Skip to content

How should non-spatial coordinates be handled ? #33

Description

@christophenoel

The current spatial convention already acknowledges explicit coordinate arrays through potential future transform types such as "lookup".

This raises a broader design question regarding non-spatial coordinates.

Today, the convention focuses on the two horizontal X/Y dimensions and affine-like spatial referencing. However, many datasets also include other coordinate dimensions such as:

  • time
  • vertical/depth
  • spectral bands
  • irregular dimensions
  • curvilinear or swath coordinates

The question is therefore more about convention boundaries and composition.

Should non-spatial coordinates be handled:

  • directly within the current spatial convention, potentially requiring a broader scope and maybe a different name,
  • or through a separate dedicated convention focused on generic coordinate systems and coordinate arrays?

If the second approach is preferred, it would be useful to discuss:

  • what the scope boundary between conventions should be,
  • how conventions compose together,
  • and what a suitable name for such a generic coordinate convention could be.

Possible naming ideas:

  • coordinates
  • coord
  • axes
  • dimensions
  • grid
  • domain

I would be interested in feedback on the intended long-term direction before additional conventions emerge independently.

I would typically expect spatial and multiscales to extend a "coordinate" convention: so the multiscale would be clearer for non-spatial dimensions.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions