the instance just a gateway to the fediverse as a whole?”

Question 2: “Do you believe the fediverse should be considered one large ecosystem? (or are some instances islands?)

Question 3: “What do you think the Mastodon project is doing well?”

Question 4: What do you think the Mastodon project could improve?

Question 5: “Do you believe instances admins have enough influence on Mastodon?”

Question 6: “Would you support establishing a steering committee? (Building community roadmaps)”

Question 7: “There is no public roadmap for Mastodon, are you aware of the planned “Groups” functionality?”

Question 8: “Does this paragraph match your expectations of how search should work in Mastodon?”

Question 9: What do you feel about the act of moderation? Do you feel the tools provided (as of July 2022) are adequate? What could be done to improve moderation?

Question 10: Do you have a problem with spam on your instance?

Question 11: Do you have a problem with abuse on your instance, or the fediverse in general?

Question 12: As a Mastodon end-user, how do you rate the UX?

Question 13: Do you or your users experience problems with the accessibility of the UX?

Question 16: As a Mastodon end-user, What front-end features would you like to see added (and prioritized) to Mastodon’s roadmap?

Question 17 As a Mastodon Administrator, What back-end features would you like to see added (and prioritized) to Mastodon’s roadmap?

Question 18 Do you find value in the Local and Federated Timelines?

Question 19 Would you like to see rich-text supported in Mastodon?

Question 20 Would you like to see twitter style “Quote Toots” in Mastodon?

Question 21 Would you want to see Tumblr style “toot remixing” in Mastodon?

Question 22 Would you like to see a more interactive (Audience participation) status types like the question/answer format?

Question 23 As both an user and admin of Mastodon, What are your wildest dreams feature requests?

Question 24 Is there anything you would explicitly would like raised that was missing from this survey?

Question 25 Final Question, What’s the best part of being a Mastodon Administrator?

Summary:

shlee

Leave a Reply Cancel reply

2022 Unofficial Mastodon Survey Results

Introduction

Survey Results

Question 1: “Do you believe your instance has a unique theme/culture? or is the instance just a gateway to the fediverse as a whole?”

Question 2: “Do you believe the fediverse should be considered one large ecosystem? (or are some instances islands?)

Question 3: “What do you think the Mastodon project is doing well?”

Question 4: What do you think the Mastodon project could improve?

Question 5: “Do you believe instances admins have enough influence on Mastodon?”

Question 6: “Would you support establishing a steering committee? (Building community roadmaps)”

Question 7: “There is no public roadmap for Mastodon, are you aware of the planned “Groups” functionality?”

Question 8: “Does this paragraph match your expectations of how search should work in Mastodon?”

Question 9: What do you feel about the act of moderation? Do you feel the tools provided (as of July 2022) are adequate? What could be done to improve moderation?

Question 10: Do you have a problem with spam on your instance?

Question 11: Do you have a problem with abuse on your instance, or the fediverse in general?

Question 12: As a Mastodon end-user, how do you rate the UX?

Question 13: Do you or your users experience problems with the accessibility of the UX?

Question 14: Do you find value in the current “Explore” functionality? (Trending Posts/Trending Hashtags/News/For You)?

Question 15: Would you support the ability to append “Licensing” metadata to toots? Similar to Creative Commons, but with discoverability and other social network elements?

Question 16: As a Mastodon end-user, What front-end features would you like to see added (and prioritized) to Mastodon’s roadmap?

Question 17 As a Mastodon Administrator, What back-end features would you like to see added (and prioritized) to Mastodon’s roadmap?

Question 18 Do you find value in the Local and Federated Timelines?

Question 19 Would you like to see rich-text supported in Mastodon?

Question 20 Would you like to see twitter style “Quote Toots” in Mastodon?

Question 21 Would you want to see Tumblr style “toot remixing” in Mastodon?

Question 22 Would you like to see a more interactive (Audience participation) status types like the question/answer format?

Question 23 As both an user and admin of Mastodon, What are your wildest dreams feature requests?

Question 24 Is there anything you would explicitly would like raised that was missing from this survey?

Question 25 Final Question, What’s the best part of being a Mastodon Administrator?

Summary:

shlee

Leave a Reply Cancel reply