Addendum (March 2013) : As per
this declaration, the WMF requires a RFA or RFA-identical process for being granted rights with access to deleted revisions, including explicitly OS and implicitly CU.
In view of the fact that
adminship is not presently a social requirement for higher privileges such as
CheckUser,
Oversight, and
Bureaucrat, it is proposed that these user groups be made self-sufficient by adding certain rights (currently found in the administrative toolset) to each bundle.
This discussion is intended to focus on addressing the technical limitations present in the
current makeup of user groups/rights. The related question - whether the administrator privilege should be a social requirement for such privileges - is being discussed
elsewhere, and carrying out this technical change would not preclude the ability of the community to implement such a requirement if consensus were found for the same.
Discussion initiated by –
xenotalk 15:13, 12 April 2011 (UTC)reply
Add viewdeleted privileges to CheckUser
Comprehensive consensus in favour of implementation.
Happy‑
melon 10:28, 15 May 2011 (UTC)reply
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
In order to remove the technical limitation that
checkusers must also be administrators to review deleted content, the deletedhistory, deletedtext, and browsearchive rights should be added to the checkuser permission group.
Support
Proposed. It is important that a CheckUser have the ability to view deleted content, as it may inform their findings during an investigation. –
xenotalk 15:13, 12 April 2011 (UTC)reply
In favour of proposal. CheckUser involves a high degree of trust already - I'd see no harm in allowing them to see deleted revisions.
Zakhalesh (
talk) 15:20, 12 April 2011 (UTC)reply
Agree that the view-deleted rights are necessary for CU work.
Cenarium (
talk) 15:30, 12 April 2011 (UTC)reply
I don't see any reason a non-admin checkuser shouldn't be able to view deleted content to assist with their investigations.
28bytes (
talk) 15:36, 12 April 2011 (UTC)reply
Good idea. Viewing deletions is no big deal, CU's are already established and respected members of the community. --Errant(
chat!) 15:38, 12 April 2011 (UTC)reply
Can't see why not.
Jafeluv (
talk) 15:41, 12 April 2011 (UTC)reply
Strong Support Thinking about
SPI, it would be really useful, the wikimedia foundation has trusted them with private information so they should be able to trust them like that. ~~
EBE123~~ talkContribs 18:56, 12 April 2011 (UTC)reply
Per my reasoning at the Village Pump. --
Avi (
talk) 20:00, 12 April 2011 (UTC)reply
Support It's assumed that people who are requesting checkuser have already been vetted by a collection of trustworthy editors and that the community at large has been given time to review the vetting and express it's thoughts about the candidate.
Hasteur (
talk) 20:01, 12 April 2011 (UTC)reply
IMO, this entire discussion is quite silly. I agree completely with xeno. If somebody's job requires them to do certain tasks, but they aren't permitted to do those tasks by their job, the entire system they're in is broken.
demize(
t ·
c) 02:18, 13 April 2011 (UTC)reply
While I won't say it is silly, it is just good systems management to ensure that each toolkit/user right has the tools to permit it to operate discretely. The current situation is largely a historical accident, because the initial user right was developed on the fly to reduce unnecessary workload on developers, and was never corrected.
Risker (
talk) 02:30, 13 April 2011 (UTC)reply
I have no problem with a non-sysop checkuser.
Reaper Eternal (
talk) 02:46, 13 April 2011 (UTC)reply
Checkusers are trusted to view non-public material, and this is an extension of that trust.--
Danaman5 (
talk) 02:48, 13 April 2011 (UTC)reply
Believe anyone trusted enough to be given this right, could also be trusted with this needed extension.
CalmerWaters 05:13, 13 April 2011 (UTC)reply
They are trusted to view much more than this; and viewing this may be necessary to determine when a check is needed.
עוד מישהוOd Mishehu 06:02, 13 April 2011 (UTC)reply
Absurd that it could be otherwise. --
SmokeyJoe (
talk) 09:50, 13 April 2011 (UTC)reply
I am also surprised that this is not already the case, and makes sense to allow non-admins to be checkusers.
CT Cooper ·talk 09:53, 13 April 2011 (UTC)reply
Checkusers already have a very high degree of community trust and should have access to deleted material.
Kudpung กุดผึ้ง (
talk) 05:12, 15 April 2011 (UTC)reply
Deleted material access is needed to do CU properly, and the selection process implies that they are trusted to view IPs which are far more sensitive. The remaining question is whether non-admin checkusers should be allowed and that's a social question, inclined to agree the tools should be self-sufficient either way (which doesn't affect that decision) - not least to support non-admin
audit subcommittee members.
FT2(
Talk |
email) 10:20, 15 April 2011 (UTC)reply
Obvious Support If they can be trusted with CU, why not seeing deleted revs.? --
Patar knight - chat/contributions 01:22, 16 April 2011 (UTC)reply
With the variety of situations CUs go through it is most definitely needed to view deleted revisions. In fact, I cannot think of a reason it has not already been done. mauchoeagle 04:54, 17 April 2011 (UTC)reply
Support—
mc10 (
t/
c) 02:05, 18 April 2011 (UTC)reply
What problem will this proposal solve? Are there a long queue of users who want to be checkusers but will not run through RFA? I can not support this solution in search of a problem.
Ruslik_
Zero 19:06, 12 April 2011 (UTC)reply
A user was recently appointed to the Audit Subcommittee and we found that they were unable to effectively investigate a case because of the inability to view deleted revisions. Further, there is nothing preventing a non-admin user from seeking advanced privileges; but without carrying out the technical change proposed herein, they would be hamstrung in carrying out their duties if they were appointed. –
xenotalk 19:18, 12 April 2011 (UTC)reply
You know it is kind of moot now.
Ruslik_
Zero 19:28, 12 April 2011 (UTC)reply
The present case will likely be rendered moot; this does not mean we should neglect to implement a technical solution for future similar cases. –
xenotalk 19:34, 12 April 2011 (UTC)reply
I think there should instead be an "auditor" role with the bundle of bits appropriate for the task.
69.111.194.167 (
talk) (was 75.57.242.120) 00:58, 13 April 2011 (UTC)reply
Sorry, but I really don't see the need to add another group, and as a consequence even more bureaucracy to this.
Ajraddatz (
Talk) 03:17, 13 April 2011 (UTC)reply
Checkuser requires the community's trust if they can not pass a RFA they do not have the community's trust.
Zginder 2011-04-28T01:55Z (
UTC)
Discussion
Was requested in
bugzilla:28440; a developer requested a clear consensus discussion be held prior to committing the change. –
xenotalk 15:13, 12 April 2011 (UTC)reply
The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
Add viewdeleted privileges to Oversight
Clear consensus in favour of implementationHappy‑
melon 10:29, 15 May 2011 (UTC)reply
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
In order to remove the technical limitation that
oversighters must also be administrators to review deleted content, the deletedhistory, deletedtext, and browsearchive rights should be added to the oversight permission group.
Support
Proposed. It is important that an Oversighter have the ability to view deleted content for a number of reason; most importantly because problematic content is typically deleted by administrators prior to being submitted for suppression. –
xenotalk 15:13, 12 April 2011 (UTC)reply
Support. Oversighters are trusted to delete the most sensitive material, so they should also be able to see it.
Zakhalesh (
talk) 15:22, 12 April 2011 (UTC)reply
Agree that view-deleted rights are necessary for OS work.
Cenarium (
talk) 15:30, 12 April 2011 (UTC)reply
Odd that they don't have those rights already.
28bytes (
talk) 15:36, 12 April 2011 (UTC)reply
As 28Bytes says, they should have had it already.. I mean, we trust them to scrub the most sensitive stuff for us! --Errant(
chat!) 15:39, 12 April 2011 (UTC)reply
It's part of their job to deal with objectionable content, so obviously they should be trusted to view it as well.
Jafeluv (
talk) 15:50, 12 April 2011 (UTC)reply
Userrights ought not need to be interdependent to function effectively.
Skomorokh 15:53, 12 April 2011 (UTC)reply
Per my reasoning at the Village Pump. --
Avi (
talk) 20:00, 12 April 2011 (UTC)reply
Support It's assumed that people who are requesting oversight have already been vetted by a collection of trustworthy editors and that the community at large has been given time to review the vetting and express it's thoughts about the candidate.
Hasteur (
talk) 20:10, 12 April 2011 (UTC)reply
Well, it would be rather silly if an oversighter had access to a "deeper" level of deletion and not the standard one.
sonia♫ 20:35, 12 April 2011 (UTC)reply
Makes sense, especially if many oversight requests are from sysops who have already (revision) deletion the pages or edits in question. –
MuZemike 21:22, 12 April 2011 (UTC)reply
It is just good systems management to ensure that each toolkit/user right has the tools to permit it to operate discretely. The current situation is largely a historical accident, because the initial user right was developed on the fly to reduce unnecessary workload on developers, and was never corrected.
Risker (
talk) 02:31, 13 April 2011 (UTC)reply
I have no problem with a non-sysop oversighter. Oversight is only for a very narrow set of issues.
Reaper Eternal (
talk) 02:45, 13 April 2011 (UTC)reply
Oversighters are trusted to view non-public material, and this is an extension of that trust.--
Danaman5 (
talk) 02:48, 13 April 2011 (UTC)reply
Very well can't properly oversight, as part of the duty is to ensure other oversighters are acting according to policy (due that they are the only ones to can view suppressed content) without being able to view deleted material
CalmerWaters 05:19, 13 April 2011 (UTC)reply
If they can be trusted to perform this highly confidential task ionvolving viewing the most secretive data posted on Wikipedia, they should be able to view all of it.
עוד מישהוOd Mishehu 06:05, 13 April 2011 (UTC)reply
Deleted material access is needed to do OS properly, and the selection process implies that they are trusted to view suppressed data which is far more sensitive. The remaining question is whether non-admin oversighters should be allowed and that's a social question, inclined to agree the tools should be self-sufficient either way (which doesn't affect that decision) - not least to support non-admin
audit subcommittee members.
FT2(
Talk |
email) 10:20, 15 April 2011 (UTC)reply
Weak support - It makes vague sense however I am not seeing any reason it is needed. OSers should have access to revisions that have been oversighted. As for anything else...I don't think so.mauchoeagle 05:03, 17 April 2011 (UTC)reply
Oversighters routinely have to address deleted revisions. Quite often oversightable material is also deleted. For example, some will have been deleted on sight by an admin, which correctly reduces the window for public viewing or spidering (eg a defamatory attack page under
CSD#G10 or a privacy breaching revision under
CRD#4). An oversighter then has to review the content and decide whether suppression applies. They also sometimes need to review suppressed edits or confirm if there has been any suppression, on a deleted page or in a user's deleted contributions. They can't reliably act as oversighters if they cannot see the deleted text.
FT2(
Talk |
email) 12:23, 17 April 2011 (UTC)reply
Support. —
mc10 (
t/
c) 02:06, 18 April 2011 (UTC)reply
I have been taking more of an active role in reporting things to oversighters, this should be obvious, Support.
TheHelpfulOne 16:25, 24 April 2011 (UTC)reply
Yes, a good idea. -- Cirt (
talk) 16:37, 9 May 2011 (UTC)reply
Replied above. –
xenotalk 19:18, 12 April 2011 (UTC)reply
Not really opposed on purely technical grounds (it's silly to let someone see oversighted edits but not deleted ones), but I'm no longer so sure it's ok to unbundle the technical and policy discussions (
code is law). I think the AUSC situation should be handlled by creating an "auditor" role with the bundle of bits necessary for the task, rather than turning individual capabilities on and off separately.
