Figure 1
Words for “mother” sampled across different languages in Pallas (1789: 10).
Figure 2
Pulmonic consonants of the variety Chevroux in the TPPSR.
Figure 3
The Phlorest project exemplifies the data curation workflow described here: Once a data type is identified, dedicated curation workflows can be supported with tools such as Python packages. Publication of datasets can then be modeled as “release” in GitHub’s standard workflow and the integration with Zenodo makes sure such releases are actually pushed to a longterm archive for scientific content.
Figure 4
Complex data in etymological dictionaries can be constructed from simpler data types which are already standardized in CLDF. In addition to the tabular data known from cognate coded wordlists the reconstruction tree adds direction and order to the reconstructed protoforms.