/Glottolog_look_up_table

Primary LanguageR

Glottolog_look_up_table

This is a small repos with an R script that cobbles together a large tsv sheet with every languoid in Glottolog (i.e. languages, dialects and families) with various meta data. Besides the "regular" Glottolog meta data (Family, long/lat, Macroarea, Countries, etc) there are also some extra info and modifications, see lists below.

Basic meta-data from Glottolog

  • Longitude/Latitude
  • Level (language/dialect/family)
  • Macroarea (Australia, South America, North America, Eurasia, Africa, Papunesia)
  • Name
  • ISO 6393-3
  • Glottocode
  • Parent-ID
  • Top-genetic unit ID
  • Path (from languoid to root of tree)
  • Countries

Added meta-data

  • AUTOTYP-area (based on this csv: https://raw.githubusercontent.com/autotyp/autotyp-data/master/data/Register.csv )
  • Name_stripped (ASCII and stripped for certain interprunctiation that certain programs, for example SplitsTree, struggles with)
  • Family_name
  • Isolate have a family name and family ID (the name and glottocode of the language) and there is a separate column that distinguishes Isolates from non-Isolates ("Isolate")
  • instead of just dialects having an ID for their parent that is a language, languages also have their own IDs as the "language_level_ID"
  • all dialects inherits meta-data from their language-level parents
  • Family_color is a distinctive colour in HEX code for every family, including seprate ones for each Isolate

todo

  • make it possible to run the python lines within R
  • deal with the "non-genealogical families" like "Sign language" etc so that they are represented in a different way (alongside the old style), see outline here: glottolog/glottolog#318 (comment)

How is x determined?

I'm just reshuffling Glottolog and AUTOTYP, I'm not making decisions about languages. If you want to know how some of these things are determined, see the list below: