It has come to my attention that Mastodon, and/or OSocial, and/or the Fediverse, are no longer what they once were. They're being flooded by Normals. Or People who Don't Speak my Language. または私のアルファベットを使用しない人。
It's becoming like ... birdland.
Or worse: Facebook.
So, here's a thought for you: Facebook was once literally Harvard.
And whatever you think of FB or Harvard, it most certainly isn't any more. Which is a thought you might want to keep in mind. Because change.
Suppose, as a thought experiment, you could create the ideal, perfect, social network or discussion board. Say, with 40 of the smartest, most creative, quirkiest, considerate people you knew. Hell: the 40 top exemplars of this on the whole planet. It would be a pretty awesome network.
(I know this because I accidentally created something like this, just by creating a small group with smart and interesting people in it. It really was surprisingly good.)
It can only get worse.
Because if you've already got the best, then /anyone else you can add/ will be less smart, less creative, less quirky, less considerate, than who's already in the group.
And at some point you'll notice. Maybe at 50 people, or 500, or 5,000, or 50k, or 500k, or 5m, or ....
For a few reasons.
* Gradation of capabilities. These are ordinalities, not cardinalities.
* Limits to common experience and interest.
* Differences of opinion. Or morals. Or philosophy.
* Just plain scale.
I joined Mastodon in what I'll call the late beginning: ten days ago.
The total Mastodon and connected userbase was under 100k users. It's at or above 300k now, tripled in less than two weeks. The largest instance, mastodon.social, had 40k users, and was double or more the size of the next largest, mastodon.cloud (which I joined, as the largest open instance). There are now several instances of comparable size, and .social is no longer the largest.
If you joined Mastodon because it was small and quirky and whatever non-mainstream group you align with felt like it was welcomed, and comfortable, and safe here, I've got some really bad news.
You didn't find a space that was inherently suited to your wants, but one that happened to support them. Because of scale and founding cohorts.
Those are the first two characteristics to be lost in any growing community. Unless it was dead-normal to start with, and even then I'd bet against it.
A while ago, in another recently-launched social network, I got in some hot water by suggesting that specific-community cultures -- LGBTQ, feminist, athiest, pro-science, pro-spiritual, academic, geek, whatever -- was going to find itself facing the fact that it is, at best, a modest-sized (if not minuscule) minority in a much larger world.
Have I mentioned Japan recently?
Over the four days, Local and Federated changed completely.
Mastodon was no longer mostly French and German, some English, and a smattering of other languages.
It was, as I commented the first time, Turning Japanese. And is now a Plurality.
It's interesting not only being a minority /language/ as a principally English speaker, but a minority /alphabet/, on an online system. And it's an interesting problem to have.
There are also several larger populations out there, not yet connected.
There are roughly 1 billion Muslims in the world. Another billion each in China and India. A billion in Africa. /Another/ two billion through the rest of Asia /not/ included in China and India. A half billion in South and Central American and the Caribbean.
And the roughly 1 billion of the US, Europe, Canada, Australia. Oh, hello, New Zealand, you're so tiny and cute I could just *pinch* you!
"Normal" itself is a minority, if you mean US, Canada, UK, EU, AU, NZ.
This also means that, say, conservative protestant Christian views ... are a distinct minority. As are white-supremacists, the KKK, Westboro Baptist Church, Libertarians, Palastinians, Israelis, Arabs, English, Irish, Waloons, Basques, ...
But if you want to find any /specific/ group with an oppressive / insurgent agenda against some other group: yeah, they're a minority too.
The good news: they're not the minority.
But consider: being a minority doesn't make you right. Or good.
Errata: that should read "The good news: they're not the /majority/."
Being /good/ makes you good.
Oh, and not just that one time.
Every. Fucking. Day.
Or at least most of them.
A similar thing about being right: you've actually got to, you know /be right/ about something. That is, your world-model has to check out when compared against the actual world. And continue to check out, including as new, potentially improved models, come along. Though those /also/ have to pass the test, and not just be wrapped in flashy marketing. But that's another essay.
