Individual tiers

At times some tiers are added into ELAN files just to be used in one study by one person or a small research group. Thereby when opening a file it is possible to find tiers which are not documented in the main template, and which may be somewhat idiosychronic. Often there are reasons why they are set up the way they are, but ELAN itself allows only little documentation of the purposes of these tiers.

Main idea here is that it has to be possible to understand every tier in ELAN file, and this information needs to be both human and machine readable. It is not OK just to add a random tier and linguistic type to an ELAN file -- those always have to be specified somehow somewhere, otherwise the files become impossible to use.

Niko Partanen

study: izva_sibilants
documentation: http://nikopartanen.github.io/izva_sibilants
prefix: sib@
parent_prefix: orth@
type_name: sibT
constraints: Included_In
time_alignable: true