obulat
commented
2 weeks ago One potential issue is the "readiness" of Openverse to introduce this functionality. Are our results sufficiently "safe" to make this assurance to users? Is there a safety threshold we could or should measure (some threshold of sensitive content reports, or something) we should determine first?
Maybe this version should use the non-user-generated media at first, as we do for some integrations?
zackkrida
Problem
It seems likely that educational facilities using Openverse may want to restrict access to sensitive content on the platform. We've received feedback in the past relating to sensitive content from folks identifying as education-affiliated:
Description
Create a version of Openverse where the sensitive content filter is "locked" and cannot be modified. This could be handled in a number of ways, that should be determined by doing some research into how institutional firewalls commonly work. Some ideas though could be derived by how Google handles this with functionality called "Lock SafeSearch for accounts, devices & networks you manage".
The tl;dr there is that they have network admins set a DNS CNAME record for forcesafesearch.google.com to www.google.com.
Alternatives
Additional context
One potential issue is the "readiness" of Openverse to introduce this functionality. Are our results sufficiently "safe" to make this assurance to users? Is there a safety threshold we could or should measure (some threshold of sensitive content reports, or something) we should determine first?