This role identifies users with some degree of domain expertise, usually represented by tag badges; and some degree of platform expertise, in the form of significant commitment to a specific tag.
Curating (old) content is one of the main differentiators between a repository of high-quality Q&A and a helpdesk, therefore the role is aligned with the overarching goal of Stack Exchange.
In my view, theThe main difference between goldtag badge holders and tag curators is that users with domain expertise who are also interested in tag curation will edit, vote-to-close, dupe-hammer and/or answer old questions.
Users with gold tag badgesexpertise who are not interested in curation, typically will only answer, and possibly cast votes only on newest questions. For them, older content is simply not interesting. Tag curators instead will also curate existing older content in their tag.
Tag badges (possibly just gold, but also silver and bronze) + number of actions (edits, close/reopen votes, delete/undelete votes1) on older content with the relevant tag where they didn't participate in tag editing. Some sort of "Archeologist""Archaeologist" badge on a specific tag could be used as a metric2.
Note: If the goal of this brainstorming is simply to define new user roles to give credit where it's due, it isit's not strictly necessary to attach privileges to it. In my view, additional privileges make sense and are a nice addition.
Deletion privileges shouldn't be affected. Personally, I do consider deletion as an important part of curation. However these are actions that remove/hide content and are harder to properly oversee. Further reducing thresholds could become a vector for abuse.
This stems from a practical issue. Tag curators who scavenge for old content, especially when looking for dupe targets, may very easily come across off-topic contentposts that didn't get moderated on timeescaped moderation. If a several-years-old off-topic question has also received answers, casting a close vote more often than not results in... nothing. The close vote will just age away in the close queue.
This is even more true for low-traffic tags, where the chances that new content gets timely moderationmoderated are lower, and off-topic material just piles up forever. When such off-topic unclosed questions receive new activity, they may getbecome eligible for concerted action in certain chat rooms, but this simply doesn't scale.