Skip to content
grid-diamondcutter

Five-minute look

You have five minutes. Here’s what to do.

Step one — clone and run the demo

Terminal window
git clone https://github.com/eir-inc/grid-diamondcutter
cd grid-diamondcutter
python power_grid_sim.py

That’s it. There are no API keys, no cloud services, no proprietary dependencies. It runs on a stock Python 3.10+ install with numpy.

The output you see should match reproducibility_hashes.json byte for byte. If it doesn’t, the repository is broken — file an issue.

Step two — look at the measurements

Open the measurements page. Every row is one published measurement. The columns are:

  • The name of the measurement.
  • Whether it ran on the toy grid (phase A) or on the monetary electricity-market substrate (phase 2).
  • What kind of question it asked.
  • Whether the predicted answer was right or wrong.

About six out of ten predictions were wrong. That is by design. A project where every prediction passes is a project where someone is changing their prediction after seeing the answer.

Step three — read the boundaries

Open the boundaries page. Each row is a claim the project has committed in writing not to make. There are thirteen of them. If a future reader thinks the project is making one of those claims, they can ask the maintainers to explain why.

If those three pass your bar

Continue with the thirty-minute walkthrough, or jump straight to the source on GitHub.