Conferences: unsolicited advice

Well, that was quite the week!

As the last few posts had possibly clued you in, we hosted a conference in Leeds this past week. We had around 50 participants, including the speakers and poster presentations. I am very grateful for everyone who came here, those that spoke and did such a wonderful job, the posters that were so fascinating, and the audience that asked questions. I'm really grateful to Philipp Schilcht for his wonderful tutorial, filled with interesting open problems, and very cool results.

Throughout the week many people said that this was really great. I enjoy organising these sort of events, and I wanted to offer some unsolicited advice to the future organisers of events such as this.

1. Plan in advance.

Have a vague idea of how many talks, people, and funding situation you want to have. Try to make a mock schedule to see what fits, then plan around that. If you start inviting people about a year in advance, it's easy for people to be free and actually show up. It's also easy to apply for additional funding far enough in advance so you can accommodate some additional support, etc. and if you have a concrete plan it's easier to write better applications.

2. Mix, mix, mix.

Make sure that you have lots of early career speakers. I am so happy that half the speakers in this conference were at the very start of their academic journey. Everyone had such interesting things to say. And it was lovely to hear Natasha Dorbinen and John Steel present their work which is built on many years of experience. On that note, try to aim for a 50/50 split in gender, and try to make that split as uniform as possible across seniority. I got a lot of wonderful feedback about how many young people showed up, as well as how well-balanced the speakers were. I am very glad that worked out for me, and it is not hard to make it work out for other people.

3. Don't overstuff your schedule.

Some time ago I was invited to a conference about realisability (after my paper on Krivine's work was out), and while I was happy to go there, it turned out that all the talks were in English solely for my sake and that the conference always run in French. That made it that I had to listen to every single talk in full details (side note: if you do that, trying to ask a question every single talk is an easy way to keep yourself interested), and especially since I am not an expert on realisability, quite the opposite, this was extremely hard. In a conference, people get bombarded with new information, new concepts, new results, etc. It is a good idea to let people have longer breaks and later mornings so they can rest between talks. This will "cost you" with a lower number of speakers.

We had 15 talks + 4 lecture tutorial + poster session, so an overall of 20 slots. Over five days. That is incredibly sparse. But people go to conferences not just to sit and hear talks, but also to meet colleagues, work out new ideas, discuss on-going work, have a late night at the pub at some point. If you have five days of 8 lectures a day, that is an insane amount of information to the point that people are exhausted and skip some nontrivial percentage of the talks.

The downside, of course, is that you can't invite everyone. That's unfortunate, and many times, people won't show up if they're not invited to give a talk. Either because they don't have funding (or that their funding requires them to present) or maybe they were already invited to two other conferences that season.