Weeks after Elon Musk took possession of Twitter and proceeded to tweet fast and break things, this Scientific American article by Joe Bak-Coleman offers a general theory of why it’s going so badly: social networks are complex systems rather than complicated ones.
On a social network, interactions between individuals create dizzying feedback loops and chaotic interactions that render simple mathematical models next-to-useless for predicting the future, let alone controlling it. Musk’s gamble is that applying his tinkering philosophy to Twitter will take him where no one has gone before: ownership of a large, healthy and profitable social network. The problem is that, unlike the moon or Mars, we have no idea how to get there—and that’s a challenge that engineering fixes cannot solve.
I saw this linked from Mastodon, where I (and now a pretty large fraction of my former Twitter network) have taken refuge. People there are generally passing it around in the spirit of schadenfreude that imbues most discussion of Twitter in the “Fediverse” distributed social network. However, I don’t see much awareness that what Bak-Coleman identifies as a problem for Musk running Twitter is also a problem for a network of mostly volunteer admins running Fediverse-connected Mastodon instances: we don’t actually know how design and moderation decisions will ultimately add up to create the decentralized social network we’re all using now.
The science fiction novel Ender’s Game is best remembered for its primary plot, about a genius child who leads Earth’s forces to genocidal victory against aliens; but it also has a secondary plot line that seems, in retrospect, terrifyingly prescient. While the protagonist Ender is learning to become the greatest space-general in history, his near-equally-gifted pre-teen brother and sister, left behind on Earth, take up politics. Peter and Valentine set up pseudonyms on a global online message board and spar theatrically, building competing followings and eventually real-world political influence. By the novel’s end Peter has leveraged his online clout into the leadership of a worldwide government.
I read Ender’s Game in the mid-1990s, when it was truly science fictional to imagine the whole world connected in a single messaging system, much less using devices portable enough to carry in a backpack. By 2004, my final year of undergrad, I acquired a bulky Dell laptop which was, most excitingly, capable of using the wifi network that had just been installed in my campus apartment complex — and I’d already gone from a hand-coded HTML personal website to a series of blogs hosted on the most obvious choice, Blogger.com. Multiple of those blogs were social affairs, shared with friends, but their connection to people elsewhere on the Internet was entirely mediated by individual “<a href=” hyperlinks. Midway through graduate school, I accepted responsibility for building a website for a conference to be hosted by my home department, and decided to try embedding a new messaging platform I’d heard about: Twitter.
Word is that Twitter is selling out to Elon Musk, whose (speculated) plans for the platform are not especially encouraging. On the one hand, Twitter privately owned by a “free-speech absolutist” may not be appreciably less pleasant for a person like me than Twitter as a publicly traded company with some nominal interest in the experience of users besides Elon Musk. On the other hand, this is as good an excuse as any to take a step back and see if I can, finally, log off.
I’m not deleting my account — not yet — but I’m going to see if I can’t get back to something like my online behavior from the era before Twitter was my first social login of the day. Way back in the Obama administration, I posted to this blog (actually, its incarnation on, yikes, Blogger) multiple times a week. I didn’t break my thoughts up into pithy little snippets, or plan longer discussions in strings of 280-character sentences. I just … wrote.
Late last week I happened to notice that I was following something like 1,400 accounts on Twitter. That seems like … a lot? So I decided to start pruning the list a little. I like Twitter for interactions with other scientists and science-y folks, for discovering new ideas and results and news, and for its overall global water cooler aspect. So with that in mind, I’ve decided to triage who I follow along these lines:
I’m only going to follow accounts that actually update regularly. Because otherwise, what’s the point?
I’m prioritizing accounts belonging to people I know personally.
In many cases, I was following both the official account for publications or organizations and accounts belonging to their staffers/contributors—and I’d get tweets about the same stuff from each. Given the choice, I’d rather follow individual people than organizations; Mark Joseph Stern over Outward.
I’m prioritizing scientists, particularly those in my field.
I’m blanket-unfollowing politicians and political organizations. I read the news; I don’t need links to press releases and official statements in my Twitter feed. And if they tweet something genuinely interesting, I should see it re-tweeted from the “real” people I follow.
I’m blanket-unfollowing parody and joke accounts. Yes, it’s funny to read the latest management tips from Captain Jean-Luc Picard, but I really don’t need a regular drip of them in my feed. As with political feeds, I’m now relying on the actual human beings I follow to show me the best stuff from these accounts.
I’m unfollowing any account if, when it comes up on my feed, I can’t remember the last time I clicked on one of its posts (unless the account falls into one of the priority lists above).
Twitter doesn’t provide any useful way to sort through a, let’s face it, ridiculously long list of account names based on anything other than the order in which I followed them, so I’ve been casting about for a third-party system. The interface at Tweepi is somewhat balky, but it does let me sort the list by how recently each account was updated, which is useful. I’m also simply keeping an eye on my main feed, and unfollowing whenever I see something that doesn’t meet the triage conditions. So far I’m down to … 1,169.◼
King penguins at the Moody Gardens aquarium, Galveston. Photo by jby.
