openstreetmap-data-intent, proof of concept

[early-draft] See https://docs.google.com/presentation/d/1V3gDPh5uzmIYtuPptFJY1GbfbvnxW_ViWuIsnSr9bhQ/edit?usp=sharing

• Installing
• Command line Usage
  • Quickstart
• The proto specification for OSM data intent file
  • Reference codes at top level
  • 1: input data
    • 1.1: natural language expalantion, from developer to developer
    • 1.2: natural language example of averange user query
    • 1.3: OpenStreetMap tags for select data
    • 1.4: OpenStreetMap tags for exclude data
    • 1.5: OpenStreetMap Wikibase Data Item
    • 1.10: Wikidata Wikibase Q items
    • 1.20: operation inside place
      • 1.20.1: operation inside place, parameter natural language
      • 1.20.2: operation inside place, parameter OSM tagging
      • 1.20.3: operation inside place, parameter OSM relation
      • 1.20.4: operation inside place, parameter Wikidata concept
  • 2: environment data
    • 2.1 prefered natural languages to restrict output if too many options
  • 3: output data

Installing

At the moment, this package is not on Python Pipi, so install from GitHub.

# pip install osmdi --upgrade
pip install git+https://github.com/fititnt/openstreetmap-data-intent.git#egg=osmdi

Command line Usage

Quickstart

osmdicli --help

# test with examples from tests/data
osmdicli tests/data/example-005.osmdi.yml

The proto specification for OSM data intent file

The general idea is based initial attempt to implement idea behind https://docs.google.com/presentation/d/1V3gDPh5uzmIYtuPptFJY1GbfbvnxW_ViWuIsnSr9bhQ/edit?usp=sharing.

Some early comments (2023-09-02) on how to do it (subject to change):

• Focused on OpenStreetMap data and tooling.
  • and it accepts input data from non OpenStreetMap data (for now, Wikidata) because allows to be used to explain relationship between concepts
    • yes, this means tools could explain how to dump data (for now, only SPARQL implemented) external to OpenStreetMap
    • the semantics on how to compare data does not have code implemented

Reference codes at top level

For files with examples, check tests/data.

• 0: metadata (optional)
• 1: input data
• 2: environment data (optional)
• 3: output data (optional)

1: input data

1.1: natural language expalantion, from developer to developer

1.2: natural language example of averange user query

1.3: OpenStreetMap tags for select data

1.4: OpenStreetMap tags for exclude data

1.5: OpenStreetMap Wikibase Data Item

1.10: Wikidata Wikibase Q items

1.20: operation inside place

1.20.1: operation inside place, parameter natural language

1.20.2: operation inside place, parameter OSM tagging

1.20.3: operation inside place, parameter OSM relation

1.20.4: operation inside place, parameter Wikidata concept

2: environment data

2.1 prefered natural languages (to restrict output if too many options)

3: output data

License

Public domain