February 8, 2023

s14e14: Someone Comes to Town, Someone Leaves Town; Streaks Broken; Aigg on your Face, Big Disgrace

s14e14: Someone Comes to Town, Someone Leaves Town; Streaks Broken; Aigg on your Face, Big Disgrace

0.0 Context Setting

A sunny, clear Wednesday morning in Portland, Oregon, on Wednesday, 8 February, 2023. Sigur Rós' Hoppípolla is on single track repeat and I Just turned over my 15 minute sand timer.

---

1.0 Some Things That Caught My Attention

1.1 Someone Comes to Town, Someone Leaves Town

Today's Thing is an adaptation of something I wrote on the cuff last night¹.

I am still, unsurprisingly, thinking about Mastodon, the application built on top of activitypub that is front-runner for "interesting social network" and a weak signal of the old kind, where a bunch of alpha geeks (itself a problematic phrase now) find themselves. As an aside, finding alpha geeks using something as a bleeding edge indicator of what's next feels like it ran its course at least five years or so ago. Maybe three. In any event, chalk it up to "regular people use the internet now".

Anyway.

I was thinking about all of the "I tried Mastodon and it was terrible" takes, most recently because Paul Thurrot (née Paul WinSuperSite) wrote that Mastodon Continues to Lose Users², a follow-up to a post about him using Mastodon since December³. This is not about Thurrot, this is a reaction to the mass of "I've tried it and it didn't work for me, for reasons" takes, of which I am not going to find and link here.

It struck me that these takes sound a bit like this: "I tried moving to a new big city on my own and it didn't work out", versus something like "I tried moving to a new big city with a bunch of friends".

The point being that I imagine if you tried to go somewhere new without a network of friends and relationships to bed into, of course it might feel like a cold, unwelcoming and confusing place. Going with friends, though, might feel warmer and more interesting.

There's a thought here about migrations of cohorts. In this particular case, Mastodon met a just-good-enough set of features (it exists, it is a bit like Twitter, indeed tongue-in-cheek referencing it, back when posts were called toots). There was enough of something that looked familiar, in other words, that it made it easier for cohorts to migrate, cohort-by-cohort, or group of friends by group of friends. It reminds me a little of the middle class notion of giving it all up and buying some cheap big houses somewhere away from the big city, but doing it with your friends and recreating your own little village or hamlet. Which is a bit like running away, and also very hard to do because it requires a whole bunch of people to coordinate the giving up of everything that's good about urban living. Plus, picking somewhere to move is somewhat difficult.

Which is very related to my point!

There are probably people who tried to move to the Mastodon network from Twitter, or who tried it, already had a bunch of "friends" there, but bounced off. And that reaction is I think entirely understandable where those people already have spaces they share with those same friends. Why move somewhere else if it's just your friends, and your friends are still hanging out with you in the other place?

So perhaps this explains a little of the lurching, collective exodus that's been happening. Again, cohort by cohort, and in limited windows of time. I'll move if you'll move, okay? Oh, you're moving? Great, let's move!

Moving with friends also, I think, addresses why even when there are people who are somewhat confused and irritated about the issue of which instance to join, there are people who aren't, who otherwise might be stressed out about it: their friends have made the decision for them. If you're moving with friends or a cohort, then you just choose the one your friends are on. But if you're trying to move on your own, which in a lot of cases might be because your presence on Twitter is more out of I'm Doing It As A Professional Who Is Expected, No, Required, To Have A Social Media Presence, either through contract or sheer weight of social expectation, then I can imagine there's people who can't just choose the one their friends are on. No, it's a job, and if it's a job, you've got to take these decisions seriously and research them, right?

Cohort-based migrations might be an interesting for a social network to support then, if you could actually do it in practice. You can handwave Metcalfe's law and say, of course the utility of the network is greater on startup if it already has the requisite number of nodes and I don't have to recreate them. Sure, you can take your graph with you, i.e. the mappings, but if the actual nodes don't go with you and there's nobody there in your cohort posting, then you've got a tattered graph with holes in in it.

Another way of looking at this is just asking: was there a time window for cohorts to move? And the bloody obvious answer was: yeah, duh-brain, it was Elon Musk and his acquisition of Twitter over the last few months, from the acquisition announcement, him joining (and not joining) the board, through the court case, and then the actual acquisition completing in November last year.

That external organizing event made it more likely for cohorts to move together in a small time window. It was the shitty landlord coming in. You could even see it: people would talk about moving publicly so of course there'd be more visible coordination and signaling of upping sticks.