I spent last week in Galveston, Texas, for the Gordon Conference on Quantitative Genetics and Genomics. It was a great time. I saw presentations of lots of cool scientific work that was right in line with my current interests—we could’ve started a drinking game with the phrase “missing heritability“—and I presented a little of my own work, and I met some very smart and friendly people. And … that’s pretty much everything I can tell you about it.
That’s because part of the Gordon Conference culture (aesthetic? ideal?) is that participants present and discuss work in progress—results that haven’t been peer-reviewed, and that may never see the light of peer review if they don’t hold up to further analysis. Attendees want to be able to discuss things freely, without seeing their words quoted all over the Internet. So it’s requested, rather firmly, that attendees don’t photograph Powerpoint slides or posters, or discuss specifics from presentations on public forums like this blog, or Twitter. (So I’ve illustrated this post with photos from an afternoon field trip to the Moody Gardens aquarium, instead.)
Souther rockhopper penguin at the Moody Gardens aquarium, Galveston. Photo by jby.
This is a pretty big departure from my usual approach to scientific conferences. When I was picked to run the website for the 2009 Evolution meetings, I made a point to try and promote online discussion of the conference proceedings. At that early stage—I started my own Twitter account largely to tweet from Evolution—participation was a bit limited. And then I went to ScienceOnline in 2010 and 2011, and saw firsthand how social media can enrich and extend a conference: the tweeted back-channel within a session, the ability to follow goings-on in other sessions, the opening up of discussion to folks beyond the physical conference. And by Evolution 2012, I saw a lot of those dynamics emerging in that meeting.
But back when I was first proposing an online presence at Evolution, I recall some of the same objections raised about blogging and Twitter that are part of the reasons for the ban at Gordon Conferences. People were going to present preliminary results; the prospect of tweeted quotes might stifle discussion; people might get scooped if their work is broadcast over the Internet.
I understand those concerns, but the counterargument to each pops into my head before I finish typing it: it’s no longer uncommon to put preliminary results online for open review (as on ArXive or another preprint server, or in an open lab notebook); are you really going to say something in front of your closest colleagues that you wouldn’t want on Twitter?; aren’t the people most likely to scoop you right there in the conference room, watching your actual presentation? And I think the experience of the Evolution meetings supports the thesis that opening up a conference to the Internet offers far more to improve and stimulate discussion at a meeting than to stifle it.
To be clear, I don’t think it’s bad that the Gordon Conferences ask participants not to tweet. The Quantitative Genetics and Genomics conference is small to the point of being intimate, and strutured to foster all sorts of interactions without the need of hashtags—everyone attends the same slate of invited presentations, everyone ate breakfast, lunch, and dinner together in the hotel dining room, and there are few enough attendees that it really is possible to catch up and chat with anyone you should want to.
Hawksbill sea turtle at the Moody Gardens aquarium, Galveston. Photo by jby.
And here’s the thing: I didn’t write a blog post or tweet about the presentations I saw, but I took notes, and I talked to people, and I’ve absorbed a whole lot of science that confirmed some things I already thought, challenged some other things, and generally changed the way I’ll think about my current and future scientific work and writing. Even if you don’t have a blow-by-blow account of that process, you’ll see its results on this site and the others where I write about science, and in the work I do from here forward.
Which is just to say that even conferences no one tweets about don’t happen in an Internet-free vacuum. One of the first new friends I made in Galveston was Emily Jane McTavish, who I already knew from Twitter as @snacktavish. It’s pretty much inevitable that our next interaction will be online—because, more and more, scientific discussion happens in cyberspace as much as—probably more than—it happens in hotel conference rooms.◼
So I saw Ed Yong do this, and it looked like a good idea: download the full text of everything I’ve tweeted since I started the @JBYoder feed—you can do this via an option under “settings,” now—and make it into a Wordle. As it happens, my very first tweet was back on the 9th of February, 2009, so this word cloud represents just a bit more than four years of my tweets.◼
The worst problem is that shortening services add another layer of indirection to an already creaky system. A regular hyperlink implicates a browser, its DNS resolver, the publisher’s DNS server, and the publisher’s website. With a shortening service, you’re adding something that acts like a third DNS resolver, except one that is assembled out of unvetted PHP and MySQL, without the benevolent oversight of luminaries like Dan Kaminsky and St. Postel.
Since starting up on Twitter, it’d occurred to me that clicking a shortened URL is a pig in a poke at best. Yet somehow it doesn’t feel as dangerous to accept a shortened URL from @apelad or @nprscottsimon as it would if I just found one in my email. Perhaps because, if they did post a hazardous link, I’d just un-follow?