0

As exposed here, the Google PlusCodes (plus.codes) resolution service is not reproducible.

A resolution service does a conversion from a namespace1 ID to other namespace2 ID. As ID we can suppose any kind of formal identifier, since a serial alphanumeric code to textual meaningful name, or mixed codeAndName. For example DOI.org is a resolution service for DOI identifier, converting DOI into URL.

To be open, the namespace must be an open dataset (e.g. the set of names of Geonames.org), or an open algorithm that encode/decode compact code to open standard like geographical coordinates of latitude and longitude (see GeoURI open standard).

PlusCodes is a resolution service that uses OLC technology for resolution of an OLC's ~10 digits geocode and (~6 digits) partial code. The OLC's partial code is part of a codeAndName identifier, where the name part is resolved by Google's location names database.

At this moment (February'2020), there are no open way to reproduce the PlusCodes service for codeAndName resolution, there are no market competitor and no open dataset for the name part. Even for a very simple and reduced scope, the official Cape Verde resolution. There are no open lookup table with PlusCode's names of Cape Verde, enabling a competitor service to run the OLC's shortening algorithms. Examples of "PlusCodes black-box" conversions:

  • The point 796RWF8Q+WF is converted into "WF8Q+WF Praia, Cape Verde"

  • The point 796RXC4C+9F is converted into into "XC4C+9F João Varela, Cape Verde"

  • The point 797R3F5Q+HW is converted into into "3F5Q+HW Cruz do Gato, Cape Verde"

  • ... and the inverse conversion, from "WF8Q+WF Praia, Cape Verde" to 796RWF8Q+WF, etc.

No way to reproduce the full behavior of PlusCodes services in this kind of conversions... Even using open databases like OpenStreetMap or Wikidata, it is impossible.

Google says that "Plus.Codes is an open technology", but, is it?
Is it open from the Open Data Community point of view?

This is the question... How to say that a resolution service, based into open data and open algorithms, is a "full-open service"?


Note

Maybe it's a broader question... For this open data discussion community,
the definition of "open data service" is flexible? Something as

Def1. Digital open data service is a service based in open data and open algorithms
(that can be "full open" or not by using some black-box complements)

or the a real open data service must to be a reproducible service?

Def2. Digital open data service is a service based in open data and open algorithms, and all service behaviour is reproducible, using the same data and algorithms.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.