When previous social networks have come along trying to upend the existing big networks (e.g. Google Plus having a go and so on), they had to compete on being better than what already existed. Mastodon, I have to say, is marginally better in what exists, for most people, and this is a horrible thing to say because for many people, Twitter is a terrible place, and Mastodon again is not much better. With Twitter's acquisition, it wasn't just Mastodon that got a bump, it was other startup networks like Post.news. But Mastodon wasn't new, it benefited from a bunch of new users in 2017/2018, the last time there was a kerfuffle around how shitty Twitter was being.

What happened this time was a bump to user acquisition thanks to an external event. People weren't choosing to join, they were finding somewhere to leave to. This feels like a trite observation, and perhaps it is. To go back to the landlord analogy, I can take it a bit further:

You were all in a pub with your mates. The barman called last orders. (This is the, uh, nice version). You looked around and made eye contact and decided sure, we'll make this a night, and then decided where to go next. You decided, together, that it was time to go now, rather than when the place actually shut and didn't want to risk your chances on a lock-in.

Sure, this is a big reason why things feel different than last time around. It wasn't just the allure of the new thing, the benefits and differences of the new thing came with sustained usage and as familiarity developed. It was the urgent, extreme change of the old thing, combined with a good-enough replacement that led to groups of people moving over together, I feel.

This isn't to say that there aren't a bunch of people who moved on their own and looked for like-minded people, then found them and made friends. Of course there are lots of people like that! This is to say that I think this time around, groups of people also moved together, and they did it more easily and they did it more quickly.

This is different from influencers or superconnectors or celebrities or those with large followings moving from one place to another, people with whom you aren't really a cohort with. I wouldn't count them as your cohort unless, you were say in the constellation of journalists with The Atlantic and the Journalism Opinion Tweeting Universe. If you weren't one of those people, and you were waiting for them to come over, and you think they're in your cohort then perhaps you're in a parasocial relationship with them? They're not going to move with their friends, I don't think. Their (sigh) alpha geek early adopter friends might already be there, frantically gesturing at them to come along, but the incentives aren't aligned. Moving because a celebrity is moving and going somewhere different is very different, I think, than heading off to the next place with your mates.

I realize that this was just a very long-winded way of saying "groups of people will leave one place for another place when they can no longer tolerate the first place."

1.2 Streak Broken

FitBit, which was bought by Google, because the technology ecosystem can't sustain something the size of FitBit without it being bought by Google, has been having, er, "service availability issues"⁴ which isn't so interesting in and of itself, but what is interesting to me is that it's unsurprisingly affected FitBit users who've lost their streaks⁵. Amusingly (or not, really), the question ("Step streak reset to zero on iOS Fitbit app") is marked as "Answered", because a moderator reported that it has been "reported to our team", which is emphatically not saying that streaks will be restored. That thread shows people who have streaks of 839 days, 500 days and 695 days that have disappeared/reset.

1.3 Aigg on your Face, Big Disgrace

I parodied Sundar Pichai's announcement of Google Bard⁶ by making it about Torment. I should probably write that up somewhere. In perhaps a fuck-up right there with Osborning yourself, Bard made a factual error in its first demo⁷. A point here is that the demo confidently asserts that the James Webb Space Telescope "took the very first pictures of a planet outside of our own solar system", and, whooeee let me interpret it this way: there's no way this demo would've gotten out without going past a bunch of (human) eyes checking it was okay. It clearly got past all those eyes. One reason why, I'm pretty sure, is that LLM-generated text is super easy to glaze over entirely because it's so predictable and we're already framed to predict what it's going to say. Which is a major point of the criticism of these models: they're great until (which is frequently) they assert something that we don't check. Which people didn't check sufficiently. And now you've got AIgg on your face, big disgrace, etc.

OK, that's it for today!

---

I did not finish this in 15 minutes. I finished the Mastodon bit first in about 17 minutes, and then I couldn't just leave well alone, so spent another 10 minutes or so writing those last two points.

How are you doing?

Best,

Dan

---

How you can support Things That Caught My Attention

Things That Caught My Attention is a free newsletter, and if you like it and find it useful, please consider becoming a paid supporter, at pay-what-you want.

Do you have an expense account or a training/research materials budget? Let your boss pay, at $25/month, or $270/year, $35/month, or $380/year, or $50/month, or $500/year.

Paid supporters get a free copy of Things That Caught My Attention, Volume 1, collecting the best essays from the first 50 episodes, and free subscribers get a 20% discount.