Mastodawn

Humi Jan 5, 2023

Arvind Narayanan

The number of active Mastodon users — those who've logged in in the last 30 days — has fallen to 1.8 million from a high of 2.6 million. This reflects the fact that while there's a wave of new users after each Musk tantrum, the majority aren't sticking around. Even the 1.8M figure is inflated by new accounts, so I expect a further drop.

Still, the number of active users is over 4 times what it used to be before the Twitter takeover.

Source: https://api.joinmastodon.org/statistics (and Wayback machine).

Arvind Narayanan Jan 4, 2023

Even though the number of Mastodon users is relatively tiny, it still feels like there's critical mass here because people in some communities are much more likely to have migrated. I previously wrote a thread about this: https://mastodon.social/@randomwalker/109548309922018593

Dan Cross Jan 4, 2023

@randomwalker this. I don't much care about the vast majority of twitter users (most of whom I didn't follow anyway). But I'm also mindful of the early days of Google+, where many, many users created accounts to kick the tires and give it a spin...but most didn't buy. Still, if the center of gravity of the communities I want to be involved with are over here, I'm happy.

Steve Bellovin Jan 4, 2023

@randomwalker What's missing thus far from Mastodon is institutional accounts: companies, governments, etc.

hannah Jan 4, 2023

@randomwalker the big problem here is the gender balance. it's bad everywhere online but i have seen few other women posting anywhere other than the art channels or ~official~ pages that just migrated. i dunno how to fix it though

malena Jan 4, 2023

@h_thoreson @randomwalker men could post less, consider not replying unless with support / additive, and boost boost boost women

hannah Jan 4, 2023

@seachanger @randomwalker I hesitate to say that anyone should post *less* but it's very obvious that there are almost no women on any of these servers. I went into the music section and it was just sad. Basically if it was recorded after 1992 or by a woman it may as well have never happened.

hannah Jan 4, 2023

@seachanger @randomwalker But if this is ever going to be a viable replacement for Twitter that has to change, a lot. Most people weren't really using Twitter for news or highbrow tech/journalism/networking, they were just there to soak up pop culture, memes, gifs from shows, share little projects they were working on, and generally hang out

malena Jan 4, 2023

@h_thoreson @randomwalker maybe we could create femme.social or similar lol

WhatCouldPossiblyHappen Jan 4, 2023

@h_thoreson @seachanger @randomwalker I found that the infosec.exchange instance has quite a few non-male persons actively contributing (some high-profile ones). It's still a predominantly male community but it seems that there was a wholesale-migration of the infosec community which left gender-balance intact and didn't make it worse.
Maybe migration speed is more gender-dependent in other communities - if everybody migrates eventually, that would only cause a temporary bias. But if migration of under-represented groups is slowed down by the current bias itself, that will be a real problem.
Unfortunately, I cannot offer any bright ideas other than trying to make mastodon a friendly and open place and hope for the best. It would be a real shame if mastodon would end up as a boys club.

hannah Jan 4, 2023

@WhatCouldPossiblyHappen @seachanger @randomwalker That seems to be one of the biggest servers in general yeah but I feel reasonably certain most people I know in real life are maybe dimly aware of what infosec even is. Part of why I went with one of the general interest ones. I don't want to have to explain myself before I get a chance to do that on my own terms.

WhatCouldPossiblyHappen Jan 4, 2023

@h_thoreson @seachanger @randomwalker sure. Didn't want to suggest joining infosec because of gender-balance. It's a special interest instance for sure.
Just wanted to point out that migration dynamics seem to be different in different groups. Maybe my point was: if the community as a whole migrates (eventually), it might do so in all its diversity - so there's hope.

hannah Jan 4, 2023

@WhatCouldPossiblyHappen @seachanger @randomwalker That's a good point, I'd love to see some network graph data that shows how people are choosing to move over. If it's in a cohort as a community that would make a lot of sense. I think back to when I joined Facebook - it was because I went to summer camp out of state with kids that went to prep schools and that was what they were allllllll talking about so I had to join Facebook too to be in touch

WhatCouldPossiblyHappen Jan 4, 2023

@h_thoreson @seachanger @randomwalker such a graph would surely be awesome! But getting reliable data for that is really hard.
Establishing identity of accounts by name or profile references might work for high-profile users, but surely not for casual ones. Not to mention the problem of getting metadata of blocked or deleted twitter accounts. Not everybody writes intro toots mentioning migration, so you can't know what you would get from those.
And if you ask people about their migration (a "migration poll"), the data will be a complete mess.

