ror-community / ror-roadmap

Central information about what is happening at ROR and how to contribute feedback
10 stars 2 forks source link

[FEATURE] Reconcile lat/long location data in ROR records with Open Street Maps #172

Open amandafrench opened 1 year ago

amandafrench commented 1 year ago

Describe the problem you would like to solve OpenStreetMap has a tagging system for university and college locations -- see https://wiki.openstreetmap.org/wiki/Tag:amenity%3Duniversity. Many of these universities and colleges undoubtedly have ROR records, but the OpenStreetMap latitude and longitude information about their locations is not in ROR.

Describe the solution you'd like A crosswalk and/or reconciliation between ROR lat/long values and OpenStreetMap, potentially via Wikidata, which also has lat/long information.

Who would benefit from this feature? Users who want to use ROR and/or OpenAlex data to map research institutions.

Additional information Add any other information or screenshots about the feature request here.

amandafrench commented 1 year ago

Have also heard from ECOOM that they would very much appreciate having a precise location for organizations in ROR records. Currently, the lat/long from Geonames only points to the city, not to the exact location of the organization itself.

mariagould commented 1 year ago

@amandafrench what are the use cases for the precise coordinates as opposed to the city-level location?

cameronneylon commented 1 year ago

@mariagould We (COKI) quite frequently deliver information in the form of maps, often relating to analysis of collaborations between organisations. This works great for long distances but not so much for within city analyses (which are of interest in terms of engagement and impact - particularly showing relationships between research orgs and stakeholder groups). There might be a case where resolution would be too high (although doing within institution maps could be interesting) but roughly campus/district level would be useful for many of our cases.

mariagould commented 1 year ago

Thanks for sharing this use case @cameronneylon - very helpful to have this context!