Wikipedia:Pending changes/Request for Comment February 2011

Request for Comment February 2011 is a multi-phase discussion of the use of the pending changes feature on the English-language Wikipedia. In accordance with consensus during the third phase, pending-changes protection was removed from all articles on Friday, 20 May 2011, with no prejudice against future reinstatement, in some form, based upon consensus and discussion.
 * The first phase was open discussion.
 * The second phase was endorsements of positions, the analysis of which is available here.
 * The third phase was a proposal on what to do with pending changes in the short term.

Background
Pending changes (PC) is a tool that underwent a two-month trial on the English Wikipedia in 2010. A straw poll at the end of the trial, which closed on September 4, 2010, showed 407 in favor of implementation in some measure and 217 opposed, with 44 other responses. Among those in favor, there was no clear consensus as to what form the implementation should take. PC remained active on some pages, but without clear guidelines as to when it should be used, or consensus on how to review changes that are pending.

Phase 1: Open discussion
Request for Comment February 2011 began on the 16th February 2011. It was not originally intended to have more than one phase. Editors were invited to discuss pending changes, in the hope that it would lead to clear and specific instructions for administrators and users requesting pending-changes protection. It was made clear that it was intended to be a discussion, not a vote. A large variety of opinions was expressed.

Phase 2: Endorsement of positions
In order to gauge how much support the different opinions had, a number of positions were stated and people were asked to put their name to those with which they agreed. A large number of positions were stated. These positions weren't all mutually-exclusive, so it was possible for editors to endorse a number of positions. The analysis of this phase showed that the three most endorsed positions were "PC helps with libel on BLPs", "PC is confusing" and "PC reduces vandalism, but so does semi-protection".

Phase 3: A proposal for the short term
It was decided that discussion of whether and how we want to use Pending Changes in the medium and long term were being overshadowed by a lack of consensus on what to do with Pending Changes in the short term. Phase 3 was an attempt to reach a consensus on that short-term issue. It was proposed to remove pending changes protection from all articles, for now, with no prejudice against reinstating it in the future, in some form, based upon consensus and discussion.

Editors were asked to state whether they support or oppose the proposal and give a brief reasoning. Comments longer than 1000 characters, responses to the comments of others and general discussion of the topic were to go in the discussion section, lower down the page.

127 editors supported the proposal and 65 opposed it. An administrator determined that there was consensus for the proposal and pending-changes protection was removed from all articles on Friday, 20 May 2011.

Possible further phases
A number of suggestions have been made for what to do next, either as an additional phase or in a new location. The suggestions include:
 * Discuss the requirements. What do people actually want from a tool like pending changes? Progress from that point in a logical fashion.
 * A questionnaire. This would enable each user to express an opinion on set questions. The responses would be analysed by a panel of trusted editors, who would bring a proposal forward to the community.
 * Essays. Users would be encouraged to write essays, analysing the pending-changes situation. They will use the space afforded by an essay to analyse evidence and construct logical arguments. It is anticipated that these would include:
 * 1) Proposals for how PC should be done
 * 2) Proposals for further trials
 * 3) Analyses of why we can't reach consensus
 * 4) Suggestions for how to make PC less confusing
 * 5) Brief histories of PC
 * Allow local experimentation. Give a nominated WikiProject a free hand to do what it likes on articles that relate to its topic, requiring only consensus amongst those involved in that WikiProject. After a certain period, the community would assess what the WikiProject had done and decide where to take the idea from there.
 * A reviewing experiment. Give the draft policy to 20 users and have them each review the same set of 50 randomly selected changes chosen from the types of articles and users that would be eligible for PC under the policy. Check how well their reviews agree with each other and collect feedback on where they felt the policy provided inadequate guidance.
 * Phased discussion. Discussion should continue, possibly in the following phases:
 * 1) The scope of articles that people would like to see protected by PC in the medium term, assuming that other issues can be sorted out. This could range from none to many articles, with various conditions for allowing or disallowing PC protection on certain articles
 * 2) Technical issues. How should the software work? Are some modifications essential and others just "nice to have"?
 * 3) Operational guidelines. How we should select reviewers, the criteria for accepting or rejecting edits, etc.