So I not much hope here. But if you find anything, please let me know. I'm very curious and happy to stand corrected.

hannah Jan 4, 2023

@WhatCouldPossiblyHappen @seachanger @randomwalker There's an obvious way to get at it indirectly, actually, but you'd need to have already collected the data unless Twitter's API has changed a lot. If you had a list somewhere and just buckets of just a raw post count pre-Elon, and then during each tantrum-induced wave of migrations, you would probably see a decline to near-zero over time for communities that migrate in a cohort.

hannah Jan 4, 2023

@WhatCouldPossiblyHappen @seachanger @randomwalker But most social media analysis tools really limit what you can do going back more than 30 days unless you already have the data stored.

WhatCouldPossiblyHappen Jan 4, 2023

@h_thoreson @seachanger @randomwalker that's a neat idea. It wouldn't tell you where people went, but you would at least see how communities leave twitter.
People maintaining both profiles in parallel for a while might be a challenge, so it won't really be a real-time observation. But 1:1 cross-posting will be more common for high-profile accounts (which cause more tweets, but they are also easier to match by metadata).
Sounds like you have a chance :smile:

Chaz Brenchley Jan 4, 2023

@h_thoreson @seachanger @randomwalker - Masto was never intended to be a replacement for Twitter, though. It's very much its own thing, with different aspirations. It is true that the influx of Twitter exiles must have an influence on how it develops hereafter, but if we push it into a "Twitter 2.0" box, then it will no longer be anything remotely resembling itself.

malena Jan 5, 2023

@h_thoreson @randomwalker @ChazBrenchley I believe we are just wanting for this to be a viable place to build community as more people realize the pitfalls of corporate socials. for that to happen, white males will need to be quieter in the replies and let women/ communities of color etc share their aspirations and experiences without being immediately reminded their experiences are wrong/not welcome etc

Jon (now at neuromatch.social)Jan 5, 2023

@h_thoreson @seachanger @randomwalker @randomwalker yep -- great point. there actually are a fair number of women here, they're just a relatively-small minority -- and guys tend to follow and boost guys, so that exaggerates the effect. Also, there's a lot of causal misogyny here ... as @ubiquity75 is one of many to point out. https://dair-community.social/@ubiquity75/109593017676917180

The same's true with racial disparities as well. It's a huge challenge.

Sarah T. Roberts, Ph.D. (@[email protected])

Content warning: Masto misogyny

Distributed AI Research Community

Pat Jan 4, 2023

@h_thoreson @seachanger @randomwalker I'm here 😊

Eric J. Korpela Jan 4, 2023

@h_thoreson @seachanger @randomwalker No women? Just went down my home feed until I found 10 unique individual posters who were identifiable as women. In the same distance I found 13 men, 4 for which the gender was unidentifiable, and 5 organizations. That's about the most equal distribution I've found anywhere online. So to me it seems OK.

I suppose it would be different if I looked in federated. Just did it in federated (on qoto.org), when I hit 10 women, there were 14 men, and 15 organizations+unknowns. Again, that's low disparity for the online world.

Sure that's small number statistics, but large enough that it's not likely to be horribly wrong.

kamenlady Jan 5, 2023

@h_thoreson @seachanger @randomwalker i started crawling the internet for music around 98. From obscure mexican industrial electro or ukranian experimental glitch to folk ballads or soundtracks. Some of the finest tunes i found, were made by women.
I'm a guy and would love a music community that is there for the music per se, no strings attached.
I'm a social disaster and would never have the ability to create a community by myself tho

baconandcoconut Jan 4, 2023

@seachanger @h_thoreson @randomwalker I've seen a few responses here to women or femmes that are basically a challenge to a feeling or a "request" to prove a point and those aren't very welcoming. (This is my anecdotal observation.)

