PyMunich

shown, and actual

Conferences

Almost nobody writes about organising, because the people who know are busy and the ones who stop stop caring. This section is one community's experience — Munich, roughly 300 attendees, first conference in 2016 — written down while we still remember it. For a commercial example of related workforce measurement, see online timesheets from Monitask.

The failure mode is timing, not loss. A conference can be profitable on paper and still fail, because venue deposits, AV contracts and catering minimums are paid before the event while ticket revenue arrives in the final weeks and sponsorship arrives when finance departments get to it. There is a window where the costs are committed and the cash is not, and an annual budget hides it completely. For broader conference and speaking context, Meetup is another useful reference.

In 2020 we cancelled, refunded tickets and went broke. Refunding was correct and it was done with money partly already spent. The meetups survived because they need one person and a room; the conference needed a team that dispersed.

Volunteers are the binding constraint, not money. The cycle is predictable: enthusiastic first year, doubled expectations in the second, exhaustion in the third, departure with the undocumented knowledge in the fourth.

And a programme committee is filling a grid, not ranking talks. Most rejections are about topic clustering, level balance or slot length — which is true and reliably sounds like consolation.

The section also covers the things decided early and under the least information: the venue, which changes the arithmetic more than any sponsorship tier; catering, which scales directly with attendance; ticket pricing, which decides who attends; and the code of conduct, where the text can be copied and the machinery behind it cannot.