69.111.194.167 (
talk) (was 75.57.242.120) 01:01, 13 April 2011 (UTC)reply
I don't think non-admin oversighters are something we need/want.
Stifle (
talk) 10:00, 13 April 2011 (UTC)reply
Oversight requires the community's trust, if they can not pass RFAs they do not have the community's trust.
Zginder 2011-04-28T02:05Z (
UTC)
RfA has nothing to do with trust, it's a beauty civility contest.
MalleusFatuorum 02:26, 28 April 2011 (UTC)reply
Even if it is a civility contest, do you want an overseer to lose his temper and release secretive information to the world?
Zginder 2011-04-28T02:32Z (
UTC)
You're not making much sense. Are you suggesting that someone who's been through RfA is guaranteed not to lose their temper and do exactly the same thing?
No, we are never guaranteed anything. I am stating it is less likely.
Zginder 2011-04-28T03:08Z (
UTC)
Discussion
Was requested in
bugzilla:28440; a developer requested a clear consensus discussion be held prior to committing the change. –
xenotalk 15:13, 12 April 2011 (UTC)reply
The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
In order to remove the technical limitation that
bureaucrats must also be administrators to review deleted content, the deletedhistory, deletedtext, and browsearchive rights should be added to the bureaucrat permission group.
Support
Proposed. It is important that a Bureaucrat have the ability to view deleted content, as it may inform their decisions when closing a requests for adminship or bureaucratship, or while carrying out renames. –
xenotalk 15:13, 12 April 2011 (UTC)reply
This also makes perfect sense. As xeno has pointed out, determination of consensus at RFA sometimes involves reviewing deleted content. Having this right included as part of the package would help make it more clear that a bureaucrat need not necessarily be an admin, too. ···
日本穣? ·
投稿 ·
Talk to Nihonjoe ·
Join WikiProject Japan! 16:00, 12 April 2011 (UTC)reply
Probably irrelevant in the current climate of deification of administrators but sensible nevertheless.
MalleusFatuorum 16:06, 12 April 2011 (UTC)reply
This was the one I was least clear upon, but, I can see the AFD argument, and the particular rights included here do not strike me as worrysome. --
joe deckertalk to me 16:13, 12 April 2011 (UTC)reply
Even if bureaucrats don't use these rights enormously often, there is no real harm in adding them. Reviewing a lone oppose vote that made mention of CSD issues, reviewing a previously promotional editor—both of those are quite valid reasons that a non-admin bureaucrat might need to use these rights.
NW(
Talk) 16:26, 12 April 2011 (UTC)reply
This seems like the least compelling of the four proposals... but on the other hand, I can't see any particular harm in it either. Xeno's arguments below are not unreasonable.
28bytes (
talk) 16:30, 12 April 2011 (UTC)reply
Agree that deleted contributions are relevant to evaluating consensus at RfA.
Jclemens (
talk) 16:46, 12 April 2011 (UTC)reply
I think bureaucrats should be admins, but since that isn't the question here: because of Xeno's description below of how viewdeleted/etc is useful for 'crat jobs, these rights ought to be attached to the bureaucrat permission group.
ErikHaugen (
talk |
contribs) 16:54, 12 April 2011 (UTC)reply
I don't think it's my place to second-guess what the 'crats say they are using in their 'crat duties.
T. Canens (
talk) 17:01, 12 April 2011 (UTC)reply
Upon further reflection, I can see the benefit of a 'crat being able to review deleted edits when reviewing strength of arguments. And no, being a 'crat is not predicated upon being an admin. It is theoretically possible that one could become a 'crat without being an admin. It is also possible that one could relinquish admin responsibilities without reliquishing 'crathood. And it should be possible to have someone's admin rights removed, without removing their 'crat rights.---BalloonmanNO! I'm Spartacus! 19:46, 12 April 2011 (UTC)reply
I think this would rarely matter at RFA, but I can see it mattering when renaming a user. ϢereSpielChequers 21:33, 12 April 2011 (UTC)reply
This is one of the less clear ones, but I think people like NW above got it right: there is an occasional need for bureaucrats to be able to view deleted content.
Ucucha 23:24, 12 April 2011 (UTC)reply
Had to think harder about this one! But, on the whole, I can't see the slightest bit of harm in it, and even if it's not needed today, it may be needed tomorrow. Much better to have something and not need it than to need it and not have it.
Pesky (
talk) 08:05, 13 April 2011 (UTC)reply
Xeno convinced me. Ultimately no real reason this is a big issue. --Errant(
chat!) 08:56, 13 April 2011 (UTC)reply
There are certain instances in which bureaucrats might need to see deleted revisions, and with this being a trusted and small user group, extending this right to them should not be a major issue. I also don't think being an admin should be a requirement on becoming a bureaucrat, as the roles are rather different.
CT Cooper ·talk 10:02, 13 April 2011 (UTC)reply
There is no reason why as crat should not be allowed to resign their admin bit if they so choose and continue performing their functions as a crat. It is not a requirement for the position, and such a technical impediment should not be an underlying reason for the unwritten "requirement" that currently exists. Jim MillerSee me |
Touch me 14:32, 13 April 2011 (UTC)reply
I think all 'crats should already be (or have been) admins because the experience informs their reading of an RFA (case in point). Viewing a deleted diff that is under discussion also informs their reading of the discussion; any non-admin that has had difficulty understanding a discussion about deleted content until some administrator emailed them a copy should instinctively understand this point. But that should not preclude someone that was once an admin and gave up the admin bit, a real possibility that Xeno raises below.--
Chaser (away) -
talk 15:23, 14 April 2011 (UTC)reply
The argument in the oppose column that the rights would be redundant holds no water in the context of this RfC if the opposer at the same time supports adding move-related rights to the group. The decision is solely whether the viewdelete-related rights are required or useful for the role of a bureaucrat or not. With two crats for, two against, I can only assume that the rights under discussion are useful for the modus operandi of some crat's. Since I see no other reason against adding these passive rights to the crat group (the supposed WMF veto against making it available for very large groups of editors does not apply), I support it.
Amalthea 15:47, 14 April 2011 (UTC)reply
I can see that viewing a deleted edit may be important to evaluating consensus at an RfA. Hut 8.5 17:58, 14 April 2011 (UTC)reply
Could be significant in the context of a rename. The tool we're discussing is the ability to view deleted content, not to delete or undelete it. If we trust someone to promote admins, then by implication we probably trust them to at least view deleted posts. Candidates seen as having a problem in this area would quickly find themselves opposed at RFB.
FT2(
Talk |
email) 10:30, 15 April 2011 (UTC)reply
Obvious Support Crats are trusted. Crats must have great community trust to be crats. --
Patar knight - chat/contributions 01:25, 16 April 2011 (UTC)reply
Support Frankly can't see what the big fuss is about. If I trust someone as a b'crat, I trust him as a sysop. By contraposition, if I don't trust someone as an admin, I surely don't trust him as a bureaucrat. The proposed change just reflects a fairly uncontroversial hierarchy of levels of trust.
Pichpich (
talk) 14:59, 17 April 2011 (UTC)reply
Sure, seems sound and sensible. -- Cirt (
talk) 16:38, 9 May 2011 (UTC)reply
Oppose
Unlike the other two userrights, I don't consider this to be a mere technical limitation. If an admin has engaged in poor conduct, then I would not be willing for them to start/continue their work as a crat after their tools have been removed, forcibly or voluntarily.
Ncmvocalist (
talk) 15:54, 12 April 2011 (UTC)reply
I don't see how it is necessary for bureaucrats to view deleted content. All bureaucrat decisions should be based on open discussion or open requests. I do not think that bureaucrats should verify on their own the evidence presented in a RFA, instead they should base their decision on the opinions provided. I also don't see how deleted contribs would matter in a rename request, seeing if the user engaged in promotion is an administrator role, not a bureaucrat role. Moreover, if the community were to consider non-admin candidates for bureaucratship, the addition of view-deleted would make it a bigger deal than it should be, and we know bureaucratship is already too big a deal.
Cenarium (
talk) 16:05, 12 April 2011 (UTC)reply
Remember that the Bureaucrats must also be sysops so I think that this is a bit useless. ~~
EBE123~~ talkContribs 19:01, 12 April 2011 (UTC)reply
Must they be? –
xenotalk 19:18, 12 April 2011 (UTC)reply
Well, when you get bureaucratship, you also become a sysop, and if you didn't get it you can change your userrights to include sysop. ~~
EBE123~~ talkContribs 21:05, 12 April 2011 (UTC)reply
And what if a duly-appointed bureaucrat wants to continue to serve as a bureaucrat but wishes to relinquish their administrative privileges? –
xenotalk 21:09, 12 April 2011 (UTC)reply
Well then the user cannot do viewdeleted. Simple. ~~
EBE123~~ talkContribs 21:13, 12 April 2011 (UTC)reply
I do not see the need for bureaucrats to see deleted edits (speaking as one). If consensus cannot be determined through on-wiki postings, how are deleted posts going to help? Same with name changes and bot flags; unlike CU and OS, I don't see how these rights are integral to performing 'crat duties. --
Avi (
talk) 20:03, 12 April 2011 (UTC)reply
May not be 'necessary' today - but what about tommorow? I have a workshop full of tools, plus those I've inherited from my father, many of which I've never used. But, tell you what, on the day I have to use one, it's a bastard having to go out and buy a new tool when I need it in my hand right that minute! Same goes for a first aid kit - you don't need it until you need it - and then you need it :o)
Pesky (
talk) 08:16, 13 April 2011 (UTC)reply
Unless you can show a reason why a 'crat would need to do this in the future, I think that giving 'crats these rights shouldn't be done. If we give the 'crats more jobs later on, we can discuss it then.
עוד מישהוOd Mishehu 05:29, 14 April 2011 (UTC)reply
The non-sysop 'crat? One who doesn't want to by a sysop, but happy to be a 'crat? Even if we don't currently have one, having a system in place which means it would be difficult to be one seems counterproductive. And also somewhat pointless.
Pesky (
talk) 08:00, 18 April 2011 (UTC)reply
Per MZMcBride. Oh, and per Avi. There's little use, and I'm of the opinion that a crat should be an admin first to give it to others. /
ƒETCHCOMMS/ 21:43, 12 April 2011 (UTC)reply
Per Ncmvocalist and Avraham. Doesn't seem compatible as opposed to the other userrights. Not to mention we've already stripped most of the rights away from bureaucrats and relegated those to sysops or Stewards, virtually making them more or less figureheads in more of a technical sense (remember that closing RFAs is a social, not a technical, right). –
MuZemike 21:21, 12 April 2011 (UTC)reply
If they are really just "figureheads", than why it is so difficult to become crat?
ArmbrustTalk to meContribs 21:56, 12 April 2011 (UTC)reply
It was to answer your question in advance, I think, that MuZemike included a bracketed reminder.
Ncmvocalist (
talk) 22:34, 12 April 2011 (UTC)reply
No reason to do this since non-admins shouldn't become bureaucrats. If a bureaucrat+sysop wants to relinquish a subset of the bits they already have, I guess I'm ok with that. They shouldn't be able to acquire the bits other than through RFA/RFB in that order, at least without a lot more prior discussion specifically aimed at that question.