So: On behalf of myself, I'd like to welcome everyone on Mastodon, OSocial, and the Fediverse to your (newly realised|long accustomed) minority status.
Get used to it.
Get used to the fact that others are, perhaps, getting used to it. Some for the first time.
Some, with more than a little resignation, not.
But realise that, unless you want to want to use some very broad classifications, there are few actual /majorities/ on Mastodon.
If you're right-handed, dark-haired, probably.
I for one welcome our dark-haired, right-handed majority.
Of course, one of the fictions of society is that majority or minority status is most significant. It's not.
There's a term that encompasses this: Oligarchy.
Rule by the few.
What discussions of minority and majority are almost always really about, as are discussions of econonomics, politics, equality, and justice, is power.
The ability to act according to your own will.
Or the ability to convince /others/ to act according to your will.
Which is to say the minority / majority, outgroup / ingroup, nonnormatives / normals discussion is one about /power/.
Accessible and clear texts on power itself are hard to find, though the classics -- Aristotle, Machiavelli, Hobbes, Locke -- are recommended. A more recent, readable, and reasonably brief example is John Kenneth Galbraith's "The Anatomy of Power".
Galbraith describes three types of power:
* Compensatory: Buying followers.
* Condign: Obedience through threat.
* Conditioned: Influence through persuasion.
It's a good start for discussion, including discussion about whether the classification is accurate or sufficient. I'll assume it is and consider it in the context of social discussion.
What, exactly, is the power that others hold over us -- me or you, or another user, say -- in the context of this platform?
Compensatory power might attempt to buy influence to say (or not say) something based on reward. That would be /inducing/ participation of a particular type, generally. In other online media, advertising and propaganda are the main exemplars of this.
Condign power might attempt to /suppress/ (or more rarely, compell) content based on direct or indirect threats.
Conditioned power would be persuasion, through example, norms, standards, codes of conduct, etc., to convince others to behave, or not behave, in specific ways.
I've seen this in several interactions on Mastodon from my first day onward, as norms have been suggested: that any news or politica discussion is unwanted or should be CW'd (an unrealistic and for many users unfeasible request), animations, porn, trolling. The usual. Some of which has somewhat worked, some not.
Culture is hard.
This doesn't mean that culture is /impossible/, but it's got to be, well, /cultivated/.
I'd welcomed a friend from elsewhere to Mastodon recently, and helped orient them after the sign-up process dumped them to a blank page.
The fact that there's no automatic welcoming / orientation message or committee, particularly on large instances (30k), is ... disorienting. And with Mastodon growing at the rate of 200,000 users every 10 days, at an accelerating rate, that's a lot of onboarding.
Remember, remember, the Eternal September.
(And yes, if there's a discussion, and I'm aware of though haven't yet signed up for the mailing list, I'll join that.)
Old-school OSocial admins have likely been around the block a few times, though I'll caution /them/ that this isn't the scale they're used to.
And there's a body of research and specialists now with literature.
Andrew Odlyzko. Stewart Brand. Howard Rheingold. Doc Searls. Scott Alexander. Biella Coleman. danah boyd. Clay Shirky.
Some of these tend to the idealistic. The California School especially thought that bringing the Minds of the World together would bring nothing but good. I find that view ... optimistic.
Clay Shirky: "I study social media. Which is to say as a first approximation: I watch people argue."
What happens when you make it easy for anyone to talk to anyone?
What happens when you make it easier for anyone to talk to anyone else is that everyone finds far more people, and ideas, to disagree with.
Somebody is /wrong/ on the Internet.
I've got some good news here: there hasn't actually been any net increase in the number of people who were wrong. It's just that they, we all, are more visible.
In an odd way, contact through media is far more mind-to-mind than you experience in shared space -- the stranger on the bus.
Or on the train, or in the car next to you, in line at a cafe.
Unless you can get them to actually /talk/ to you -- something we often don't do in public spaces. But do on the Internet.
And it turns out that direct mind-to-mind connections can be ... problematic.
(They can also be wonderful, and I've already met some really incredible people via Mastodon.)
