NE Scala in the Time of Social Distancing

A non-Dotty-related topic today: the NE Scala 2020 conference wrapped yesterday. This is an annual shindig, a co-production of the New York, Boston and Philadelphia Scala Meetups, and has been running for ten years now; it’s become a highlight of my year. This time, it was the New York team’s turn to take the lead.

It is really three conferences on three successive days. The first day is the Typelevel Summit, a set of talks curated by the Typelevel community. Day two is NE Scala proper, where the ticket-holders get to vote on which talks they would like to see. And day three is the Scala Unconference, a bit of wonderful emergent behavior that starts with a blank spreadsheet of rooms and time slots: folks grab slots to talk about whatever Scala-related topic they are passionate about, and everybody breaks out into the sessions that interest them.

It was broadly successful in its planned goals — we had a great lineup of talks as usual, and the conference’s long-term objective of supporting diversity in the Scala community was reflected nicely in the lineup of talks and speakers. (There’s a long ways to go yet, but this year was significantly better than most tech conferences in that respect.)

It was also successful in the not-originally-planned goal of still having a conference in the face of a sudden health crisis. As of a week ago, we were still confident that things were going to run as expected, but matters began to evolve even faster than I expected. Over the course of Tuesday, my own viewpoint shifted from “of course we can do this” to “this is starting to look irresponsible”; by the end of the day, we had a rough consensus that we needed to change tack quickly, and shift to a purely online conference. By the time things started on Thursday (and even more, by the time we finished on Saturday), as the term “social distancing” suddenly made its way into common parlance, it was becoming clearer that we had made the right choice.

Most of the credit for pulling this off goes to Ryan Williams, this year’s conference chair, who did an amazing amount of work in putting together both the original plan and the changed one.

Which brings me to the main point: he has posted a debrief of the conference that I would highly recommend to anyone who is trying to navigate these waters. We wound up using a combination of Zoom and Slack to make it happen, and while it wasn’t the same as the in-person conference would have been, it worked solidly well for the talks and adequately for the socializing. I’m sure that it’s possible to improve on the way we did things, but it worked remarkably well for such a last-minute plan, and it serves as a good proof of concept that this sort of thing can work, at least for some sorts of gatherings.

With any luck, next year will be back to in-person in Boston, but there’s a lot of food for thought here. While all-online wasn’t ideal, it did have a lot of advantages. We lost some members who were unhappy about the lack of the physical venue, but wound up gaining about as many who hadn’t expected to be able to be there in person anyway. For next year, we’re starting to think about how a hybrid modality would work — not just live-streaming the talks, but trying to be more broadly inclusive using online tools. If you’d be interested in helping to figure that out, please ping me and let’s talk in the coming months. I think that, in the long run, we can come out stronger due to the lessons learned this time around.

In the meantime: I hope you all stay reasonably well through this crisis, and that I’ll have more chances to meet y’all at future conferences…

Lifelong programmer and software architect, specializing in online social tools and (nowadays) Scala. Architect of Querki (“leading the small data revolution”).