69.111.194.167 (
talk) 01:07, 13 April 2011 (UTC)reply
I believe that if we are going to open the possibility of bureaucratship becoming a more standalone role that does not necessarily require adminship, there needs to exist some differentiation with adminship, rather than it simply being a sort of "admin-lite". The difference should be that bureaucrats rely on open discussions to make their decisions, with the counsel of administrators and checkusers regarding deleted and private material. We shouldn't be thinking so much in terms of "self-sufficient" roles, and should instead consider the complementarity of roles.--
Danaman5 (
talk) 02:35, 13 April 2011 (UTC)reply
As 'crats promote admins, I think that a bureaucrat should be an admin first. This is especially true as admins attempt to assess consensus at AFD, etc.
Reaper Eternal (
talk) 02:44, 13 April 2011 (UTC)reply
Well, 'crats promoting admins isn't itself the issue, since stewards can also do that (without being local admins).
69.111.194.167 (
talk) 04:39, 13 April 2011 (UTC)reply
That is on the smaller Wikis that do not have local Bureaucrats. I can't foresee an instance where a steward would unilaterally promote an admin on EN; except to correct a de-sysops error.
CalmerWaters 05:04, 13 April 2011 (UTC)reply
Per above, I can see no situation in which a bureaucrat would not be an admin.
Ajraddatz (
Talk) 03:18, 13 April 2011 (UTC)reply
I don't see why a 'crat needs this - any relevant deleted data for RfAs should be made accessable to non-admins participating in the RfA.
עוד מישהוOd Mishehu 07:02, 13 April 2011 (UTC)reply
In the unlikely event that a non-admin becomes a 'crat being able to view deleted edits would have no bearing on their job. In the one or two cases where it would (I can't see more than that) asking an admin or leaving it for another 'crat would work just as well.
Sven ManguardWha? 07:11, 13 April 2011 (UTC)reply
This isn't a good idea. Crats should be admins because they need that experience to help them close RFAs. --Rschen7754 07:17, 13 April 2011 (UTC)reply
Silly. A bureaucrat must be an experienced administrator in good standing. --
SmokeyJoe (
talk) 09:53, 13 April 2011 (UTC)reply
Must one be? It may be the de facto standard, but in no way is it the de jure standard. Theoretically, I could go do an RfB and pass right now, according to policy.
demize(
t ·
c) 10:19, 13 April 2011 (UTC)reply
I believe the de factor standard should be hard written as de jure. Bureaucrats have very limited additional privileges, but what they do have places them clearly as leaders, even elites, and thus a subset, of the administrators. I support bureaucrats acquiring the power and role of de-admining (per a community consensus discussion), and, with a dim view on any individual wearing many hats, aka
separation of powers, support de jure demarcations, such as forbidding combined bureaucrat/arbiter --
SmokeyJoe (
talk) 13:32, 13 April 2011 (UTC)reply
I think that you'll find a lot of folks disagree with you - neither administrators nor bureaucrats are leaders or elites, nor should they be seen as such. –
xenotalk 14:35, 13 April 2011 (UTC)reply
Whether they are supposed to be or not, admins are seen as elites and in many cases act as elites. Admins also tend to be far more protective of admin rights (not breaking them apart, demanding that other people become admins before getting other rights, etc) than non admins. A class system has developed, and it's a turn off for many. Continuing to deny it at this point is counterproductive. Right now, adminship is a big deal, period.
Sven ManguardWha? 20:36, 13 April 2011 (UTC)reply
Per MuZemike. Although I have no problem with bureaucrats not being admins, that doesn't change the fact that viewing deleted pages has absolutely nothing to do with bureaucrat duties. As a bureaucrat, I cannot think of a single time where I have needed the ability to view deleted pages to carry out my bureaucrat duties. --
(ʞɿɐʇ) ɐuɐʞsǝp 14:53, 13 April 2011 (UTC)reply
Per above. Quite redundant. -FASTILY(TALK) 19:07, 13 April 2011 (UTC)reply
per Sven Manguard. --
Ϫ 22:30, 13 April 2011 (UTC)reply
Expect that bureaucrats should also be admins so should not need this. I would not expect an RFB to pass without an RFA also being passable.
Graeme Bartlett (
talk) 12:16, 14 April 2011 (UTC)reply
On the simple grounds that this one is not like the others, and the rationale at the top of this page doesn't apply to it.
Chick Bowen 05:22, 15 April 2011 (UTC)reply
Oppose – Unnecessary for normal 'crat work. —
mc10 (
t/
c) 02:06, 18 April 2011 (UTC)reply
Oppose unlike the other proposals its not clear really why Bureaucrats need this right. --
Eraserhead1 <
talk> 21:39, 19 April 2011 (UTC)reply
The rationales advanced in favour of this proposal are decidedly unconvincing.
Skomorokh 11:03, 23 April 2011 (UTC)reply
There is no reason a crat should need to view deleted edits.
Zginder 2011-04-28T01:56Z (
UTC)
I see absolutely no need for 'crats to have viewdlete privileges taking into account the situations that bureaucrats handle....unnecessary. mauchoeagle(
c) 19:48, 9 May 2011 (UTC)reply
Neutral
I don't have anything against this proposal, but at the moment all 'crats have admin tools as well. And I don't think anybody can become a 'crat without being an admin first.
ArmbrustTalk to meContribs 18:33, 12 April 2011 (UTC)reply
Bit of a catch-22, no? All 'crats current have admin tools because (it is so posited that) they need to view deleted revisions (and
more) to effectively do their job. If the necessary privileges were available in the bureaucrat package, a bureaucrat might decide to relinquish their administrative privileges while still serving as a bureaucrat. Thus, one of the reason all 'crats currently have admin tools is because they can't give them up without becoming less effective - a self-fulfilling prophecy. –
xenotalk 18:40, 12 April 2011 (UTC)reply
Maybe, but I don't see why a 'crat would want to remove their admin user right. I mean there's no policy (at least I don't know about it), which says you can't use some of the rights if you don't use them all. If you want to use only deletedhistory, deletedtext, and browsearchive from the admin package, then just simply don't use the other parts. I don't think, that it's too difficult. If many non-admins would be promoted to 'crats, than these question would be more important.
ArmbrustTalk to meContribs 21:56, 12 April 2011 (UTC)reply
A 'crat would want to remove the admin bit simply because they don't want to do the tasks anymore. If you have the bit, and you're active somewhere, people will still be asking you to do admin stuff.
demize(
t ·
c) 02:23, 13 April 2011 (UTC)reply
I suppose it is a sensible argument that if a crat needed to see deleted material, they could temporarily give themselves the admin bit. --Errant(
chat!) 10:25, 13 April 2011 (UTC)reply
There have been multiple cases in the past where crats have relinquished their admin rights while maintaining their crat bits. (
X! ·
talk) ·
@843 · 19:14, 13 April 2011 (UTC)reply
Question can a 'crat make himself an admin?
Zginder 2011-04-28T03:18Z (
UTC)
They have the technical ability to do so, but there are very few situations where that would be appropriate.
28bytes (
talk) 03:22, 28 April 2011 (UTC)reply
That would mean that a 'crat must have passed a RFA and not have lost the mop in disgrace. Otherwise a 'crat could give itself powers the community does not.
Zginder 2011-04-28T03:27Z (
UTC)
More generally, 'crats have the technical ability to give anyone the sysop bit, but again, they only do so under prescribed circumstances. Just to put the proposal here in context, it has been suggested that some 'crats would like to relinquish their sysop bits to focus on 'crat work full-time. This proposal would make that technically possible, since there's currently no policy reason a 'crat couldn't relinquish the sysop bit but continue to perform 'crat work.
28bytes (
talk) 03:30, 28 April 2011 (UTC)reply
I have no problem with a 'crat giving up the mop in fact
I would like to see it required, but they must have still passed an RFA (and not have been desysoped in disgrace), if they could at any time give the mop to themselves. Also I do not think that 'crats need to see deleted pages to do their job.
Zginder 2011-04-28T04:09Z (
UTC)
Discussion
Was requested in
bugzilla:25752; a developer requested a clear consensus discussion be held prior to committing the change. –
xenotalk 15:13, 12 April 2011 (UTC)reply
I'm not opposing, but Bureaucrats don't really need to view deleted material as it is not in their "job spec". I don't buy the argument that they would often (if at all) need it for RFA closures because they are only judging consensus and not making a decision on the suitability of the candidate themselves. Also; there is not practical likelihood that a non-Admin would become a bureaucrat any time soon --Errant(
chat!) 15:42, 12 April 2011 (UTC)reply
As regards deleted edits and RfA, see my below reply to Jafeluv. Viewing deleted edits may also be necessary when determining whether or not to carry out a rename (i.e. to see if the user has been making promotional edits that were since deleted). While the community may never endorse a non-admin standing at RfB, an existing bureaucrat might wish to resign their administrative privileges. Please try to focus on the technical aspect of the proposal - assume that someone is a bureaucrat but not an administrator: should they be able to view deleted revisions? –
xenotalk 15:55, 12 April 2011 (UTC)reply
While I have nothing against this, I don't quite see how closing an RfA would require looking at deleted contributions. After all, the closing crat is supposed to evaluate the discussion, not the candidate. Not sure I see a clear relevance to user renaming, either.
Jafeluv (
talk) 15:45, 12 April 2011 (UTC)reply
Sometimes opposition is raised in related to a candidate's speedy deletion tagging, and sometimes administrators opposing reference specific deleted edits - viewing the deleted contributions of a candidate may be required to verify statements made by administrators opposing on these ground. –
xenotalk 15:55, 12 April 2011 (UTC)reply
I am inclined to oppose this pending an explanation as to why it is necessary for the role. I don't buy the argument in the nomination for a minute, so unless these rights are essential to the CHU or bot flagging aspects of the role, this would seem an unwelcome conflation of unrelated functions. Could bureaucrats comment on how central these tools are to their work currently? Thanks,
Skomorokh 15:56, 12 April 2011 (UTC)reply
I am a bureaucrat; see my above remarks. I wouldn't say that it is central to the role, but I have used it numerous times when determining whether to carry out a rename (e.g. Has this user been _so_ promotional with their prior edits that their rename should be declined and they should simply be blocked?) and I have viewed the deleted contributions of RfA candidates prior to closing the request (e.g. Does this user
really use A1 erroneously far too often, even though the targeted articles may eventually be deleted under more applicable criteria?). –
xenotalk 15:57, 12 April 2011 (UTC)reply
I think all bureaucrat work can and should be done entirely based on open discussion or requests. I doubt that you'd need to check deleted contribs to see if a user excessively engaged in promotion, you just need to check his user talk page. Making further investigation, considering a block, would be an admin job, not bureaucrat job. And when closing an RFA, bureaucrats should consider the opinions provided, not directly investigate the evidence.
Cenarium (
talk) 16:10, 12 April 2011 (UTC)reply
As someone who actually performs the work in question, I disagree - but respect your opinion nonetheless. –
xenotalk 16:13, 12 April 2011 (UTC)reply
Well that's my opinion on how the work should be done, but of course others may have different opinions.
