PyMunich

shown, and actual

Free Labour

A community conference is built from unpaid work. So are the talks in it, most of the libraries discussed, and the documentation everybody reads. For a commercial example of related workforce measurement, see the 7-minute rule for payroll from Monitask.

That arrangement produces a great deal of value and it has a distributional consequence that is worth stating: who can afford to work for free decides who is visible. For independent technical background, PyPI provides additional documentation and examples.

Where the hours come from

Organising: roughly a year of evenings from four to six people, which is the constraint that ends most community events.

Speaking: twenty to sixty hours per talk, plus travel.

Maintaining: open source packages that the entire programme depends on, largely maintained by people doing it around a job.

And reviewing: two hundred proposals read by volunteers in evenings.

None of it appears in any budget, including ours, which is why a conference that breaks even has actually consumed a great deal of value that nobody counted.

Who can afford it

The part that matters.

People whose employers pay for it. Some companies give conference days, sponsor travel, or count community work as work. That is generous and it means those employees can participate at a level others cannot.

People without caring responsibilities. A year of evenings is available to some people and not to others, and the difference tracks gender and life stage closely.

People with financial slack. Travelling to speak, unpaid, requires being able to absorb the cost of two days not earning.

And people already established, who get invited rather than applying, and for whom the marginal talk is cheaper.

So the visible community is systematically composed of people who could afford to be visible, which is a different thing from the people doing the work — and it is invisible from inside precisely because everybody in the room passed the same filter.

What can be done about it

Concrete, and none of it solves the problem.

Travel grants, funded from the corporate sponsorship tier. The single most effective intervention per euro.

Speaker support as standard, not on request. Asking is itself a barrier.

Bounded organising roles, so participation does not require a year of evenings. Six hours a month is available to far more people than "help run the conference".

Childcare, and saying so on the website.

And remote participation, which is the cheapest way to include people who cannot travel at all.

What organisers should be honest about

You are asking for unpaid work, and you should say so plainly when recruiting — including how much, in hours, and over what period.

Vagueness about the ask filters for people who can absorb an unknown commitment, which is the same group the rest of this page describes.

And thank people specifically. It is not compensation and it is the minimum.

The tension nobody resolves

Paying people would change what these events are.

Paid organisers means a budget several times larger, which means higher ticket prices or commercial sponsorship at a scale that changes the programme — which is the thing a community conference is not.

So the free labour is load-bearing and its costs are unevenly distributed, and both of those are true at once. Nobody in this ecosystem has resolved that, including us.

What is available is being honest about it, and spending the money you do have on the barriers rather than on the event's appearance.

The short version