RafaPolit
commented
3 years ago How would this work in practice?
- Is this a one time notification: "Please be warned that there could be shocking images in this collection"?
- Is this a persistent warning symbol on cards and entity view on that template to showcase that it could be traumatic?
- Is this a blurred image replacing ALL the contents of that template until you actively un-mask the image with an action?
- Is this an automated detection of images that could be shocking (through AI or something similar)?
- Is this a manual tagging of the image as "not safe" or something like that, that would require user curation of the content?
I see so many options here, ranging from a simple message to very complex algorithms. It would be interesting to know what are the expectations?
In use cases where the partner is documenting violations directly from primary sources or is crowdsourcing information from social media and other sites on on-going conflicts, the nature of information that is collected within Uwazi can contain violent and/or traumatic imagery and videos. Evidence of violations documented by citizens/bystanders in the form of images and videos are increasingly becoming common that are then brought into our databases for analysis and verification. In keeping with the standards to maintain the well-being of Uwazi users (both admin and end user) who would access this database, it could be beneficial to explore the idea of being able to configure a cautionary message for certain templates in Uwazi.
User story: As an admin user, I want to be able to display a trigger warning message for specific templates in my database that may contain traumatic video/imagery so that both my team members as well as the end user is able to prepare themselves before being subjected to potentially disturbing content.