Cenarium (
talk) 16:17, 12 April 2011 (UTC)reply
I agree that the bureaucrats should be guided by the community's comments in closing RfAs (and not their own investigations), but verifying that a statement based on a deleted edit is an accurate representation of the situation (rather than taking it at face value) is still an important task. –
xenotalk 16:18, 12 April 2011 (UTC)reply
Yes it's an important task, but one that can be taken by the users (with view-deleted rights) participating in the RFA. I feel that all of the bureaucrat's reasoning should be based on the public discussion.
Cenarium (
talk) 16:28, 12 April 2011 (UTC)reply
Fair enough - reasonable minds can differ on this. –
xenotalk 16:31, 12 April 2011 (UTC)reply
Please see the oppose and neutral section of
this RfA. If I had closed that RfA, I would have reviewed the deleted contributions of the candidate while determining the outcome of the discussion, in particular, the red links posted by Moonriddengirl in the neutral section. –
xenotalk 16:07, 12 April 2011 (UTC)reply
I think bureaucrats should base their decision on only what is public, the opinions provided by the users, so that all users can follow their reasoning and interpretation of consensus. Investigating deleted evidence on one's own goes beyond the role of bureaucrat, they should refer to the comments made on it.
Cenarium (
talk) 16:14, 12 April 2011 (UTC)reply
Should they take any comment based on a deleted edit at face value, or ensure that the comment is an accurate representation of the situation by viewing the deleted edits themselves? –
xenotalk 16:21, 12 April 2011 (UTC)reply
Replied above to avoid duplicating discussion.
Cenarium (
talk) 16:28, 12 April 2011 (UTC)reply
No one has mentioned it but part of our copyright policy is that although copyright violations are not truly deleted, only a certain small and approved group can see them. We need to carefully consider expanding this ability to another group that may not truly need the ability lest granting access to this material through several processes and to diverse groups dilutes our defense. Adding access to oversighters seems obvious, though.
Rmhermen (
talk) 16:55, 12 April 2011 (UTC)reply
Given that the number of non-admin bureaucrats is unlikely to ever exceed the total number of current bureaucrats, the number of current bureaucrats is a tiny fraction of the current administrator corps, and under the current practice those (currently hypothetical) non-admin bureaucrats would be removed from the admin corps, I see it as close to a zero-net-sum item. Right now, every bureaucrat can see deleted revs. If Xeno gets his wish and is allowed to keep being a bureaucrat without the admin bit, that's a net change of... zero.
Jclemens (
talk) 01:01, 13 April 2011 (UTC)reply
The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
Add move-related privileges to Bureaucrat
Clear consensus in favour of implementationHappy‑
melon 10:31, 15 May 2011 (UTC)reply
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
In order to remove the technical limitation that
bureaucrats must also be administrators to properly carry out renames, the move-subpages, suppressredirect, and tboverride rights should be added to the bureaucrat permission group.
Support
Proposed. It is important that a Bureaucrat have the ability to move subpages, to move pages without creating a redirect, and to override the title blacklist, especially when carrying out renames. –
xenotalk 15:13, 12 April 2011 (UTC)reply
Required to make the role "self-sufficient" so sensible to add, even though from a practical perspective it won't be needed. --Errant(
chat!) 15:51, 12 April 2011 (UTC)reply
Necessary for the job.
Cenarium (
talk) 15:52, 12 April 2011 (UTC)reply
Makes sense. Otherwise bureaucrats would either need to be admins themselves or ask for an admin to assist when performing renames.
Jafeluv (
talk) 15:53, 12 April 2011 (UTC)reply
Bureaucrats should have the rights they need to do their jobs; right should be self-contained.
ErikHaugen (
talk |
contribs) 17:19, 12 April 2011 (UTC)reply
Unlike view-deleted privileges, this does affect the core functionality of bureaucrats, and so I support. --
Avi (
talk) 20:04, 12 April 2011 (UTC)reply
Personally, I would never support someone to be a crat without being a sysop first, but at least moving pages is relevant to what a crat does (renames). /
ƒETCHCOMMS/ 21:44, 12 April 2011 (UTC)reply
Seems like it might be unnecessary, but also little harm in it.
Beyond My Ken (
talk) 22:15, 12 April 2011 (UTC)reply
I can think of an editor who would more easily pass RFB than RFA. Might as well make this possible. ϢereSpielChequers 22:17, 12 April 2011 (UTC)reply
Straightforward. Bureaucrats need these privileges in order to rename users.
Ucucha 23:26, 12 April 2011 (UTC)reply
If bureaucrats have the ability to rename users, then it makes sense for them to be able to move those pages, as well. The reason this is not the case is because, in the MediaWiki defaults, sysops normally hold right to rename users. Nobody probably bothered to look at the fact that this right was not given to bureaucrats because it's our community norm that bureaucrats be sysops first when, in actuality, they need not be. –
MuZemike 00:38, 13 April 2011 (UTC)reply
Trust the 'crats on this one. Again, I believe that each toolkit/user right should have the tools required to operate independently from any other permissions.
Risker (
talk) 02:34, 13 April 2011 (UTC)reply
Unlike the privileges related to viewing deleted material above, this set of move-related privileges directly affects the ability of bureaucrats to carry out their core duties following an open consensus discussion.--
Danaman5 (
talk) 02:41, 13 April 2011 (UTC)reply
Definitely - the 'crat ability should be self-sufficient, and all 'crat jobs should be doable in full without any other user rights. This is some times necessary for user renames.
עוד מישהוOd Mishehu 07:03, 13 April 2011 (UTC)reply
Can't see the slightest bit of harm in it, and even if it's not needed today, it may be needed tomorrow. Much better to have something and not need it than to need it and not have it.
Pesky (
talk) 08:04, 13 April 2011 (UTC)reply
I fine with this technical change being done: while it would be rare, one of the bureaucrats may not want to remain as an administrator, or merely want to take a break from being one, but are still happy doing the bureaucrat tasks.
Acalamari 11:39, 13 April 2011 (UTC)reply
Xeno hits the nail on the head. I've thought about this one problem before, actually. --
(ʞɿɐʇ) ɐuɐʞsǝp 14:54, 13 April 2011 (UTC)reply
Could be significant in the context of a rename. The tool we're discussing is easily reversed (like all admin tools) if misused. If we trust someone to promote admins, then by implication we probably trust them to at least handle the related page moves properly. Candidates seen as having a problem in this area would quickly find themselves opposed at RFB.
FT2(
Talk |
email) 10:30, 15 April 2011 (UTC)reply
Agree it is useless. Another solution in search of a problem.
Ruslik_
Zero 19:08, 12 April 2011 (UTC)reply
See my replies to Juliancolton below. –
xenotalk 19:36, 12 April 2011 (UTC)reply
It should be that bureaucrats must also be administrators, thus nullifying the premise. --
SmokeyJoe (
talk) 09:57, 13 April 2011 (UTC)reply
Although people with advanced permissions tend to deny it quite vocally, there is a class system in Wikipedia. Although people tend to accept it but can't agree on how to fix it, RfA is a horrid system at this point. I think Pesky inadvertently responded to you in the support section too, by the way.
Sven ManguardWha? 20:33, 13 April 2011 (UTC)reply
Discussion
Was requested in
bugzilla:25752; a developer requested a clear consensus discussion be held prior to committing the change. –
xenotalk 15:13, 12 April 2011 (UTC)reply
Leaning towards supporting this, but pausing per what I said below.
Ncmvocalist (
talk) 16:23, 12 April 2011 (UTC)reply
Renameuser extension apparently allows bureaucrats to move all user subpages of the renamed user. So, move-subpages userright is not necessary.
Ruslik_
Zero 08:41, 19 April 2011 (UTC)reply
The userright will still be useful to cleanup in case a particular rename hangs. –
xenotalk 14:10, 19 April 2011 (UTC)reply
In this case all restrictions discussed below will apply.
Ruslik_
Zero 07:35, 20 April 2011 (UTC)reply
The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
General discussion
I think these are good proposals and am very much in favor especially for the first two, but on the other hand, is there any reason for not having these "higher" userrights restricted to admins? I don't mean that we should prevent non-admins from applying for checkuser rights, for example, but if an editor can be trusted with checkuser tools, can they not also be trusted with the "regular" mop?
Zakhalesh (
talk) 15:24, 12 April 2011 (UTC)reply
How is adminship not a social requirement for bureaucratship? I admit, I don't get around to RFA much, but in my experience it is very much a social requirement. Unless you meant to say that adminship is not a technical requirement for bureaucratship, in which case you would be correct.--
Danaman5 (
talk) 15:34, 12 April 2011 (UTC)reply
There is no requirement that an editor must be an administrator prior to their standing for bureaucratship, and nothing preventing a bureaucrat from resigning their administrative privileges while retaining their bureaucrat rights. Please do keep in mind that best practices suggest that user groups should be self-sufficient and self-contained, and making the technical changes to the bureaucrat privilege would not prevent the community from opposing candidates at RfB based on their not having successfully completed RfA. –
xenotalk 15:38, 12 April 2011 (UTC)reply
I understand your point, but when you say it is not a "social requirement", it implies to me that editors who apply for bureaucratship without first being an administrator are not opposed based on that fact (i.e. their doing so is accepted by the community), but that does not seem to be the case to me. Perhaps just a case of unclear wording.--
Danaman5 (
talk) 15:41, 12 April 2011 (UTC)reply
That the community has not yet bestowed bureaucratship on an editor who has never been an administrator does not mean that it will not one day occur, and there is no policy-based (i.e. social) requirement that one be an administrator prior to standing for bureaucrat. Furthermore, more than one bureaucrat has indicated a desire to step down from adminship while continuing to serve in their capacity as a bureaucrat. This proposal strictly seeks to make higher-level userrights self-sufficient and self-contained, and does not seek to change the community's attitude towards a non-administrator standing for bureaucratship. –
xenotalk 15:46, 12 April 2011 (UTC)reply
This provides a separable ability for there to exist a 'cat-without-sysop, which seems like a quite useful combination. Even if the community decided that was a bad combination, such a prohibition could be made on a policy basis, so I see no particular harm done by implementing this even in that case. And the community may very well wish to consider making some crats-without-sysop going forward in some cases, e.g., if renameuser backlogs got to be a problem. --
joe deckertalk to me 15:56, 12 April 2011 (UTC)reply
The community may very well wish to do a lot of things; but there's a time and a place - if such a situation arises which would justify the need for this, then I'd prefer to implement it after knowing what the circumstances are and/or what the wiki climate is like at that time.
Ncmvocalist (
talk) 16:13, 12 April 2011 (UTC)reply
Mmmm, it's true that I phrased that in a very "someday" sort of way, and that that part of my argument is premature. However, it sounds to me as if Xeno has considered and may still be considering asking to drop his sysop bit but not his 'crat bit. I don't think it's premature to ask what that would look like. Moreover, Xeno is right as near as I'm aware about one thing, we have no policy that says you can't have that combination of bits. As I'm more of a "that which is not prohibited is allowed" rather than a "that which is not specifically allowed is prohibited" kinda guy, I think it does seem worth asking "what would 'crat-sans-sysop" look like now as a result, whether that discussion happens here or elsewhere. *shrug* --
joe deckertalk to me 16:20, 12 April 2011 (UTC)reply
What about the fact that if view-deleted rights were added to bureaucrat rights, this would make it a bigger deal for non-admins to pass RFB ? Yes it doesn't seem needed to me.
