I know I'm late to the game commenting in here, but I think it's important to distinguish whether we're talking about people merely adding descriptors to their NSFW posts or whether we're talking about a whole tagging system being implemented into the Lemmy protocol. Adding a full fledged tagging system to Lemmy similar to Mastodon would be a pretty big undertaking. But anything short of that would feel like a kludgey, bolted-on solution. The NSFW flag + user added descriptors is the only way for users to go about it for now.

I do think a conversation about adding tags into Lemmy is worthwhile because tags could potentially allow for a lot of neat things. E.g. we might eventually be able to start making custom feeds by aggregating content from multiple similar communities across instances. It would also let users filter tags they don't want to see at all (regardless of whether it's NSFW) or don't want to see by default (i.e. blur the images until they're selected). But I know the Lemmy devs probably don't want it to end up looking like a Reddit/Mastodon hybrid or just become kbin, so this is something they'll consider carefully. I think right now they're rightly focused on database and federation performance and the broader dev community is helping with more immediate needs like spam account mitigation. They've gotta stop the bleeding and get a solid foundation established before building major changes like a full fledged tagging system.