GUDA-friendly Data

What does it mean for data to be "GUDA-friendly"?

GUDA-friendly data is open access.

For ASL data, open access means that anybody with an internet connection can see the primary sign data in a video format, along with the annotation files that have made the video data searchable, and with additional metadata about the video data.

GUDA-friendly data is annotated consistently.

Annotation files are created to render the primary video data searchable. In order for the annotation files to be easily searchable, they must be in the same format and follow the same conventions.

What are the GUDA annotation conventions?

ELAN is an annotation tool for video data that has been developed at the MPI. The creators of ELAN prefer that you cite it as software in your published research, or otherwise cite one of their papers.

GUDA-friendly annotation is done in ELAN, and so the resulting annotation file will be in .eaf format. Video(s) are linked to the files rather than embedded in them; you always need a movie file and an .eaf. There are several resources on the web for tips using ELAN, including some that Julie has made.

ASL Signbank is a database that pairs sign forms with consistent labels known as ID-Glosses. The database is constantly growing. Many of the signs can be accessed without logging in, but many more signs are available to those who register (for free).

GUDA-friendly annotation in ELAN uses the ASL Signbank so that the annotation files will be consistent. You can use the "Lexicon Service" in ELAN to link your file to the ASL Signbank. Julie has a video explaining how! ELANport also has a lot of great short video explanations.

The Sign Language Annotation, Archiving, and Sharing (SLAASh) annotation conventions grew out of an earlier project to annotate and share child ASL data. These conventions have also been the foundation for several subsequent projects.

Julie has made a figshare collection with SLAASh information. In that collection, you can find previous coding manuals and posters summarizing previous iterations of the conventions. Julie also has a link to the current annotation conventions on figshare.

The important thing to keep in mind with annotation conventions is that they may change slightly due to the nature of the project(s) and the team's research goals. But we want to balance this with keeping things as consistent as we can. So, please use SLAASh as your foundation!