Cenarium (
talk) 16:31, 12 April 2011 (UTC)reply
I think that's a red herring. Each role should have all the required privileges needed to do any of the tasks required by that role, without relying on any other (presumably subordinate or prerequisite) role. If roles are considered "social" prerequisites, then the addition of these privileges will create no new risks, because each of these proposals overlap with what has already been required.
Jclemens (
talk) 16:52, 12 April 2011 (UTC)reply
I don't consider that viewing deleted pages is needed for the bureaucrat role, but never mind.
Cenarium (
talk) 17:01, 12 April 2011 (UTC)reply
Unless I'm dreadfully misreading the proposal, I don't see that there's any real problem to go along with these solutions. People fail RfB for being an admin for only 10 months, yet we're going to set up non-admins to run?
Juliancolton (
talk) 18:58, 12 April 2011 (UTC)reply
The problem is that the community
seems to largely agree that adminship is not a social requirement for certain higher privileges (nor should it be), but it is currently a technical requirement because some rights necessary for the tasks are inherited from the administrative package. Try to focus on the technical aspects of the proposal and set aside your worries that we might have proposed a solution without a problem - even if this were the case, there is no inherent harm that would be caused by the technical change. –
xenotalk 19:07, 12 April 2011 (UTC)reply
But it's more than just a technical issue. Userrights shouldn't be self-sufficient; they should compliment each other depending on what a given user is interested in doing on the project. It's true that there's not a huge risk in implementing these proposed changes, but I guess I just feel it's a lot of confusion for an as-of-yet unverified hypothetical situation.
Juliancolton (
talk) 19:25, 12 April 2011 (UTC)reply
I don't make these proposals because I like arguing with users who think the solutions I am proposing have no problems to go along with them. In 2010, I actually intended to relinquish my administrative privileges after
bugzilla:25752 was fulfilled, and I know of at least one other bureaucrat who has considered doing so as well. –
xenotalk 19:31, 12 April 2011 (UTC)reply
Are you hesitantly apologizing, or indicating confusion at my response? –
xenotalk 19:38, 12 April 2011 (UTC)reply
"Userrights shouldn't be self-sufficient; they should compliment each other depending on what a given user is interested in doing on the project" <-- Does this mean you think that the privileges found in 'rollback', 'reviewer', 'file mover', 'autopatrolled', & 'account creator' should be removed from the administrative rights package and sought individually by administrators? If not, is there a reason that userrights redundancy is desirable in the administrative package but not other higher privileges? –
xenotalk 19:51, 12 April 2011 (UTC)reply
To be honest, I really don't care either way with most of these. I'm not (and never have been), a big "bugzilla" reader anyway. It occurs to me that either you "trust" someone, or you don't. I know the whole Arb/Crat/Admin/Editor thing isn't supposed to be hierarchical in structure, but in practice it often works out that way. Although, I can see where "Crat" is more of a horizontal position to "Admin" in function. Just don't recall ever seeing some of these rights being assigned to users who aren't members of a particular group. — Ched :
? 22:58, 12 April 2011 (UTC)reply
Comment Xeno, it sounds like you really want to go the whole hog and allow turning each permission bit on and off separately. There is some logic to that at lower user levels (like rollback or reviewer), which are mostly about competence rather than trust. At the upper levels, you're proposing to separate out what was a few roles activated by well-established processes, into a bit vector with a configuration space of exponential size in the number of bits, with individual bits being potentially given out on an ad hoc basis. Switching from roles to bits is potentially pretty drastic and I think should take considerable mulling over before implementation. I'd be ok sticking with roles (adding new ones like "auditor" if necessary), and allowing role-holders (who got the roles through normal processes) voluntarily relinquish bits that they received as part of their roles. (But if want bureaucrat-like duties without sysop bits, why not run for Steward?).
If I understand SlimVirgin and Cenarium's concerns, it's something like a
code is law worry that anything that any capability enabled by the software will be used and then entrenched, if it delivers the past of least resistance to some ephemeral goal once or twice. Looking at it this way, and with the increased complexity of the current RFC, I'm uncomfortable with having people acquire the bits through less formal processes than what we have now, therefore have reservations about putting the technical framework in place for it. I suppose this is a reversal from what I was saying earlier.
69.111.194.167 (
talk) (new address) 00:54, 13 April 2011 (UTC)reply
No... that is not really what I am asking for here, though I think that is what MZMcBride is
suggesting on the talk page. I don't really follow your "bit vector" example, but I don't think that is what I am asking for either. I am asking for a few user groups to receive permissions they require for their role - permissions which are usually inherited by virtue of most higher permission holders also being administrators. While this has worked until now (since it is rare for a non-administrator to seek higher permissions, and rarer still for them to be granted), there is no harm in remedying a situation that has largely come about by organic growth (the higher permissions were apparently put together quickly on an urgent need basis). If I relinquished my administrator rights, I could still rename users, but I wouldn't be able to move their subpages during the rename. That is just a technical glitch that needs to be overcome, and that is what this proposal seeks to remedy. (As for your bracketed suggestion - Stewards are not permitted to act on their local wiki, and indeed are not to carry out renames with their steward hat on projects with bureaucrats.)
At the risk of sounding repetitive: Not everyone wants to be an administrator, and not everyone wants to be an administrator forever. Further, 'administrator' is not some kind of "super user" or prerequisite for advanced privileges, nor does the RFA process or the flipping of the administrator bit suddenly make a user any more or less trustworthy than they were prior to them receiving administrative privileges. –
xenotalk 14:06, 13 April 2011 (UTC)reply
The abusefilter right seems to me to be a very obvious example of why the present system is absurd. Almost all of the current admins were "promoted" before those filters existed, and hardly any understand how to write regular expressions. The present system of packaging user rights is simply lazy, and a historical artefact.
MalleusFatuorum 02:24, 14 April 2011 (UTC)reply
Xeno, I guess I wonder what you think "administrator rights" means, in terms of relinquishing them. Do you mean just the technical capabilities of +sysop that aren't in +bureaucrat? For example, say you're not on arbcom, you're still a bureaucrat, and you've relinquished "administrator rights". There is an AE request about user XYZ's editing in some articles under discretionary sanctions. Can you close the request by placing user XYZ under a topic ban, notifying him on his talk page, and updating the appropriate logs? That doesn't use any sysop buttons, but it's an action reserved for "uninvolved admins". So I'm wondering if you'd consider yourself still authorized to do that. Me, I think adminship is no longer about "buttons" and RFA has a higher bar than before, because admins have greater authority than before, in areas like AE/ANI/etc. We never hear about "rouge admins" any more, since the out-of-process "rouge" stuff they used to do to keep the site running, has since morphed into recognized and accepted processes. But that makes adminship a much "bigger deal" than before. As I see it, the big deal is authority, not buttons.
69.111.194.167 (
talk) 03:45, 14 April 2011 (UTC)reply
I think that is somewhat off-topic (a philosophical discussion about what it means to be an administrator probably belongs elsewhere) - nevertheless - no, if I turned in my administrative privileges, I wouldn't close a request at AE because it calls for uninvolved administrators to close requests. –
xenotalk 12:20, 14 April 2011 (UTC)reply
Comment. This proposal, even if adopted, will not make bureaucrats self-sufficient. They still will not be able to move user css and js subpages without 'editusercss' and 'edituserjs' userrights. They are also will not be able to move move-protected userpages without 'protect' userright.
Ruslik_
Zero 19:01, 14 April 2011 (UTC)reply
Hadn't thought about that - thank you for pointing that out. This is probably something that could be dealt with "under the hood", so to speak - rather than adding those rights to the proposal. –
xenotalk 19:37, 14 April 2011 (UTC)reply
It would be a good idea to include them in this, but now it is a hassle to add after so many have voted. Ruslik0, are there any more rights that would be used? What about the user editnotices?
Graeme Bartlett (
talk) 21:45, 14 April 2011 (UTC)reply
I wonder if a separate RfC would be appropriate that posed the question more directly: "Should admins who become bureaucrats be allowed to later relinquish their adminship but retain the technical abilities needed to fully function as a bureaucrat?" I think the current proposal assumes the answer is yes, and while it might be the case that the community agrees with this (I would support it personally), I think the lack of context and the tradition that bureaucrats have always been admins first is leading a lot of people to assume this proposal has no practical point given that all 'crats are (currently) admins.
28bytes (
talk) 00:04, 15 April 2011 (UTC)reply
How could they miss that context with me clamouring about it all over the page? ;> I think the simplest solution to the problem highlighted by Ruslik would be to ask the developers to make the RenameUser interface move all the target user's pages regardless of protection status or whether they end in .js or .css. –
xenotalk 13:03, 15 April 2011 (UTC)reply
Of course, the RenameUser interface can be made to move js/css files as well as to move subpages, not to create redirects etc. However, this also would make the proposal under discussion meaningless. The truth is that without those userrights that I mentioned above your dream of being a purified bureaucrat will remain just a dream. It seems impossible to be an effective bureaucrat without being also a sysop.
Ruslik_
Zero 19:29, 15 April 2011 (UTC)reply
How so? The discussion seems to confirm that users do not have an issue with bureaucrats continuing to rename users while not being sysops. As such, the software should be fixed to allow this. (In any case, the majority of users who seek rename do not have custom js or css and do not have moveprotected pages, based on my observations since you've mentioned this.) –
xenotalk 19:34, 15 April 2011 (UTC)reply
Every time you are renaming an account you will need to check whether there exist js/css subpages (this is simple) and whether some subpages are move-protected (more difficult). Otherwise you may end up scattering various subpages over two accounts. The system likely will not display any warnings. And if you find them you will need to ask somebody else to do the job. So, the life of the hypothetical purified bureaucrat will not be easy.
Ruslik_
Zero 19:49, 15 April 2011 (UTC)reply
Yes, which is why the RenameUser extension should be fixed to not be interdependent on other userrights. –
xenotalk 20:02, 15 April 2011 (UTC)reply
The likely result is WONTFIX.
Ruslik_
Zero 13:12, 16 April 2011 (UTC)reply
For what reason? –
xenotalk 13:16, 16 April 2011 (UTC)reply
Developers have a lot of more pressing issues to deal with. This is simply a non-issue from the technical point of view.
Ruslik_
Zero 17:06, 16 April 2011 (UTC)reply
I don't follow your logic; but as Happy-melon has helpfully explained, this section of the discussion is moot because the RenameUser extension already largely operates how it ought (i.e. independently of lower-order userrights). –
xenotalk 19:06, 18 April 2011 (UTC)reply
The 'editusercss', 'edituserjs' and 'protect' permissions are not tested for in the RenameUser extension, only that the user has the 'move' permission. To use the terminology of this thread, it has already been 'fixed'. If it hadn't, it would have been a one-line change to one boolean parameter to make it so, and would have been something I'd very happily have done. Developers are, in general, volunteers and work on whatever interests them. Members of other projects should generally not try to either second-guess, or dictate, what those interests are or should be.
Happy‑
melon 10:02, 17 April 2011 (UTC)reply
I appear to be wrong. However, I do not understand why it checks 'suppressredirect' but not other three permissions? Is there any logic here?
Ruslik_
Zero 16:02, 17 April 2011 (UTC)reply
I guess because it's a whole feature in itself, not a partial feature; if you have 'suppressredirect' it shows the checkbox and lets you do it; if you don't it doesn't even let you try. Similarly the option to move pages is only available if you have the 'move' permission. Incidentally, the 'move-subpages' permission is not checked either, so I don't really know why that's included in this discussion (presumably innocent ignorance in the same way).