Gustavo 🎮🔭🇵🇷Jan 4, 2023

@seachanger @h_thoreson @randomwalker ok so not my custom to argue against opinions, but telling us to simply “post less” doesn’t make any sense. Is this a social network?

KatM Jan 4, 2023

@h_thoreson @randomwalker I guess I'm on .art so my experience is skewed, but I was just thinking how I have more women mutuals here than I did on Twitter. Maybe from following craft groups/hashtags...but also women scientists...

hannah Jan 4, 2023

@KatEmm @randomwalker Right, I did note that the art channel seems more balanced than most of the rest of the servers I have looked at

Nonie Jan 5, 2023

@KatEmm @h_thoreson @randomwalker
This is my #strategy - #follow, follow, follow. I'd rather follow too many rather than risk missing out on someone interesting because I wasn't aware of them. If someone just isn't my cup of tea, I can always unfollow later. My feed right now is pretty #diverse and very interesting!

wythran Jan 5, 2023

@h_thoreson my solution to this was to make a deliberate effort to even out the gender balance of my follows. Doing this for a little while has dramatically improved my experience in this regard. Feel free to check my follows for people of interest.

Moon Jan 5, 2023

@h_thoreson in light of @randomwalker ‘s post what might this imbalance mean, particularly considering the recent trends on Twitter? I might have expected the imbalance to favor women instead.

hannah Jan 5, 2023

@lunar @randomwalker women thought elon musk was a rich loser long before the twitter dumpster fire. men only changed their mind after seeing him ruin a bunch of stuff they liked

Moon Jan 5, 2023

@h_thoreson @randomwalker but the imbalance remains unexplained. If true, where have the women gone if not to Mastodon? Simply lurking or flat out dropped Twitter without replacement? I’m not conversant in social media usage stats so pardon my naïveté.

hannah Jan 5, 2023

@lunar @randomwalker instagram and tiktok

Moon Jan 5, 2023

@h_thoreson @randomwalker 🤔 hmm, maybe the collective state of our feeble minds has rendered Twitter a broken means of sharing a public space. That is, we’re the problem, not so much Twitter. (Another way of saying not a sufficient critical mass of grownups in the room to make it work)

dragonfrog Jan 5, 2023

@h_thoreson @lunar @randomwalker I mean, I knew he was a rich loser for approximately as long as I knew he existed. He just wasn't the rich loser who owned Twitter until recently.

Kim Scheinberg Jan 5, 2023

@h_thoreson @randomwalker
I see lots of women in part because I follow lots of women. If you're looking for suggestions for women to follow, this thread is helpful:

https://mas.to/@heidilifeldman@mastodon.social/109594737479973050

Heidi Li Feldman (@[email protected])

Please reply to this post with suggestions of one or two women to follow here on #Mastodon, anybody you find interesting. I want to make sure I hear their voices. #SuggestedFollows

Mastodon

Olav Jan 5, 2023

@h_thoreson @randomwalker follow more women and queer folks. Boost their posts. My experience here is a helluva lot more diverse than Twitter ever was

Rebecca Nordquist Jan 5, 2023

@h_thoreson @randomwalker Agreed on this, I've been actively following and boosting women for the past few weeks for this reason. I do see a change in my timeline, but it is gradual.

That also means *not* boosting posts by men, even if they are cool, because otherwise that shift never happens. I have mixed feelings on this, but am keeping it up for another few weeks to see if I can get my corner of Mastodon more balanced.

Michał "rysiek" Woźniak · 🇺🇦Jan 5, 2023

@renordquist as a man I implore you: please do not feel bad about not boosting my posts.

Make your timeline your own, and absolutely do boost the signal you feel needs boosting, even if that means some things that are cool remain not boosted. 👍

Thank you for making fedi better and more inclusive!

@h_thoreson @randomwalker

mkb Jan 5, 2023

@renordquist @h_thoreson @randomwalker

Many of us have your back.