I'm coming to think that a part of this is that there are limits to how much humanity we can handle. Or, again, scale matters.
Dunbar's number is a well-known concept now: the idea that people can maintain stable social relationships within groups of about 50-150. It's not just the relationship between yourself and each of the others, it's how /they/ relate to one another: does Bob like or hate Sue, or Jose, or Gita? With 150 people, there are 22,500 relations. With 30,000, there are nearly one billion.
There are reasons for suggesting and encouraging smaller instances, as OSocial admins recommend.
I've also been looking at media, messaging, context, and especially the trade-off between /quantity/ and /attention/, over the past six months or so.
James Gleick: "When inforation is cheap, attention becomes expensive."
dredmorbius: Rivality is the counterpoint of virality.
The distinction between /cardinality/ and /ordinality/ is one I'm increasingly aware of.
Cardinal numbers are the counting set: 1, 2, 3, ...
Ordinal numbers are the ranking set: 1st, 2nd, 3rd, ...
When you increase the size of a set, or a network, you increase its /cardinality/. Again: Mastodon's growth from 80k to 300k users in the past ten days.
What does /not/ change with scale is the ordinality. If you double the size of a network, there are still precisely ten "top ten" slots. But there's twice the competition for them, or twice as much content (assuming linear scaling) to be filtere to find that set.
And there are only 24 hours in a day.
Studies of Internet usage -- Pew Research is a recommended source -- have shown roughly 40 minutes/day of primary social media (FB) use.
Divide that 40 minutes by messages, and you get the time consideration for each message. 10 messages = 4 minutes each. 100 messages = 24 seconds each. More time on one message is less, or none, on the others.
I've found a few citations for daily information consumption by experts, which are interesting.
Stephen Wolfram, creator of Mathematica and Wolfram Alpha, has tracked life metrics since 1989, including daily outbound emails. The data are noisy, but averaged by month, works out to about 50 per day, with a mode (most frequent value) of 20-25. /Inbound/ email ranges about 300.
ReCode's Walt Mossberg reports about 350 emails/day.
The NYTimes comment moderation team handle 800 comments/day, each.
I'm suspect these values are upper bounds, and in the NY Times case, pretty close to the peak, of what someone processing information and having to make /some/ decision or determination based on it -- a post / no-post decision in the case of the Times, a brief, possibly extended, reply in case of Wolfram's email -- might be expected to handle.
And that the true value is probably in the 10 - 100 items range.
Another useful statistic is "how many articles does a newspaper publish in a day". In the case of the Washington Post, New York Times, Wall Street Journal, and Buzzfeed, the anser is about 150 - 500 items per day of original, locally-sourced content.
News-wire services such as AP, Reuters, and UPI offer 2,500 - 5,000 items. Numbers researched investigating questions of "fake news".
@dredmorbius If anything, my biggest issue with Diaspora was that normal people weren't there. It seemed to be 100% people who wouldn't fit in Facebook or Twitter, and I kinda want a mix of everyone.
@ocdtrekkie Places which appeal to, and attract, /only/ societies misfits ... tend to become communities of misfits.
I suspect that was a part of Imzy's problem as well.
I'd like to see a broader appeal. I'm seeing it.
@ocdtrekkie This. I very much enjoy the new people I'm meeting here, and it's nice to hang out with people among who choice of distro is no weirder than Windows or Mac. But I miss interactions with people whom I have established connections. Which is important to me in a social network. @dredmorbius
@dredmorbius interesting, you are the bearer of bad news I understand...as I read more of your posts.
@dredmorbius I get the feeling you are sounding the dooms day trumpets because you are on the fourth largest instance.
When we sort out account migration, you will be able to move. Less people in an instance usually equates to higher specialization of that instance.
Just going to leave this link here, if you want to help speed development. https://github.com/tootsuite/mastodon/issues/177
@NthTensor Actually, the complaint isn't mine at all. I'm enjoying the ride.