Happy‑
melon 17:32, 17 April 2011 (UTC)reply
I does not appear to check $wgMaximumMovedPages as well. So, any number of subpages can be moved. It also does not check if the user is blocked. So, blocked bureaucrats can still rename users.
Ruslik_
Zero 06:32, 18 April 2011 (UTC)reply
Hmm... not limiting the number of pages moved is correct, I would say. Not checking blocked status is definitely not; fixed that in
r86369.
Happy‑
melon 23:47, 18 April 2011 (UTC)reply
Ok, I noticed that in title.php isValidMoveOperation checks whether the target title is protected. This check is not influenced by $auth.
Ruslik_
Zero 07:27, 19 April 2011 (UTC)reply
Thank you for explaining that, Happy-melon. Yes, I (and Ruslik, it seems) presumed that because 'suppressredirect' was checked, the other rights would be checked as well. –
xenotalk 19:01, 18 April 2011 (UTC)reply
Wouldn't it be better/easier to grant the
researcher right to all crats/CUs/OS non-admins instead of adding viewdeleted privileges to these user groups, in order to prevent software's internal redundancy (in view of the fact that a lot of administrators are also bureaucrats and/or checkusers)?
Ruy Pugliesi◥ 04:11, 1 May 2011 (UTC)reply
Short answer, Researcher is Foundation-administered, not Arbcom-administered.
Jclemens (
talk) 04:30, 1 May 2011 (UTC)reply
Ok. We could implement a similar "ArbCom -administered" permission or authorize "researcher flag" being administered in this way, instead of adding more privileges to crats and checkusers. For example: on Portuguese Wikipedia, there is the
eliminator, which allows non-admins to view deleted contributions and also delete/restore pages.
Ruy Pugliesi◥ 20:37, 1 May 2011 (UTC)reply
From the better-late-than-never department: researcher does not allow one to view deleted revisions (
Special:UserGroupRights). –
xenotalk 15:32, 7 September 2011 (UTC)reply
Addendum (March 2013) : As per
this declaration, the WMF requires a RFA or RFA-identical process for being granted rights with access to deleted revisions, including explicitly OS and implicitly CU.
In view of the fact that
adminship is not presently a social requirement for higher privileges such as
CheckUser,
Oversight, and
Bureaucrat, it is proposed that these user groups be made self-sufficient by adding certain rights (currently found in the administrative toolset) to each bundle.
This discussion is intended to focus on addressing the technical limitations present in the
current makeup of user groups/rights. The related question - whether the administrator privilege should be a social requirement for such privileges - is being discussed
elsewhere, and carrying out this technical change would not preclude the ability of the community to implement such a requirement if consensus were found for the same.
Discussion initiated by –
xenotalk 15:13, 12 April 2011 (UTC)reply
Add viewdeleted privileges to CheckUser
Comprehensive consensus in favour of implementation.
Happy‑
melon 10:28, 15 May 2011 (UTC)reply
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
In order to remove the technical limitation that
checkusers must also be administrators to review deleted content, the deletedhistory, deletedtext, and browsearchive rights should be added to the checkuser permission group.
Support
Proposed. It is important that a CheckUser have the ability to view deleted content, as it may inform their findings during an investigation. –
xenotalk 15:13, 12 April 2011 (UTC)reply
In favour of proposal. CheckUser involves a high degree of trust already - I'd see no harm in allowing them to see deleted revisions.
Zakhalesh (
talk) 15:20, 12 April 2011 (UTC)reply
Agree that the view-deleted rights are necessary for CU work.
Cenarium (
talk) 15:30, 12 April 2011 (UTC)reply
I don't see any reason a non-admin checkuser shouldn't be able to view deleted content to assist with their investigations.
28bytes (
talk) 15:36, 12 April 2011 (UTC)reply
Good idea. Viewing deletions is no big deal, CU's are already established and respected members of the community. --Errant(
chat!) 15:38, 12 April 2011 (UTC)reply
Can't see why not.
Jafeluv (
talk) 15:41, 12 April 2011 (UTC)reply
Strong Support Thinking about
SPI, it would be really useful, the wikimedia foundation has trusted them with private information so they should be able to trust them like that. ~~
EBE123~~ talkContribs 18:56, 12 April 2011 (UTC)reply
Per my reasoning at the Village Pump. --
Avi (
talk) 20:00, 12 April 2011 (UTC)reply
Support It's assumed that people who are requesting checkuser have already been vetted by a collection of trustworthy editors and that the community at large has been given time to review the vetting and express it's thoughts about the candidate.
Hasteur (
talk) 20:01, 12 April 2011 (UTC)reply
IMO, this entire discussion is quite silly. I agree completely with xeno. If somebody's job requires them to do certain tasks, but they aren't permitted to do those tasks by their job, the entire system they're in is broken.
demize(
t ·
c) 02:18, 13 April 2011 (UTC)reply
While I won't say it is silly, it is just good systems management to ensure that each toolkit/user right has the tools to permit it to operate discretely. The current situation is largely a historical accident, because the initial user right was developed on the fly to reduce unnecessary workload on developers, and was never corrected.
Risker (
talk) 02:30, 13 April 2011 (UTC)reply
I have no problem with a non-sysop checkuser.
Reaper Eternal (
talk) 02:46, 13 April 2011 (UTC)reply
Checkusers are trusted to view non-public material, and this is an extension of that trust.--
Danaman5 (
talk) 02:48, 13 April 2011 (UTC)reply
Believe anyone trusted enough to be given this right, could also be trusted with this needed extension.
CalmerWaters 05:13, 13 April 2011 (UTC)reply
They are trusted to view much more than this; and viewing this may be necessary to determine when a check is needed.
עוד מישהוOd Mishehu 06:02, 13 April 2011 (UTC)reply
Absurd that it could be otherwise. --
SmokeyJoe (
talk) 09:50, 13 April 2011 (UTC)reply
I am also surprised that this is not already the case, and makes sense to allow non-admins to be checkusers.
CT Cooper ·talk 09:53, 13 April 2011 (UTC)reply
Checkusers already have a very high degree of community trust and should have access to deleted material.
Kudpung กุดผึ้ง (
talk) 05:12, 15 April 2011 (UTC)reply
Deleted material access is needed to do CU properly, and the selection process implies that they are trusted to view IPs which are far more sensitive. The remaining question is whether non-admin checkusers should be allowed and that's a social question, inclined to agree the tools should be self-sufficient either way (which doesn't affect that decision) - not least to support non-admin
audit subcommittee members.
FT2(
Talk |
email) 10:20, 15 April 2011 (UTC)reply
Obvious Support If they can be trusted with CU, why not seeing deleted revs.? --
Patar knight - chat/contributions 01:22, 16 April 2011 (UTC)reply
With the variety of situations CUs go through it is most definitely needed to view deleted revisions. In fact, I cannot think of a reason it has not already been done. mauchoeagle 04:54, 17 April 2011 (UTC)reply
Support—
mc10 (
t/
c) 02:05, 18 April 2011 (UTC)reply
What problem will this proposal solve? Are there a long queue of users who want to be checkusers but will not run through RFA? I can not support this solution in search of a problem.
Ruslik_
Zero 19:06, 12 April 2011 (UTC)reply
A user was recently appointed to the Audit Subcommittee and we found that they were unable to effectively investigate a case because of the inability to view deleted revisions. Further, there is nothing preventing a non-admin user from seeking advanced privileges; but without carrying out the technical change proposed herein, they would be hamstrung in carrying out their duties if they were appointed. –
xenotalk 19:18, 12 April 2011 (UTC)reply
You know it is kind of moot now.
Ruslik_
Zero 19:28, 12 April 2011 (UTC)reply
The present case will likely be rendered moot; this does not mean we should neglect to implement a technical solution for future similar cases. –
xenotalk 19:34, 12 April 2011 (UTC)reply
I think there should instead be an "auditor" role with the bundle of bits appropriate for the task.
69.111.194.167 (
talk) (was 75.57.242.120) 00:58, 13 April 2011 (UTC)reply
Sorry, but I really don't see the need to add another group, and as a consequence even more bureaucracy to this.
Ajraddatz (
Talk) 03:17, 13 April 2011 (UTC)reply
Checkuser requires the community's trust if they can not pass a RFA they do not have the community's trust.
Zginder 2011-04-28T01:55Z (
UTC)
Discussion
Was requested in
bugzilla:28440; a developer requested a clear consensus discussion be held prior to committing the change. –
xenotalk 15:13, 12 April 2011 (UTC)reply
The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
Add viewdeleted privileges to Oversight
Clear consensus in favour of implementationHappy‑
melon 10:29, 15 May 2011 (UTC)reply
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
In order to remove the technical limitation that
oversighters must also be administrators to review deleted content, the deletedhistory, deletedtext, and browsearchive rights should be added to the oversight permission group.
Support
Proposed. It is important that an Oversighter have the ability to view deleted content for a number of reason; most importantly because problematic content is typically deleted by administrators prior to being submitted for suppression. –
xenotalk 15:13, 12 April 2011 (UTC)reply
Support. Oversighters are trusted to delete the most sensitive material, so they should also be able to see it.
Zakhalesh (
talk) 15:22, 12 April 2011 (UTC)reply
Agree that view-deleted rights are necessary for OS work.
Cenarium (
talk) 15:30, 12 April 2011 (UTC)reply
Odd that they don't have those rights already.
28bytes (
talk) 15:36, 12 April 2011 (UTC)reply
As 28Bytes says, they should have had it already.. I mean, we trust them to scrub the most sensitive stuff for us! --Errant(
chat!) 15:39, 12 April 2011 (UTC)reply
It's part of their job to deal with objectionable content, so obviously they should be trusted to view it as well.
Jafeluv (
talk) 15:50, 12 April 2011 (UTC)reply
Userrights ought not need to be interdependent to function effectively.
Skomorokh 15:53, 12 April 2011 (UTC)reply
Per my reasoning at the Village Pump. --
Avi (
talk) 20:00, 12 April 2011 (UTC)reply
Support It's assumed that people who are requesting oversight have already been vetted by a collection of trustworthy editors and that the community at large has been given time to review the vetting and express it's thoughts about the candidate.
Hasteur (
talk) 20:10, 12 April 2011 (UTC)reply
Well, it would be rather silly if an oversighter had access to a "deeper" level of deletion and not the standard one.
sonia♫ 20:35, 12 April 2011 (UTC)reply
Makes sense, especially if many oversight requests are from sysops who have already (revision) deletion the pages or edits in question. –
MuZemike 21:22, 12 April 2011 (UTC)reply
It is just good systems management to ensure that each toolkit/user right has the tools to permit it to operate discretely. The current situation is largely a historical accident, because the initial user right was developed on the fly to reduce unnecessary workload on developers, and was never corrected.
Risker (
talk) 02:31, 13 April 2011 (UTC)reply
I have no problem with a non-sysop oversighter. Oversight is only for a very narrow set of issues.
Reaper Eternal (
talk) 02:45, 13 April 2011 (UTC)reply
Oversighters are trusted to view non-public material, and this is an extension of that trust.--
Danaman5 (
talk) 02:48, 13 April 2011 (UTC)reply
Very well can't properly oversight, as part of the duty is to ensure other oversighters are acting according to policy (due that they are the only ones to can view suppressed content) without being able to view deleted material
CalmerWaters 05:19, 13 April 2011 (UTC)reply
If they can be trusted to perform this highly confidential task ionvolving viewing the most secretive data posted on Wikipedia, they should be able to view all of it.
