Home
This guide is designed to be a starting point or model for researchers planning to develop their own web-based corpora. It documents the software design, and deployment process for, Crow, the corpus & repository of writing, which is located at https://crow.corporaproject.org.
The software discussed here is not a plug-and-play end product. Making it work with your corpus data requires significant configuration, and assumes access to substantial knowledge of both web development in general and Drupal and Angular development in specific.
Read the executive summary to determine if this approach fits your corpus goals.
The dataset that makes up the Corpus and Repository of Writing (Crow) is a large-scale learner corpus of English writing samples from university foundational writing courses, as well as pedagogical materials used in those courses. It is designed to contain tens of thousands of individual texts, searchable by word, phrase, or metadata.
- Overview: the website backend
- Importing corpus texts
- Importing repository materials
- Deploying to a server
- Updating software dependencies
- Performance considerations: What to expect for page load times
- Email integration
The user interface for the Crow corpus Interface Design is designed to caterer to multiple audiences -- corpus researchers, writing teachers, and students. Registration is required, with different tiers of access.