I've /definitely/ seen grousing by others, though. And wanted to address a few points:
* The perception is real (As is the pain)
* This is inevitable. Fighting this will make you /very/ unhappy
* The FB:Harvard thing. I mean, c'mon, Facebook used to be /The Facebook/ and now it's ... /Facebook/
* Looking for solutions, including community accomodation, /and/ moderation, is necessary
@dredmorbius That's true though, it is a minority. doesnt mean it's wrong to say that.
@dredmorbius omg - I haven't heard that song in years ... someone should figure out a way to sort by language ... I don't have an issue with other cultures but having the majority of a feed be unreadable isn't really that ... compelling? ... yeah - I noticed the language problem when I clicked on the federated feed - unfortunate
@Xenophrenia The ability to specify (or cycle through) language(s) of interest would be useful, as would be embedded translation.
I believe both are on the map.
@dredmorbius please explain the scoring at the botton this time it's 18/. I love your writing BTW (not sarcastic)
@Publicvent Sequence numbers.
@dredmorbius fascinating...does bob have children, what car does he drive, where does he work a million things to consider when knowing everyone. They say you have the capacity to know 50-150 ppl but i will never forget every person ive met. even you
@Publicvent Should I or should I not tell Bob's wife about his mistress. Or guy-on-the-side ...
@dredmorbius I was listening to a #podcast (You Are Not So Smart; "Fallacy Fallacy" episode) about arguing & internet discourse & one of the the things that struck me was how there's a pervasive belief that we write less/are less literate because of the internet, where the reality is we collectively write way more in total and on a daily basis than we (collectively) used to.
@frankiesaxx That'd be interesting to get a real look at. I suspect we're exposed to far more causual writing now than in quite some time.
As of 1800, only about 25% of Europe's population was literate, and earlier than that rates may have been as low as 5% in rural areas. The culture of letter-writing itself had to be developed, and use of postal mail was a fairly newish tradition in the 19th century, for the general population. Books gave specific guidance on style.
@dredmorbius This is the podcast episode: https://youarenotsosmart.com/2016/01/22/yanss-067-the-fallacy-fallacy/
I can't remember if he says where the info is from; it might be the Stanford study mentioned here:
@frankiesaxx Thanks, listening.
Woah: Typical person, after high school / college: 3-10 written documents for the entire rest of their life, prior to the Internet.
I'm ... trying to think of how much I've written.
@dredmorbius Haha I know! But it was specialized labor. Amazing to think.
@frankiesaxx I'd like to know what "typical person", and distribution of writing was.
In clerical / office work, you'd see a /lot/ of writing, probably several memos, reports, and later email, per week. Easily 100+ documents/year.
Outside that, quite likely less.
@dredmorbius Yeah. I would guess you would see it specialized. It would be interesting to see that broken down.
@dredmorbius @frankiesaxx on this note I sometimes read really old stuff from Google books and one thing I noticed was it was as common for scientists writing for European audiences to write their entire papers in French as it might have been in English, even if neither were the scientists native languages (Alessandro Volta's papers about batteries and electricity for the Royal Society are a good example)
Prior to the invention of the printing press, writing (and copying) books was hugely more expensive than reading. It was far easier to standardise on language, and train readers in that common language, than to produce books in each localised language.
In labour time, a book represented 1/3 - 1/2 a man-year of labour, for the scribe alone. Plus 300 calves for the vellum. About $0.5 - $1 million.
With the printing press, paper, cheaper production costs, /increased literacy/, the equation shifted. Literacy expanded from 5% of the population to 25%, then 90%. It was cheaper to produce works in the vernacular (and print many) than teach everyone Latin (though that happened).
In the 19th century, printing rates increased from about 180 impressions/hr. to > 1 million impressions/hr., per press.
Cast iron, steam, steel, electricity, rotary presses, webbed presses.
@dredmorbius @frankiesaxx although this issue does not change in centuries : https://medievalfragments.wordpress.com/2013/02/22/paws-pee-and-mice-cats-among-medieval-manuscripts/
The social network of the future: No ads, no corporate surveillance, ethical design, and decentralization! Own your data with Mastodon!