Over on Birdsite a decade ago, @anildash did something similar. Anil made a rule for himself to only retweet women for a year.

https://medium.com/the-web-we-make/the-year-i-didnt-retweet-men-79403a7eade1#.kxrd4ptzh

The Year I Didn’t Retweet Men - The Only Woman in the Room - Medium

In its early days, Twitter used to have a suggested user list, people whose Twitter accounts were recommended to people signing up for the service. If you got added to the list, you ended up a bit…

The Only Woman in the Room

@renordquist I'm very much involved in #OpenStreetMap and am running a user survey. The #genderBalance in that community is very bad as well, with about 90% identifying as male. For now, there are more non-binary and genderqueer people who responded then women... (and I'm very glad to have them in our community, but it is striking to end up with such ratios - it means women are vastly underrepresented here). But I will try to fix it a little bit! (Thanks to #NlNet)

Micaela Olds Jan 5, 2023

@h_thoreson @randomwalker fix your personal network. No algorithm is going to do the work for you.

💀🌇 FKA kevin Jan 4, 2023

@randomwalker If only we could pair these data with Twitter usage. Folks might not stick around here but do they return to Twitter? Or do they abandon the infinite scroll altogether?

Chris Messina Jan 4, 2023

@driscoll @randomwalker had the same thought, re: Twitter usage.

Nate Jan 5, 2023

@chrismessina @driscoll @randomwalker a retention problem is a retention problem - no comps required.

Chris Messina Jan 5, 2023

@innonate @driscoll @randomwalker 100%

💀🌇 FKA kevin Jan 5, 2023

@chrismessina @innonate @randomwalker agreed--it's important to think about retention independent of any other system.

But I also want to know more about this moment of collective exploration. I don't know how to interpret the decline in Mastodon usage without knowing if there is an overall shift away from this type of social media. It's possible that the downward trend is much worse over at FB and Twitter!

Chris Messina Jan 5, 2023

@driscoll @innonate @randomwalker it could also be seasonal — based on the holidays?

Nate Jan 6, 2023

@chrismessina @driscoll @randomwalker my point is that at this phase for a social product and this scale you don’t want secular trends or seasonality to be noticiable.

💀🌇 FKA kevin Jan 6, 2023

@innonate @chrismessina @randomwalker have you seen this kind of retention issue before?

Anti-monopolist Jan 5, 2023

@driscoll @randomwalker I’m here because my Twitter account was permanently suspended a couple months ago and they won’t tell me why. Never even had a temporary suspension or a reprimand before this.

💀🌇 FKA kevin Jan 5, 2023

@maberuth @randomwalker you ain't alone, it seems

knova is testing Akkoma Jan 6, 2023

@driscoll @randomwalker sample size 1, but I made a deliberate choice to never go back. It's nice not doomscrolling anymore!

Hello57vis Jan 4, 2023

@randomwalker
Thank you very much
amazing how #mastodon users, just like #twitter users created a giant meme ["the great migration" which is a mirage

poratlu Jan 4, 2023

@Hello57 @randomwalker 4x is a mirage?

Hello57vis Jan 4, 2023

@poratlu @randomwalker

it is a lot from the perspective of tusker, but from the perspective of twitter, no
and twitter itself is small compared to other sites

PeggyJoan ☮️💙🌈🐶Jan 5, 2023

@Hello57 @randomwalker
Thank you for allowing me to sit in your class! You have taught me a lot about Planet M.🤗

Wu Evar 🇪🇺🇺🇦Jan 4, 2023

@randomwalker Do crosspost accounts (or lets call it what they are: mostly bots of guys who pretend to care about Mastodon but in reality are probably like junkies on Twitter) count into number of active users?

Arvind Narayanan Jan 4, 2023

@maschinentraum I'm pretty sure crossposts count as activity, so yes, the number of people who are actually using Mastodon is even lower than the graph shows.

Wu Evar 🇪🇺🇺🇦Jan 4, 2023

@randomwalker That's sad :(

Expected so. Used the migration tools when they became available. Most of the big accounts I follow(ed) on Twitter have accounts here, but don't post or don't interact at all.

Same thing as usual: "XYZ is evil! Let's boycott! We can do this!" - Ah well, we have to take some minor drawbacks - "Well XYZ _might_ be evil, but I need my likes.. um I mean we have to fight evil at it's root and not give up, only cowards run away. 🤷"

Wu Evar 🇪🇺🇺🇦Jan 5, 2023

@randomwalker oh and with the most popular Crossposter shutting down pretty soon… this event will be a good indicator how many users care.