WORK IN PROGRESS This is a tool to help pseudonymise site content for research purposes. Pseudonymisation (https://en.wikipedia.org/wiki/Pseudonymization) differs from anonymisation in that identifying fields are replaced with human-readable artificial identifiers. Data may also be altered slightly to further distance research data from source individuals. TO DO: Data fields (e.g. dates, grades) will be altered by introducing a small amount of random variance (within 0.1 SD of the data values), to protect user identities from inference attacks (not yet implemented). Long text fields are still filled with random text matching the length of the original text. These will be replaced with "lorem ipsum" text of the same length. WARNING: This will alter data throughout a whole site. DO NOT use on production sites. To install this tool... 1. Copy your site, including the code, database and data folder. 2. Copy the 'pseudonymise' folder to the /local folder of your copied site (or clone from Github). 3. Visit the Notifications page (Site administration -> Notifications). The tool can then be accessed from: - The web interface: Site administration -> Development -> Pseudonymise (can result in a timeout on a large site) - Command line interface: local/anonymise/cli/pseudonymise.php Note that you need to select 'Pseudonymise all' option to pseudonymise all site's data. Depending on what options you select not everything in the site is pseudonymised. Don't give your private data away. Examples of generated names: Courses (randomized): Course Independent Study in Analysis of Musculoskeletal Employment and Forecasting Sections (serialized): Section Dialogue Qualifications and Folklore Activities (serialized): Forum Web Organization and Citizenship Page Abstract Beverage Syntax and Hospitality File Abstract Dialogue Wealth and Mathematics Assignment Yeast Vertebrates and Neuroscience Wiki Rennaisance Nanoscience and Kant URL Rennaisance Journalism and Jazz Workshop Zebras and Xenophilia IMS content package Zygotes Quiz Yeast Vertebrates and Yersinia Pestis Users: Adriana Shkreli Alejandro Abbasi Benita Korhonen Note: This plugin benefits from the use of Gravatar to generate unique avatar images based on the email address, e.g. using the robohash algorithm (https://robohash.org/). See https://docs.moodle.org/en/User_policies#Gravatar_default_image_URL for more details.