In this GitHub issue I opened, I briefly mentioned how I do not thinking properly handling explicit content by overloading the tag system for that purpose is the smart solution long term. That said I think there is a need to expedite the handling of NSFW images on steemit.com, so I am in favor for now of using the #​nsfw tag to hide post summaries (which is not yet implemented on steemit.com as of the time I am writing this comment) and to mark the post with the NSFW label.

That is an easy enough fix that should buy us enough time to work through the details of a better metadata solution. Perhaps something like the one in the OP. However, I do need to modify the OP a little bit to work nicely with the new tag system instead of the old category-only system. I think a modified version of the explicit standard in the OP would be a much more elegant solution than the stop-gap solution of filtering based on the #​nsfw tag.

@pharesim brought up a valid concern in the slack about wasted UI space by requiring a checkbox to mark a post as NSFW. My response is that it would first require a single extra line (which is really an expandable div) that asks the user if they wish to customize the metadata of the post/comment they are writing. For users who have not been defaulted by the client into certain metadata (set explicit, language, etc. fields in the metadata) due to the client's heuristics (e.g. names of tags included in the post/comment or that of tags included in the top-level post of that comment's discussion thread, explicit or language metadata settings of the post that a comment is replying to, etc.), this would be the only additional UI elements added to the post draft UI. For posts that are defaulted into some of these metadata fields (won't be for most users), they will additionally see some of those defaulted settings above the Post button, so that they know what will be submitted to the blockchain and so that they have opportunity to change it if they think it is necessary. I think this can be a very elegant solution to handle not just NSFW content, but all kinds of explicit content ratings, as well as other standards discussed in steem-standards, such as the language standard.

In summary, the idea is that if you don't want to customize or fill in extra metadata yourself and you aren't defaulted into any metadata, the only wasted line is that single line (which is an expanding div) asking if you want to customize metadata for the post. Also, you can always at a glance tell what metadata has been set for the post (either by you or by default by the client due to heuristics) in a location that is close to the Post button so that you can confirm it is good before clicking the Post button and actually submitting the post (+ metadata) to the blockchain.