User:Sj/2010 global sysops discussion

This is a summary of the major sentiments aired during the discussion, originally drafted with Pathoschild, with a focus on those that were the subject of heated discussion:

1. Voting
   1. The topic really motivated input, and a significant majority of responders were in favor.
   2. This was perceived by some participants as a pure vote, and by others as a consensus decision. It was advertised as / named as a vote, despite the language of consensus being used elsewhere. As a result many people voted with no reason
      • There was some additional minor issues re: who gets to weigh in and how the discussion/vote is concluded that could have been clarified sooner and more crisply.
      • Identification of votes based on whether their wiki 'would be affected' was generally considered inappropriate, especially considering the impact of global[un]blocking.
   3. There is a general sense that 80% is a normal threshhold for such discussions to have reasonable consensus. Some think this means there is room for interpretation when the balance is close to that, others do not.

2. Scope
   1. The "10" and "3 / 2 months" values were not explained. The process for identifying which wikis would be affected was clearer, but left a small number of people confused.
   2. Some powers granted global sysops were considered by many to be unnecessary or inappropriate. global[un]block and editusercssjs were mentioned most often. Globalblock affects more than just the wikis otherwise helped by the proposed role.
      • One of the most frequent concerns was that admins from one background wouldn't be familiar with the style and policies of another wiki, or with the local language. (as a result, autopatrol/patrol/markbotedits might also be considered inappropriate)
   3. There was general support for the idea of tackling vandalism more comprehensively.
      • Many requested more concrete data about the rate and quality of current vandalism issues on small wikis, and the tools/time needed to address most of them.
   4. A specific process for optin/out was not proposed.

      Compare the opt-in process used for the global bot policy: anyone can propose on the wiki that it adopt the global policy, and if there is no opposition after a reasonable period, it is adopted. Anyone can thereafter suggest unadopting the global bot policy, but noone does. That has avoided precisely the concerns voiced here.

3. Criteria
   1. The process for letting potential commentors know about a global sysop nomination was unclear.
      • Some commenters suggested explicit term limits. The proposal itself discussed potential removal for unsuitability. The way g-sysop selection falls between all-project calls to participate for steward elections and more casual one-week polls for individual project sysop votes made some people feel they would have little control over who was given the right to come administrate their wiki.
      • Some suggested that 'being apolitical' be a requisite. Others suggested starting this role with more limited powers.
      • The meaning and scope of applicability of the phrase "only in urgent cases of abuse, or for non-controversial maintenance" was not clear enough for some commenters.
      • There is a sincere concern among many commentors that the most active admins and contributors from large wikis would end up imposing their sense of propriety/style/administration on smaller wikis. (Incnis Mrsi summarized that point of view well.) It may be a pessimistic concern, but bears addressing with more sympathy. By the definition of who gets to vote on this proposal, and by the nature of larger wikis, most people weighing in have been large wiki editors.