In January we held a closed demo for the working group — for the first time we showed an interactive prototype where the key scenarios already work: submitting a problem, forming positions, voting, discussion, forming a working group around a specific topic. This is not the MVP — much of it is done "by hand", some data is hardcoded, and a number of screens remain as static mockups. But for the first time you can see the platform as a whole, not as a set of separate ideas.
The main goal of the demo was to stress-test the concept. We wanted to see whether our decisions add up to a logically coherent whole when a user follows a real scenario rather than reading a presentation.
We revised half of our assumptions about user behavior within the hour-and-a-half demo. It was painful, but that's exactly what it was for.
After the demo, the working group identified three blocks of critical feedback: the evidence interface turned out to be overloaded, the transition from discussion to forming a working group was non-obvious, and the reputation mechanics were too complex to grasp at first glance. A separate mini-project to rework each point has been launched.
What went well, and what didn't
Not all the feedback was bad. Some of the core decisions passed the test and carry over to the MVP without significant changes.
- The issue structure split into "overview / law / states / positions / discussion / timeline" — keeping it.
- The position-voting mechanics — keeping them, simplifying visually.
- The evidence interface — fully redoing it.
- The transition from discussion to a working group — redesigning it.
- Karma and reputation — simplifying the model, keeping only two visibility levels.
A full report on the demo has been published for members of the founding network. A public summary of the key takeaways will appear together with the site launch.