עוד מישהוOd Mishehu 06:05, 13 April 2011 (UTC)reply
Deleted material access is needed to do OS properly, and the selection process implies that they are trusted to view suppressed data which is far more sensitive. The remaining question is whether non-admin oversighters should be allowed and that's a social question, inclined to agree the tools should be self-sufficient either way (which doesn't affect that decision) - not least to support non-admin
audit subcommittee members.
FT2(
Talk |
email) 10:20, 15 April 2011 (UTC)reply
Weak support - It makes vague sense however I am not seeing any reason it is needed. OSers should have access to revisions that have been oversighted. As for anything else...I don't think so.mauchoeagle 05:03, 17 April 2011 (UTC)reply
Oversighters routinely have to address deleted revisions. Quite often oversightable material is also deleted. For example, some will have been deleted on sight by an admin, which correctly reduces the window for public viewing or spidering (eg a defamatory attack page under
CSD#G10 or a privacy breaching revision under
CRD#4). An oversighter then has to review the content and decide whether suppression applies. They also sometimes need to review suppressed edits or confirm if there has been any suppression, on a deleted page or in a user's deleted contributions. They can't reliably act as oversighters if they cannot see the deleted text.
FT2(
Talk |
email) 12:23, 17 April 2011 (UTC)reply
Support. —
mc10 (
t/
c) 02:06, 18 April 2011 (UTC)reply
I have been taking more of an active role in reporting things to oversighters, this should be obvious, Support.
TheHelpfulOne 16:25, 24 April 2011 (UTC)reply
Yes, a good idea. -- Cirt (
talk) 16:37, 9 May 2011 (UTC)reply
Replied above. –
xenotalk 19:18, 12 April 2011 (UTC)reply
Not really opposed on purely technical grounds (it's silly to let someone see oversighted edits but not deleted ones), but I'm no longer so sure it's ok to unbundle the technical and policy discussions (
code is law). I think the AUSC situation should be handlled by creating an "auditor" role with the bundle of bits necessary for the task, rather than turning individual capabilities on and off separately.
69.111.194.167 (
talk) (was 75.57.242.120) 01:01, 13 April 2011 (UTC)reply
I don't think non-admin oversighters are something we need/want.
Stifle (
talk) 10:00, 13 April 2011 (UTC)reply
Oversight requires the community's trust, if they can not pass RFAs they do not have the community's trust.
Zginder 2011-04-28T02:05Z (
UTC)
RfA has nothing to do with trust, it's a beauty civility contest.
MalleusFatuorum 02:26, 28 April 2011 (UTC)reply
Even if it is a civility contest, do you want an overseer to lose his temper and release secretive information to the world?
Zginder 2011-04-28T02:32Z (
UTC)
You're not making much sense. Are you suggesting that someone who's been through RfA is guaranteed not to lose their temper and do exactly the same thing?
No, we are never guaranteed anything. I am stating it is less likely.
Zginder 2011-04-28T03:08Z (
UTC)
Discussion
Was requested in
bugzilla:28440; a developer requested a clear consensus discussion be held prior to committing the change. –
xenotalk 15:13, 12 April 2011 (UTC)reply
The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
In order to remove the technical limitation that
bureaucrats must also be administrators to review deleted content, the deletedhistory, deletedtext, and browsearchive rights should be added to the bureaucrat permission group.
Support
Proposed. It is important that a Bureaucrat have the ability to view deleted content, as it may inform their decisions when closing a requests for adminship or bureaucratship, or while carrying out renames. –
xenotalk 15:13, 12 April 2011 (UTC)reply
This also makes perfect sense. As xeno has pointed out, determination of consensus at RFA sometimes involves reviewing deleted content. Having this right included as part of the package would help make it more clear that a bureaucrat need not necessarily be an admin, too. ···
日本穣? ·
投稿 ·
Talk to Nihonjoe ·
Join WikiProject Japan! 16:00, 12 April 2011 (UTC)reply
Probably irrelevant in the current climate of deification of administrators but sensible nevertheless.
MalleusFatuorum 16:06, 12 April 2011 (UTC)reply
This was the one I was least clear upon, but, I can see the AFD argument, and the particular rights included here do not strike me as worrysome. --
joe deckertalk to me 16:13, 12 April 2011 (UTC)reply
Even if bureaucrats don't use these rights enormously often, there is no real harm in adding them. Reviewing a lone oppose vote that made mention of CSD issues, reviewing a previously promotional editor—both of those are quite valid reasons that a non-admin bureaucrat might need to use these rights.
NW(
Talk) 16:26, 12 April 2011 (UTC)reply
This seems like the least compelling of the four proposals... but on the other hand, I can't see any particular harm in it either. Xeno's arguments below are not unreasonable.
28bytes (
talk) 16:30, 12 April 2011 (UTC)reply
Agree that deleted contributions are relevant to evaluating consensus at RfA.
Jclemens (
talk) 16:46, 12 April 2011 (UTC)reply
I think bureaucrats should be admins, but since that isn't the question here: because of Xeno's description below of how viewdeleted/etc is useful for 'crat jobs, these rights ought to be attached to the bureaucrat permission group.
ErikHaugen (
talk |
contribs) 16:54, 12 April 2011 (UTC)reply
I don't think it's my place to second-guess what the 'crats say they are using in their 'crat duties.
T. Canens (
talk) 17:01, 12 April 2011 (UTC)reply
Upon further reflection, I can see the benefit of a 'crat being able to review deleted edits when reviewing strength of arguments. And no, being a 'crat is not predicated upon being an admin. It is theoretically possible that one could become a 'crat without being an admin. It is also possible that one could relinquish admin responsibilities without reliquishing 'crathood. And it should be possible to have someone's admin rights removed, without removing their 'crat rights.---BalloonmanNO! I'm Spartacus! 19:46, 12 April 2011 (UTC)reply
I think this would rarely matter at RFA, but I can see it mattering when renaming a user. ϢereSpielChequers 21:33, 12 April 2011 (UTC)reply
This is one of the less clear ones, but I think people like NW above got it right: there is an occasional need for bureaucrats to be able to view deleted content.
Ucucha 23:24, 12 April 2011 (UTC)reply
Had to think harder about this one! But, on the whole, I can't see the slightest bit of harm in it, and even if it's not needed today, it may be needed tomorrow. Much better to have something and not need it than to need it and not have it.
Pesky (
talk) 08:05, 13 April 2011 (UTC)reply
Xeno convinced me. Ultimately no real reason this is a big issue. --Errant(
chat!) 08:56, 13 April 2011 (UTC)reply
There are certain instances in which bureaucrats might need to see deleted revisions, and with this being a trusted and small user group, extending this right to them should not be a major issue. I also don't think being an admin should be a requirement on becoming a bureaucrat, as the roles are rather different.
CT Cooper ·talk 10:02, 13 April 2011 (UTC)reply
There is no reason why as crat should not be allowed to resign their admin bit if they so choose and continue performing their functions as a crat. It is not a requirement for the position, and such a technical impediment should not be an underlying reason for the unwritten "requirement" that currently exists. Jim MillerSee me |
Touch me 14:32, 13 April 2011 (UTC)reply
I think all 'crats should already be (or have been) admins because the experience informs their reading of an RFA (case in point). Viewing a deleted diff that is under discussion also informs their reading of the discussion; any non-admin that has had difficulty understanding a discussion about deleted content until some administrator emailed them a copy should instinctively understand this point. But that should not preclude someone that was once an admin and gave up the admin bit, a real possibility that Xeno raises below.--
Chaser (away) -
talk 15:23, 14 April 2011 (UTC)reply
The argument in the oppose column that the rights would be redundant holds no water in the context of this RfC if the opposer at the same time supports adding move-related rights to the group. The decision is solely whether the viewdelete-related rights are required or useful for the role of a bureaucrat or not. With two crats for, two against, I can only assume that the rights under discussion are useful for the modus operandi of some crat's. Since I see no other reason against adding these passive rights to the crat group (the supposed WMF veto against making it available for very large groups of editors does not apply), I support it.
Amalthea 15:47, 14 April 2011 (UTC)reply
I can see that viewing a deleted edit may be important to evaluating consensus at an RfA. Hut 8.5 17:58, 14 April 2011 (UTC)reply
Could be significant in the context of a rename. The tool we're discussing is the ability to view deleted content, not to delete or undelete it. If we trust someone to promote admins, then by implication we probably trust them to at least view deleted posts. Candidates seen as having a problem in this area would quickly find themselves opposed at RFB.
FT2(
Talk |
email) 10:30, 15 April 2011 (UTC)reply
Obvious Support Crats are trusted. Crats must have great community trust to be crats. --
Patar knight - chat/contributions 01:25, 16 April 2011 (UTC)reply
Support Frankly can't see what the big fuss is about. If I trust someone as a b'crat, I trust him as a sysop. By contraposition, if I don't trust someone as an admin, I surely don't trust him as a bureaucrat. The proposed change just reflects a fairly uncontroversial hierarchy of levels of trust.
Pichpich (
talk) 14:59, 17 April 2011 (UTC)reply
Sure, seems sound and sensible. -- Cirt (
talk) 16:38, 9 May 2011 (UTC)reply
Oppose
Unlike the other two userrights, I don't consider this to be a mere technical limitation. If an admin has engaged in poor conduct, then I would not be willing for them to start/continue their work as a crat after their tools have been removed, forcibly or voluntarily.
Ncmvocalist (
talk) 15:54, 12 April 2011 (UTC)reply
I don't see how it is necessary for bureaucrats to view deleted content. All bureaucrat decisions should be based on open discussion or open requests. I do not think that bureaucrats should verify on their own the evidence presented in a RFA, instead they should base their decision on the opinions provided. I also don't see how deleted contribs would matter in a rename request, seeing if the user engaged in promotion is an administrator role, not a bureaucrat role. Moreover, if the community were to consider non-admin candidates for bureaucratship, the addition of view-deleted would make it a bigger deal than it should be, and we know bureaucratship is already too big a deal.
Cenarium (
talk) 16:05, 12 April 2011 (UTC)reply
Remember that the Bureaucrats must also be sysops so I think that this is a bit useless. ~~
EBE123~~ talkContribs 19:01, 12 April 2011 (UTC)reply
Must they be? –
xenotalk 19:18, 12 April 2011 (UTC)reply
Well, when you get bureaucratship, you also become a sysop, and if you didn't get it you can change your userrights to include sysop. ~~
EBE123~~ talkContribs 21:05, 12 April 2011 (UTC)reply
And what if a duly-appointed bureaucrat wants to continue to serve as a bureaucrat but wishes to relinquish their administrative privileges? –
xenotalk 21:09, 12 April 2011 (UTC)reply
Well then the user cannot do viewdeleted. Simple. ~~
EBE123~~ talkContribs 21:13, 12 April 2011 (UTC)reply
I do not see the need for bureaucrats to see deleted edits (speaking as one). If consensus cannot be determined through on-wiki postings, how are deleted posts going to help? Same with name changes and bot flags; unlike CU and OS, I don't see how these rights are integral to performing 'crat duties. --
Avi (
talk) 20:03, 12 April 2011 (UTC)reply
May not be 'necessary' today - but what about tommorow? I have a workshop full of tools, plus those I've inherited from my father, many of which I've never used. But, tell you what, on the day I have to use one, it's a bastard having to go out and buy a new tool when I need it in my hand right that minute! Same goes for a first aid kit - you don't need it until you need it - and then you need it :o)
Pesky (
talk) 08:16, 13 April 2011 (UTC)reply
Unless you can show a reason why a 'crat would need to do this in the future, I think that giving 'crats these rights shouldn't be done. If we give the 'crats more jobs later on, we can discuss it then.
