life beyond googlegroups, or how we un-trainwrecked our listserve

Have you ever been in an online group that had a major pile-on conflict? If you’re alive in 2024… you probably have! I’ve witnessed and heard about so many of these regrettable situations, from professional listserves to social justice groups to massive Facebook groups of like-minded people, like parents of adopted children to people with curly hair to neighbors in a specific geography. The listserve management conundrum has flummoxed me and many of my peers and collaborators for years now, leading some groups to never form listserves, and others to permanently shut them down… sometimes with little or no notice. These same vexing digital communities also facilitate some of the most profound resource sharing I’ve ever known, and in many ways form the lifeblood of communities across space, time, and identities.

The listserve that inspired this post also inspired:

• a sensational Halloween costume
• multiple sermons
• a pervasive sense of doom
• many tears
• much handwringing
• uncountable acts of mutual aid, hesed, and solidarity

This listserve brought out our extremes; is it an exaggeration to say that it brought out our best and our worst? And I don’t only mean the “bell curve” of the community but also, if I’m being truly honest, my inner best and my inner worst.

listserve woes

The sheer volume of the listserve was staggering in scope, whether measured in number of people, number of emails, or number of topics discussed. This led some people to shy away for fear of addressing such a large audience. That being said, many other members loved the scale of the group, and attributed its efficacy to exactly this quality. Because if you are looking far and wide for childcare, plant starts, or a ride, you want to ask the most people possible! This had three unintended consequences, which could be characterized as:

• Tragedy of the commons – overuse of community resource by a small group of people
• Bystander effect – everyone thought someone else would reply/help
• Saturation – too much email volume led people to tune it out, which defeats the purpose of having a listserve all together

I’m choosing to write about the structural issues here, rather than the high drama listserve conflicts. Why? Well, I’m sure you can imagine listserve conflicts of your own. I don’t need to air my community’s dirty laundry. But more importantly, the conflicts represented a tiny fraction of the listserve usage, and they happened a few times a year and typically quieted down relatively quickly. (I don’t want to minimize the stress/agony of these moments, but the previous sentence is still factual!).

More interesting than the substance of the conflict is the technology/people/culture/governance questions that they would raise. Constantly looming were questions of whether we should shut down the listserve (temporarily or permanently), and if temporarily, who decides when to turn it back on? The many listserve hiatus moments were a tremendous relief, but the truth was that we all missed the opportunity to borrow a proverbial cup of sugar from each other, and did not have any viable alternative methods to do so. Sure, the most socially-well-connected listserve members could lean on their friendships “offline” – and to be sure, they did! But, don’t we also want to care for the least socially-well-connected, for whom this community support had become somewhat of a lifeline? This is one of many paradoxes of community building during late stage capitalism.

I just noticed that I’m writing this in the past tense. I should lay my cards on the table… this post is an elegy for a listserve that had tremendous impact, but ultimately outlasted its useableness. And that’s exactly what I want to write about.

the process

Last fall, community leaders decided to indefinitely pause the listserve while we set aside time for a new tool to emerge. Sadly, the listserve had become unwieldy. There were sometimes dozens of emails exchanged per day; I can’t say that all of these were on topics that were squarely relevant to the community. There were many instances of insulting and bullying via email, which created an ethical dilemma for me about whether the listserve was actually promoting antisocial behavior by providing a platform for it. I also observed an equity issue, in that the people who are “chronically online” (hi, it me) would have “dibs” on free stuff, would often become quickly reactive (in an unhelpful way), and could ignite a big conflict before the community’s teachers, food service workers, therapists, and manual laborers had even seen their inbox. These issues were very much top of mind when I joined a committee to research and develop a new system for digital engagement (ew why does that term sound so icky?) that would replace the listserve, retaining as much of the good stuff as possible, while minimizing the harmful interactions.

I know this sounds so trite to readers who hear about software companies and their “values” day in and day out, but I’m going to say it anyway. My community already had 10 guiding values written down and we wanted to use them to influence the technology we selected AND the way we encouraged/enabled people to use it. We ended up forming somewhat of a code of conduct which explains how each of those 10 values relates to the ways we treat each other online. (Send me a message if you want to see the code of conduct, I just won’t post it here to preserve a shred of anonymity for the community I’m writing about). The best, publicly available code of conduct that I found in my research was from Discourse. Here’s a pull quote:

The conversations we have here set the tone for everyone. Help us influence the future of this community by choosing to engage in discussions that make this forum an interesting place to be — and avoiding those that do not.

This has really changed my thinking. I’ve often called this blog, a “lil corner of the internet.” Now, I’m thinking a lot more about how I choose “places” on the internet and how we, collectively, make them “an interesting place to be.”

It was immediately clear to us that this was a technology AND culture problem. But isn’t that always true?! I researched literally 12 different platforms, from the sublime to the ridiculous, but we knew that we would just be moving into a new home with the same, old, furniture. That being said, we also realized that the technology design and capabilities were to some extent shaping (or allowing) behaviors, and if we had technology with a different design and guardrails in place, we would have better chances of cultural change in a positive direction.

We decided that we could not stay on the platform GoogleGroups even though it was in our previous comfort zone. An exhaustive search of listserve/online forum type of software revealed some that were WAY more complicated than we could handle, and others that looked they were developed in 1982. For us, the Goldilocks option was a platform called Groups.io.

in need of a new metaphor

Later in this post, I share more about how we constructed new settings and community norms for Groups.io. But first, I want to write about why we decided that the foundational idea of a listserve where everyone emails everyone was no longer serving our needs.

