[Places] Quality of data project update?

Discussed in https://github.com/orgs/OvertureMaps/discussions/206

^{Originally posted by danmanix August 20, 2024}

Category Feedback

Add details

• Expected behaviour you didn't see
  Clean, and approaching definitive listing of less transient places (ie, more permanent places like airports/ galleries over restaurants.

• Unexpected behaviour
  A lot of duplication/ inconsistent naming in the data of some of the larger places (airports, see below). This seems to come from multiple sources, where perhaps they aren't as interested in data quality as, say, OS, but data from OS seems to be missing.

• suggested functionality
  -- Just looking for an update in the data clean-up that was announced in June? I know it will be a massive job!!, but transport hubs are key to my project :)
  -- Where are the other data sources?
  -- Should there be differentiation between passenger/ cargo/ private jet terminals in the categories? (this sort of thing may apply to other sub-cats).

• Steps to reproduce the problem
  The below query will produce a lot of duplication, inconsistent naming conventions, places incorrectly marked as an airport/ terminal, incorrect/ incomplete addresses and more.

SELECT addresses[1] as address, names.primary as name, categories.main as category, sources[1]['dataset'] as src
FROM read_parquet('..../release/theme=places/type=/', filename=true, hive_partitioning=1) WHERE addresses[1].country='GB' AND names.primary ilike '%heathrow%' AND categories.main in ('airport','airport_terminal');

Dependency with other categories, if any.

None