Fig 1.
Linguistic map of the Lesser Sunda region.
Members of the Austronesian family are shown in brown, Timor-Alor-Pantar languages in green.
Fig 2.
Map of Indonesia, with the locations mentioned in the text.
Fig 3.
→ denotes the target of a reference field (e.g. the Lect_ID reference in forms.csv points to an entry in lects.csv).
Fig 4.
The concept ‘fish’/‘ikan’ in LexiRumah online.
Fig 5.
The Lamaholot-Lerek lect in LexiRumah online.
Fig 6.
Map showing the locations of lects currently in LexiRumah.
Lects are color-coded according to the number of lexical items present.
Table 1.
Number of counterparts for Lesser Sunda lects included in databases covering the region.
Numbers in the table denote the number of concept-form pairs given in a database for a given language or dialect cluster. Where a database has multiple sources for the same lect, this has been reported as the size of the largest source with the number of sources (“src”), except in the case of ASJP, which has a much more uniform wordlist length and many more different lects overall.
Fig 7.
Vocabulary pattern plots for the Alorese lects in LexiRumah.
(A) All 596 concepts in the database, sorted along the x-axis by their number of cognate classes across all lects. (B) Concepts 20–30 and (C) Concepts 530–550 from (A). Every unit bar represents one similarity class in the database attested in at least one AL lect. Similarity classes are coloured according to their distribution in different language groups: similarity classes shared only with TAP languages are green, concepts distributed as if they have a hierarchical history of inheritance are orange, and concepts with classes that are unique for AL are purple. Classes marked yellow have a complicated distribution. Close-ups are provided of concepts 20–30 from the more stable half of the plot (Fig 7B) and of number 530–550 from the less stable half of the plot (Fig 7C).