We had some illuminating conversations where we realized that a platform and norms which support the ability to reply-all to everything results in a lot of (1) noise/clutter (irrelevant emails); (2) when disagreements/conflicts inevitably occur, there is amplification of unkind words; (3) overwhelming amount of replies leads to polarization in over-engagement and dis-engagement.

What if we removed the ability to “reply-all” and instead supported people to privately message someone about a topic?

We started calling this the “Bulletin Board” model. We imagine the listserve like a community bulletin board, where you metaphorically “take a number” if you are interested in contacting the original poster. Moreover, if you don’t find a post compelling, you simply move on (rather than post a response to everyone).

I tried to find a software solution for an actual “virtual” Bulletin Board. I found a few, and I even built one myself, but none had the robust features that we knew we would need. So, we took the core concepts and tried to apply them to existing “forum” or “listserve” style options. In other words, we wanted a system where one person could email many people, but many people cannot email many people in response. Everyone started talking about the “one-to-many, not many-to-many” rule, which means I accomplished my secret agenda of teaching everyone database science (kidding!). Outside of the committee, our peers found this language as clear as a plate of spaghetti. We stuck to the “bulletin board” metaphor and even named the new system [org name’s] Bulletin Board and used the thumbtack emoji as a symbol.

Not only jettisoning “Reply All,” but also, introducing a new model of how we engage with each other, liberated us from the confines of GoogleGroups.

the criteria

At this point, we knew we wanted technology that aligned with our Bulletin Board idea (which was not possible in GoogleGroups). These are some of the criteria that influenced our decision:

• We wanted to slow down the pace of communication. This disqualified platforms that we otherwise love, like Slack or Discord. NO red notification bubbles! No dopamine hits! We also wanted to have a “daily digest” feature to allow people to “catch up” without having to read every post. We made the daily digest feature the default.
• It was essential that the platform “integrate” with email for accessibility reasons, and because it’s just useful. However, we did not want the platform to be ONLY in email inboxes. We wanted it to have a homepage and a browser-based option.
• We decided from day one that we would prohibit the ability to reply-all. This has been our single most impactful decision, and it also disqualified many otherwise great platforms.
• Then there’s the basic stuff. It had to be cheap or free, not too difficult to use, reliable and maintainable over time.

If you want to use Groups.io and model your Settings similar to ours, here’s a screenshot:

Here are some cultural changes that complemented the technology change:

• We created a technology rule that people could not reply-all to previous posts. However, there is no prohibition to starting new posts, which is the whole point of the platform. Therefore, we needed to reinforce our technology change with cultural change. We made a firm boundary that anyone who used a new post to reply to a previous post, especially about a controversial topic, would be removed from the listserve until they completed a repair process.
• We created content expectations for what is appropriate to post. No public shaming, please!
• We created an opt-in system for joining the new platform, consistent with our community value of Consent. (We previously just added everyone!). This will keep the platform limited to people who actually want to be there.
• We created transparent consequences for not abiding by community norms. These consequences do not require moderating/approving every single post, which is not feasible and they are based in restorative justice.

As we developed all of these new tools and systems, we looked for feedback from community members (but not SO much feedback that we would inevitably disappoint people). A small focus group did initial testing and we presented a live demo at the organization’s annual membership meeting. We sent out multiple community-wide emails explaining the process and providing alternative means to continue mutual aid efforts in the absence of the listserve. When it was time for rollout, we created 4 videos to demonstrate how to use core platform features. We wanted everyone to have support as they learned the new platform (which wasn’t particularly hard to use, but CHANGE IS HARD!).

I was afraid that moving from one platform to another would cannibalize everything that we loved about the old listserve. What if no one joined it? What if the same problems carried over? I am glad that I, and we, faced our fears. Life is SO much better on the other side of this big change. We have 176 people who joined as of today (which is about 25% of the former list), but everyone is actively opting in to this new era of digital engagement. I’m enjoying the daily digest of topics, which I can reply to if I have something to add; my replies do not go to everyone; and MERCIFULLY everyone’s replies do not go to me!

In retrospect, I notice that my community fell into the classic Leftist pitfall that is articulated so well in Jo Freeman’s Tyranny of Structurelessness. Without having a clear structure, each person was left to their own to determine how to have listserve decorum. This resulted in some people taking on a “policing” role (yikes!), others adopting an avoidance posture, some people crossing boundaries (unbeknownst to them), others feeling resentful about that, and the listserve becoming unmanageable and toxic. While I don’t think Freeman’s example of cliques fits our scenario as well as a Cinderella slipper, I did mention above that socially-connected members were more able to get their needs met than socially-disconnected members. Informal networks, outside of the listserve, could quickly undermine the listserve at it’s best. At the leadership level, I think we were secretly hoping that the community and the listserve would somehow manage itself out of this problem. It is daunting to use leadership authority to intervene in community proceedings, even though that is absolutely what we are supposed to do. We on the Left, broadly speaking, fear power because it is so often abused. However, in this case it was precisely our aversion to our own power to change the status quo that led to the proliferation of issues. Let me balance this out by saying, it was not ONLY our aversion to power but also that we DID NOT KNOW WHAT TO DO. If we had known what to do (like we do now) we would have just done it, which is why I am writing this blog post!

Against the backdrop of many organizations shutting down dialogue due to inability to constrain conflict, we found a path to expand dialogue in a way that felt safer and more productive for our community members. We used technology to do it, but we also had to take a long, hard look at ourselves and develop new community norms. I think it helped that the listserve was offline for about 10 months!

If readers are just hearing about Groups.io and/or want to see some of my community’s materials, please leave me a comment. I would love to hear from you! It would make me so happy to see more communities move away from GoogleGroups and toward a platform with more customizable settings, which promote the best outcomes of online forums. We deserve to have spaces on the internet where we “want to be.”