Topos opens up

Today the Topos repository goes public, the first release is tagged, and this site exists. Here is what that means in practice.

What’s online

• The code, under MIT: 2D/3D tracking, PSN, OSC, ADM-OSC and WebSocket outputs, laser-meter calibration, dashboard, a simulation mode to try it without hardware. 296 tests, CI on Python 3.11 and 3.12.
• The anchor firmware and the cases in OpenSCAD + STL, under CERN-OHL-P.
• The complete guide, from ordering the modules to calibrating in the venue, imported here from the repository at every site build — it cannot drift from the code.
• The roadmap, including what we won’t do.

What’s missing

Photos. Printed cases, anchors rigged on stage, a tag on a performer. They’re coming — you have to be in the theatre to do them properly. Videos too.

French and Chinese translations of the guide. English is there; the rest comes page by page.

What we already know has to change

Topos has been running in production at Douze Dix-Huit since spring. Two things came out of the first shows.

Setup still takes too many steps to keep the five-minute promise. We have a list; it will become issues.

Art-Net — driving moving heads directly from Topos — worked through Soma and failed live. Managing fixtures on the Topos side stacks error layers, and it isn’t the job of a spatial translator. We will probably refocus Topos on what it does well, positions, and leave fixture control to the console or to Soma. Nothing is removed until the path is clear.

Why open now

Because it works, not because it’s finished. A neighbourhood theatre that wants tracking doesn’t have the budget of closed solutions, and doesn’t need their complexity. If Topos can serve a stage other than ours, better to know early — and better that this stage tells us what’s wrong.

Questions go to Discussions, venue feedback to a venue report.