Sciveyor began as two separate projects: RLetters, a Ruby package that was responsible for the website and analysis code, and evoText, a particular installation of the RLetters software with a database of articles drawn from “the evolutionary sciences,” broadly speaking. That project began in 2009, and was initially hosted at the University of Notre Dame, then at Louisiana State University, and finally followed Charles Pence to the Université catholique de Louvain.
In 2020, we made the decision to combine the RLetters and evoText projects, and rebrand them as Sciveyor. There were a few reasons for this. First, evoText as a project was originally limited in scope – we intentionally made some choices not to include journal articles that fell outside of the evolutionary sciences. That made sense at the time (including to our funders), but as we expanded the corpus, we began to realize (and hear from users!) that the presence of general-science journals in our corpus, such as Nature and the catalog from PLoS, meant that evoText was a useful tool for scientists, philosophers, historians, and others interested not just in the uses of evolution, but in analyzing any number of sciences. The name “evoText” had thus become too narrow, and tended to make users underestimate how useful the site could be.
Second, it was becoming confusing to separate “RLetters,” as the software that powered the website, from evoText itself. That software has also radically changed, and will soon allow developers to contribute in a variety of languages, not just Ruby. In that sense, the name of RLetters had also gone out of date.
Since 2020, then, Sciveyor is both a piece of software and a website. If you would like to build your own corpus of journal articles and analyze them in the same ways that we do, you can find all of our source code at https://codeberg.org/scieveyor, and you can read more about how to contribute to our development process here in our developer documentation.