On Replacing Science Online

That's "Science Online" as in the conference that folded, not "science, online" as in the practice of trying to understand the universe from in front of a networked computer. Specifically, I'm posting about David Zaslavsky's call for help in putting together a replacement meeting. There was a lot of talk about this right when Science Online went under, but that's pretty much died down, at least in public. David's trying to get something more active going.

This is, of course, a massive undertaking, and something fraught with peril. And it's not like I have any free time to make really constructive contributions on matters logistical. (254 days.) But in the fine tradition of keyboard commandos, I'll throw out some comments about what I think ought to be done.

As I've said a few times in different forums, I think there are some inherent contradictions in the whole business of what Science Online became, that contributed to its downfall. The biggest and most problematic is the tension between "professional conference" and "party with friends"-- somebody on Twitter referred to this issue as needing to make a decision as to whether it's a conference or a convention, and I think that's a good way of putting it.

This tension plays into most of the stuff that was awkward about Science Online: the regularly praised "intimate" feeling of the meeting demands a restricted attendee list, which led to the crazy "register within five minutes of a particular time or go into the lottery" system. The desire to have certain specific people there also restricted the program possibilities, and efforts to broaden the participation by limiting the number of items for specific people either further reduce the space available, or lead to some big names not being there, which then reduces the attraction for people who want to see them. Needing to know that the right people will be there demands setting a lot of the program in advance, but that undercuts the "unconference" thing. And of course, limiting attendance drives up the cost per attendee a bit.

The final issue was that the meeting was very much the creation of a small group of people-- Bora, Anton, and Karyn-- and very strongly identified with them. Which made it really hard to recover when those specific people did things that offended or disappointed the core community.

I don't think it's possible to perfectly replicate what Science Online was, and I'm not sure it's advisable to try. That is, a lot of the things that people cite as the key positive features are also key factors in its eventual downfall, and trying to duplicate the glory days is likely to lead to the downfall of a successor organization in a manner that's broadly similar to the original.

These are, of course, problems faced by all sorts of organizations that hold regular meetings, so in a sense they're the Previously Solved Problems beloved of mathematicians. And there are three general classes of solutions, each with pros and cons. One thing to do is just to own the exclusivity-- make it a pure invite-only Event, with a select cast or rock stars and everybody else chosen by lottery, and be the TED of online science communication. Another approach is to move toward a more formal membership organization kind of model, like the American Physical Society and others, where you make the meeting as big and formal as it needs to be, and much more of a professional conference. I don't think there's a great deal of enthusiasm for that, though.

The door number three option is to go the "convention" route in the sense of SF conventions like Worldcon. This is a more amorphous kind of structure, rotating around between various local committees who put the meeting on one year, then hand matters off to the next group. With a behind-the-scenes crew of volunteers who show up pretty much every year and handle a lot of the logistical stuff that requires consistent experience.

This has its problems, to be sure-- there's a lot of variability in the quality of the experience from year to year-- but it fixes a number of the problems associated with the original Science Online without completely sacrificing the informal aspects that people really liked. If something goes wrong one year, well, next year it's an entirely different bunch of people, who won't make the same mistakes the same way. And by being largely volunteer-driven, you have a bit more freedom to bring new people and ideas in.

Of course, the down side is that being largely volunteer-driven makes it a little difficult to plan too far in advance, because you need a different group of people to step up each year and do the insane amount of work needed to put one of these things on. The big advantage of the original small-group model (or the big membership-organization model) was that you had a fixed base, and the same set of folks who knew what they were doing putting the meeting on every year. I'm not sure that there's the necessary community support to do that, though, especially starting essentially from scratch.

Anyway, I think that's a model that's probably worth considering-- don't look for a single group or location to put on a successor meeting, but try to find 3-4 groups to take turns hosting future meetings. But that's a much harder thing to do, so... As I said, it's a Hard Problem.