It's been about a month since the deal closed on Twitter, and a sizable portion of the DH community that I count on for collaboration and community began trying out Mastodon. Between travel and a holiday week, it's been a busy month -- figuring out a new platform, helping people find each other, and trying to make conversation in a new space was not what I had planned. But we're up to over 430 people on the DH Mastodon list, and I'm glad for the nudge to get back into blogging, with posts in the first week (resources for getting started), second week (lessons from running IT infrastructure), third week (a sad brain dump about loss), and now this.

I'd meant to write about moderation; it's been interesting to watch the arc of discussion go from setting up servers, to challenges with scaling, and now to the question of moderation. For many Black users, the moderation issues were immediately evident, and the lack of affordances central to Black Twitter were likewise visible. I've been following Dr. Johnathan Flowers' posts (after a month, still can't bring myself to write "toots") on the topic; this interview with him covers a lot of ground, though not the later interjection of what DH folks would recognize as classic hack vs yack warfare. Meanwhile, Jon Bell wrote an interesting thread about some of the features Mastodon has deliberately not included (such as quote-tweets), and how the impact when Twitter added them is not what people may have assumed.

I have thoughts on moderation, after years of running directories (n.b. never run a directory) which are the equivalent of playing the moderation game on tutorial mode. There's some interesting ideas floating around for things like moderators' co-ops with services that could be purchased by server admins. Perhaps next week's musings will be about all that. But I've realized that underpinning so many of these issues -- from moderation to hosting costs to who's running all these new servers -- are a set of questions that we were able to leave unanswered on Twitter.

What are we doing here, actually?

Because the techno-social logistics of Twitter were simply handled by Twitter, we never needed to figure out an answer to that question over there. Sure, those of us who've taught workshops on social media for grad students have had to gesture towards answers if only in the for of "do's" and "don'ts"; Heather Froehlich's "How I use Twitter as an academic" is a classic in that genre, e.g.:

  • Twitter is a space for networking and making friends, but also your seniors are watching you.
  • You like the thing you study, so tweet about what you are doing. Be generous about what you know.
  • Yelling about politics does not make you a better person., but it does make you feel part of a larger culture of dissatisfaction.

Twitter was a space for networking, making friends, surveillance of junior scholars, sharing scholarly process and expertise, and yelling about politics, and many more things. And it could be all those things at the same time, because the company backing the platform saw all those uses as valid, or at least not sufficiently contrary to their goals to be worth cracking down on.

The fact that "Mastodon" (or the Fediverse) is made up of servers that distinguish themselves from one another by their scope or topic already nudges people towards a more defined sense of what they're doing. Why join a regional instance if you don't want to partake in discourse about a specific location (that you probably live in or are from)? Why join a sci-fi server if you're not interested in discussing sci-fi? Why join an academic (or even more niche, a disciplinary) server if you're not planning on using it to discuss -- or watch conversations about -- the field? Many of these places explicitly say you're welcome to discuss other things too, but they may have suggestions (or requirements) around the use of content warnings for certain topics, including US politics.

What are our academic infrastructure resources for?

"You can keep the conversation going on Twitter" was an easy thing to say at a conference; it was a space that (in its own, and sometimes-problematic way) was maintained for us, at no direct financial cost to us. We're approaching the 10 year anniversary of the defunding of Project Bamboo; I've seen firsthand how expensive it can be to plan, let alone implement, let alone sustainably run scholarly infrastructure for the humanities. Twitter worked (for a given value of "worked") for academics across a huge swath of fields, from epidemiology to linguistics to history, not to mention many people on the border of academia and industry, or academia and journalism. Without that kind of centrally-run infrastructure, we're left trying to figure out who's willing to run what, and who can find resources, from where, to keep these individual servers going.

I think scholarly communication within and across fields is essential. But I also came of age in DH in the era of Twitter, which has shaped so many of my projects and the way I've gone about trying to build community within the field. Is it as important as I believe it to be, or is it just that my way of being within academic spaces has evolved for the tropics, but we're now facing an ice age?

