This Proposal does not offer a solution, but rather tries to initialize an update for the existing oversight policy. This one shall be adjusted to the new core feature RevisionDelete (suppressrevision).
Currently, the Oversight policy allows hiding of potentially libellous information only when the subject concerned specifically demands Oversight (or on advice of the Wikimedia Foundation, which almost never happens). This leaves the question, what happens with libellous information on which the subject has not made any statement, because he is not aware of it. As opposed to the policy on EN Wiki where anyone can request oversight for such information - not just the victim.
- Revise and rethink the Oversight policy regarding the use on potentially libellous information.
- Adapt policy to fit RevisionDelete and Hideuser features, rather than the old Extension:Oversight
In 2006, the new user group "oversight" was created "to permanently remove individual edits from an article's history" with the extension Oversight. While there seems to be no Draft or Resolution on wmf-wiki the policy is used since September 2009 in these three cases:
Because that "Oversight" extension had disadvantages like irreversibility (just possible for users with database access), only removal of entire revisions etc. that policy was made so strict. With the new core feature RevisionDelete which was written to improve the work of oversights and stewards it's not any longer needed to remove entire revisions. Libellous (resp. insulting) or private informations publishing user names (which daily occur dozens of times) can be hidden while it's not needed to remove the revision or summary (besides, it's possible to remove summaries which contain those libellous or non-public informations). With the new feature "hideuser" it's even possible to do that for all contributions of one user with one single-click. Libellous or even "just" single user insulting/offending user names could be easily and revertably hidden without destroying the history. It would not be needed to be "specifically asked by the subject" as the policy says, because nothing else happens than one stroked out user name, the history will not be destroyed.
Part 2 of the oversight policy could be changed to a more practical way: Removal of potentially libellous (or insulting?) information (without the need of specifically asked by the subject). Some stewards and oversights act in that way yet and mostly ask the subjects afterwards to confirm that action. All users confirmed and were glad that those user names (which they sometimes haven't even seen) did not any longer appear in the article's history.
In case of libellous information, practically only few users ask for the information to be removed, because:
- They are not aware of the information (no one is able to check every edit for potentially libellous information concerning him)
- This could be countered by informing the user, unless:
- They are not even on Wikipedia. Often it is impossible to contact the aggrieved person.
- This could be countered by informing the user, unless:
- If they are aware of the information, they might not be aware of Oversight and the Oversight policy.
Those who are not informed about the libellous edits on Wikipedia, are disadvantaged against those who are, because they don't have a chance of the libellous information to be removed. This should be reconsidered.
Additionally, one should keep in mind, that the current Oversight policy was designed for the old mw:Extension:Oversight, which has been replaced by mw:RevisionDelete. RevisionDelete is much more transparent (for example: revisions are still in the version history, their existence is no secret, they are simply crossed out. With the old Oversight system, revisions appeared as "does not exist and never existed") and changes can be easily verified and undone. The use of RevisionDelete is probably less controversial than the use of Extension:Oversight.
What should we do? Some of these options can be combined, for example "Opt-In for requests" and "Inform users if possible".
|Case-specific requests||In each case, the user has to explicitly request Oversight|
|Opt-In for requests||A user may choose to Opt-In for removal of all libellous information|
|Inform users if possible||Send Emails if an Email address can be spotted and ask for permission to use OS|
|Opt-Out for requests||For users opposing Oversight|
|Hybrid model - OS without request in exceptionally severe cases, but normal cases require requests||Question: What are severe cases?|
|OS all cases of libellous information||Could lead to a large number of OS actions|
- If there is a hybrid model (OS without request in exceptionally severe cases), how should we define what cases can be oversighted without a request? It's hard to draw a line.
- If users should be informed, whose job is this? The job of Oversighters?
- In what way does the change of the technical basis (old vs. new OS) change the way we should think about OS? How shall the existing oversight policy adjusted to the new core feature RevisionDelete?
- Which are the rules for new features like "hideuser"?
- No monetary costs
- Board has to write/approve an updated policy
- Documentation has to be changed
- Users who are already aware of "Oversight" have to be informed about the new policy
- Wikipedia:Requests for oversight The EN Wiki policy which does allow anyone to request that anyone's Personal data and libel be oversighted.
Do you have a thought about this proposal? A suggestion? Discuss this proposal by going to Proposal talk:Rewriting of oversight policy.
Want to work on this proposal?
- .. Sign your name here!