The initial version of the data schema, with the basic structure and some complex functionalities for each class, including normalization. This will be mainly focused on electronic structure calculations, but we plan to implement functionalities for handling atomistic simulations.

This initial version includes:

• The main data section Simulation with the sub-sections ModelSystem - ModelMethod - Outputs.
• The hierarchical ModelSystem implementation, and all the important sub-sections relevant for system: ChemicalFormula, Symmetry, Cell.
• AtomicCell(Cell) and its main quantities and sub-sections: AtomsState and OrbitalsState.
• GeometricSpace.
• DFT, GW, BSE, DMFT inheriting from ModelMethod.
• NumericalSettings as a sub-section of ModelMethod and which can be: Mesh, BasisSet, SelfConsistency
• Outputs and SCFOutputs(Outputs) as a way of parsing self-consistenly calculated properties
• A base class for PhysicalProperty and Variables. This implementation depends on Area D and affects directly the main NOMAD and the FAIRmat taxonomy task force.
• A list of output properties and their normalizations: ElectronicBandGap, etc. Each lives as a sub-section under Outputs.
• Add extensive unit testing for everything, to achieve a good coverage (min 90%).
• Add this plugin as a default to nomad.

The implementation leaves out some todos, but mainly we will bump the version after the initial release when we:

1. Start implementing the schema in parsers, and come back to refine the definitions and functionalities.
2. Create a documentation pages and receive feedback from PIs
3. Exchange feedback with Areas A and B, and decide which classes must be moved to the main nomad src code under nomad/datamodel/metainfo/basesections.py.

Before releasing a 1.0.0 version, we need to:

• Control normalization in the main parsers (VASP, FHIaims, exciting, ABINIT, YAMBO, w2dynamics, solid_dmft, GROMACS, LLAMPS...)
• Have a clear documentation on how to use this
• Finish todos (TODO) and fix warnings (!) comments in the code

For a future release, we will:

1. Map data into results.
2. Produce plots of our properties in a flexible way.
   This will mainly depend on changing the NOMAD src code.