Academic infrastructure is not well funded or supported, and the pool of funding tends to be smaller the more specific the discipline: is this how we should be spending what little we have? Is this how those resources would do the most good? If I had to choose between sustaining Voyant and a Digital Humanities Mastodon server, what would I pick? What are we doing and are we willing to give up to keep supporting it?

The "right" answer here is that instead of fighting over scraps and crumbs, we need to advocate for a bigger pie. We need more resources to do this work well and should fight for more funding for academic infrastructure. And while I can't disagree in principle, 1) that's a long-term dream; and 2) it feels like it's willfully overlooking the current landscape of massive funding cuts, especially in the humanities -- we're lucky anymore to not lose whole programs, and that feels like the place one would need to start with repairing damage already done.

Who wants to fund community infrastructure?

So where does this leave us now? There are grant programs and funding schemes and whole meta-infrastructures like DARIAH that we could appeal to with the pitch of funding Mastodon servers as a form of core scholarly infrastructure. But community-oriented infrastructure is always a harder sell than its more technical counterparts, especially technical infrastructure that leans on the buzzwords du jour, like AI or kubernetes. The more clearly you can frame the server as a platform for academic collaboration, knowledge-sharing, dissemination of resources and announcements, the easier it would be to fund... and if you were to actually moderate in a way to keep the conversation within those bounds, people would bail for other, less-restrictive spaces.

Who pays, then? Will the Patreon/Ko-fi/etc. individual one-off and recurring donation model that other Mastodon servers have used for years work here, too? Maybe at a certain scale, but large-scale is at odds with the federated-small-community concept. What has happened in the last couple weeks in academic servers has had the opposite momentum, towards more and smaller fragments. There's hcommons but also h-net; 18th centuryists were there early, but recently an instance for historians has made its debut. Even before factoring in the costs of paying moderators, it's hard to imagine that there are simply enough people with enough left over at the end of the month that they'd want to make an adequately large recurring donation to support the servers. Many of these academics have research funds, but it's hard to sell faculty on directing those funds towards paying for common-good infrastructure instead of things specific to their own research project. (Trust me, I worked on setting up HPC at a university once.)

My own perspective here might be shaped by "doing Mastodon wrong" by being on a large and general server, but I find myself wondering if we would be better off devoting community infrastructure resources to developing guides, tutorials, and supporting the ongoing development of software, rather than the literally endless task of running servers. Because, at least in principle, reputable Mastodon servers federate, on a certain level we don't need -- and perhaps shouldn't encourage -- proliferation of hyper-specific servers. Give pointers to well-run servers and encourage people to go there; the community doesn't have to be the server, and the server doesn't have to be the locus of community. Build a community around a hashtag, or an a.gup.pe groups. Connect people, get them talking to one another, wherever they're connecting through. A Mastodon account is on some level a portal, there are many places people can get one, and just as academic organizations (let alone sub-fields) don't try to set up their own ISPs, maybe some of the same principles could hold true here? The value is in the communication, the content, the connection with others -- not how you log into the place where that happens.

Where are we (going) now?

It's hard to imagine what sum of money you'd have to give me to talk me into running a Mastodon server, and I think it's worth noting here that it's not the technical-staff-at-DH-center types who have been rushing to set them up. I have no horse in this race, other than wanting to see this whole endeavor of a community migration away from Twitter to work, if only for the sake of my own highly-interconnected workflows.

I think it'd help if the people running academic servers talked to one another. It's a long (honestly, maybe endless) road ahead; getting some sense of good practices, what options are out there, how well different approaches are working, and potential pitfalls others have experienced, strikes me as the sort of thing I would want to do if I found myself in the unenviable situation of running a server. So I've written up a survey along those lines.

Because I'm hoping for responses from old-time Mastodon admins along with new, I'm struggling through a rather painful to use Drupal 7-based form-creation platform, out of respect for the privacy-oriented sensibilities common around Mastodon. Perhaps it's just further penance for Drupal for Humanists. I'll be curious to see what kind of uptake I end up getting, and if it leads to anything useful. But better models for scholarly infrastructure don't simply come into being. I don't have the mandate (or the money) to return to my DH roots with grant visions of the future of cyberinfrastructure a la Project Bamboo -- nor would I want to -- but trying to nudge along conversations feels like a small and feasible step towards something better, at least we can hope.