1. Introduction¶

This document describes how the ExELang Consortium works: who can do what with which data, under what conditions, and who is responsible for what. There are several parties involved or related to the ExELang Consortium, as described in the following table.

1.1 Who is this booklet by and for?¶

This booklet was initially drafted by members of the ExELang team but is maintained by the custodians, to whom questions or suggestions should be addressed. It is primarily intended for custodians and corpus collectors, as it aims to clearly spell out rights and responsibilities. New members of custodians' teams should read, in addition to the current booklet, the internal Custodian Handbook the internal handbook. It can also be useful to inform institutional review boards, legal and compliance departments, and scientific archives.

1.2 Why was the ExELang Consortium created?¶

The ExELang team standardized a large group of datasets during the ExELang grant period. As the ExELang funding was coming to an end, the ExELang team proposed to facilitate a continuation of the positive synergy between people who collect data and people who create tools via the ExELang Consortium. The positive synergy is as follows: custodians benefit because they need data to create their models; corpus collectors benefit because the presence of the dataset in model training and benchmarking increases the likelihood that models will perform well on similar data collected by the same corpus collector in the future.

1.3 Who are the custodians?¶

The current custodians are Okko Räsänen (Tampere University), Marvin Lavechin, and Sho Tsuji.

1.4 Who are the corpus collectors?¶

At the time of writing, we do not know for sure which datasets will ultimately be included, and who are the corpus collectors for those datasets. We also do not know whether other people with similar datasets will be able to join in.