עוד מישהוOd Mishehu 05:29, 14 April 2011 (UTC)reply
The non-sysop 'crat? One who doesn't want to by a sysop, but happy to be a 'crat? Even if we don't currently have one, having a system in place which means it would be difficult to be one seems counterproductive. And also somewhat pointless.
Pesky (
talk) 08:00, 18 April 2011 (UTC)reply
Per MZMcBride. Oh, and per Avi. There's little use, and I'm of the opinion that a crat should be an admin first to give it to others. /
ƒETCHCOMMS/ 21:43, 12 April 2011 (UTC)reply
Per Ncmvocalist and Avraham. Doesn't seem compatible as opposed to the other userrights. Not to mention we've already stripped most of the rights away from bureaucrats and relegated those to sysops or Stewards, virtually making them more or less figureheads in more of a technical sense (remember that closing RFAs is a social, not a technical, right). –
MuZemike 21:21, 12 April 2011 (UTC)reply
If they are really just "figureheads", than why it is so difficult to become crat?
ArmbrustTalk to meContribs 21:56, 12 April 2011 (UTC)reply
It was to answer your question in advance, I think, that MuZemike included a bracketed reminder.
Ncmvocalist (
talk) 22:34, 12 April 2011 (UTC)reply
No reason to do this since non-admins shouldn't become bureaucrats. If a bureaucrat+sysop wants to relinquish a subset of the bits they already have, I guess I'm ok with that. They shouldn't be able to acquire the bits other than through RFA/RFB in that order, at least without a lot more prior discussion specifically aimed at that question.
69.111.194.167 (
talk) 01:07, 13 April 2011 (UTC)reply
I believe that if we are going to open the possibility of bureaucratship becoming a more standalone role that does not necessarily require adminship, there needs to exist some differentiation with adminship, rather than it simply being a sort of "admin-lite". The difference should be that bureaucrats rely on open discussions to make their decisions, with the counsel of administrators and checkusers regarding deleted and private material. We shouldn't be thinking so much in terms of "self-sufficient" roles, and should instead consider the complementarity of roles.--
Danaman5 (
talk) 02:35, 13 April 2011 (UTC)reply
As 'crats promote admins, I think that a bureaucrat should be an admin first. This is especially true as admins attempt to assess consensus at AFD, etc.
Reaper Eternal (
talk) 02:44, 13 April 2011 (UTC)reply
Well, 'crats promoting admins isn't itself the issue, since stewards can also do that (without being local admins).
69.111.194.167 (
talk) 04:39, 13 April 2011 (UTC)reply
That is on the smaller Wikis that do not have local Bureaucrats. I can't foresee an instance where a steward would unilaterally promote an admin on EN; except to correct a de-sysops error.
CalmerWaters 05:04, 13 April 2011 (UTC)reply
Per above, I can see no situation in which a bureaucrat would not be an admin.
Ajraddatz (
Talk) 03:18, 13 April 2011 (UTC)reply
I don't see why a 'crat needs this - any relevant deleted data for RfAs should be made accessable to non-admins participating in the RfA.
עוד מישהוOd Mishehu 07:02, 13 April 2011 (UTC)reply
In the unlikely event that a non-admin becomes a 'crat being able to view deleted edits would have no bearing on their job. In the one or two cases where it would (I can't see more than that) asking an admin or leaving it for another 'crat would work just as well.
Sven ManguardWha? 07:11, 13 April 2011 (UTC)reply
This isn't a good idea. Crats should be admins because they need that experience to help them close RFAs. --Rschen7754 07:17, 13 April 2011 (UTC)reply
Silly. A bureaucrat must be an experienced administrator in good standing. --
SmokeyJoe (
talk) 09:53, 13 April 2011 (UTC)reply
Must one be? It may be the de facto standard, but in no way is it the de jure standard. Theoretically, I could go do an RfB and pass right now, according to policy.
demize(
t ·
c) 10:19, 13 April 2011 (UTC)reply
I believe the de factor standard should be hard written as de jure. Bureaucrats have very limited additional privileges, but what they do have places them clearly as leaders, even elites, and thus a subset, of the administrators. I support bureaucrats acquiring the power and role of de-admining (per a community consensus discussion), and, with a dim view on any individual wearing many hats, aka
separation of powers, support de jure demarcations, such as forbidding combined bureaucrat/arbiter --
SmokeyJoe (
talk) 13:32, 13 April 2011 (UTC)reply
I think that you'll find a lot of folks disagree with you - neither administrators nor bureaucrats are leaders or elites, nor should they be seen as such. –
xenotalk 14:35, 13 April 2011 (UTC)reply
Whether they are supposed to be or not, admins are seen as elites and in many cases act as elites. Admins also tend to be far more protective of admin rights (not breaking them apart, demanding that other people become admins before getting other rights, etc) than non admins. A class system has developed, and it's a turn off for many. Continuing to deny it at this point is counterproductive. Right now, adminship is a big deal, period.
Sven ManguardWha? 20:36, 13 April 2011 (UTC)reply
Per MuZemike. Although I have no problem with bureaucrats not being admins, that doesn't change the fact that viewing deleted pages has absolutely nothing to do with bureaucrat duties. As a bureaucrat, I cannot think of a single time where I have needed the ability to view deleted pages to carry out my bureaucrat duties. --
(ʞɿɐʇ) ɐuɐʞsǝp 14:53, 13 April 2011 (UTC)reply
Per above. Quite redundant. -FASTILY(TALK) 19:07, 13 April 2011 (UTC)reply
per Sven Manguard. --
Ϫ 22:30, 13 April 2011 (UTC)reply
Expect that bureaucrats should also be admins so should not need this. I would not expect an RFB to pass without an RFA also being passable.
Graeme Bartlett (
talk) 12:16, 14 April 2011 (UTC)reply
On the simple grounds that this one is not like the others, and the rationale at the top of this page doesn't apply to it.
Chick Bowen 05:22, 15 April 2011 (UTC)reply
Oppose – Unnecessary for normal 'crat work. —
mc10 (
t/
c) 02:06, 18 April 2011 (UTC)reply
Oppose unlike the other proposals its not clear really why Bureaucrats need this right. --
Eraserhead1 <
talk> 21:39, 19 April 2011 (UTC)reply
The rationales advanced in favour of this proposal are decidedly unconvincing.
Skomorokh 11:03, 23 April 2011 (UTC)reply
There is no reason a crat should need to view deleted edits.
Zginder 2011-04-28T01:56Z (
UTC)
I see absolutely no need for 'crats to have viewdlete privileges taking into account the situations that bureaucrats handle....unnecessary. mauchoeagle(
c) 19:48, 9 May 2011 (UTC)reply
Neutral
I don't have anything against this proposal, but at the moment all 'crats have admin tools as well. And I don't think anybody can become a 'crat without being an admin first.
ArmbrustTalk to meContribs 18:33, 12 April 2011 (UTC)reply
Bit of a catch-22, no? All 'crats current have admin tools because (it is so posited that) they need to view deleted revisions (and
more) to effectively do their job. If the necessary privileges were available in the bureaucrat package, a bureaucrat might decide to relinquish their administrative privileges while still serving as a bureaucrat. Thus, one of the reason all 'crats currently have admin tools is because they can't give them up without becoming less effective - a self-fulfilling prophecy. –
xenotalk 18:40, 12 April 2011 (UTC)reply
Maybe, but I don't see why a 'crat would want to remove their admin user right. I mean there's no policy (at least I don't know about it), which says you can't use some of the rights if you don't use them all. If you want to use only deletedhistory, deletedtext, and browsearchive from the admin package, then just simply don't use the other parts. I don't think, that it's too difficult. If many non-admins would be promoted to 'crats, than these question would be more important.
ArmbrustTalk to meContribs 21:56, 12 April 2011 (UTC)reply
A 'crat would want to remove the admin bit simply because they don't want to do the tasks anymore. If you have the bit, and you're active somewhere, people will still be asking you to do admin stuff.
demize(
t ·
c) 02:23, 13 April 2011 (UTC)reply
I suppose it is a sensible argument that if a crat needed to see deleted material, they could temporarily give themselves the admin bit. --Errant(
chat!) 10:25, 13 April 2011 (UTC)reply
There have been multiple cases in the past where crats have relinquished their admin rights while maintaining their crat bits. (
X! ·
talk) ·
@843 · 19:14, 13 April 2011 (UTC)reply
Question can a 'crat make himself an admin?
Zginder 2011-04-28T03:18Z (
UTC)
They have the technical ability to do so, but there are very few situations where that would be appropriate.
28bytes (
talk) 03:22, 28 April 2011 (UTC)reply
That would mean that a 'crat must have passed a RFA and not have lost the mop in disgrace. Otherwise a 'crat could give itself powers the community does not.
Zginder 2011-04-28T03:27Z (
UTC)
More generally, 'crats have the technical ability to give anyone the sysop bit, but again, they only do so under prescribed circumstances. Just to put the proposal here in context, it has been suggested that some 'crats would like to relinquish their sysop bits to focus on 'crat work full-time. This proposal would make that technically possible, since there's currently no policy reason a 'crat couldn't relinquish the sysop bit but continue to perform 'crat work.
28bytes (
talk) 03:30, 28 April 2011 (UTC)reply
I have no problem with a 'crat giving up the mop in fact
I would like to see it required, but they must have still passed an RFA (and not have been desysoped in disgrace), if they could at any time give the mop to themselves. Also I do not think that 'crats need to see deleted pages to do their job.
Zginder 2011-04-28T04:09Z (
UTC)
Discussion
Was requested in
bugzilla:25752; a developer requested a clear consensus discussion be held prior to committing the change. –
xenotalk 15:13, 12 April 2011 (UTC)reply
I'm not opposing, but Bureaucrats don't really need to view deleted material as it is not in their "job spec". I don't buy the argument that they would often (if at all) need it for RFA closures because they are only judging consensus and not making a decision on the suitability of the candidate themselves. Also; there is not practical likelihood that a non-Admin would become a bureaucrat any time soon --Errant(
chat!) 15:42, 12 April 2011 (UTC)reply
As regards deleted edits and RfA, see my below reply to Jafeluv. Viewing deleted edits may also be necessary when determining whether or not to carry out a rename (i.e. to see if the user has been making promotional edits that were since deleted). While the community may never endorse a non-admin standing at RfB, an existing bureaucrat might wish to resign their administrative privileges. Please try to focus on the technical aspect of the proposal - assume that someone is a bureaucrat but not an administrator: should they be able to view deleted revisions? –
xenotalk 15:55, 12 April 2011 (UTC)reply
While I have nothing against this, I don't quite see how closing an RfA would require looking at deleted contributions. After all, the closing crat is supposed to evaluate the discussion, not the candidate. Not sure I see a clear relevance to user renaming, either.
Jafeluv (
talk) 15:45, 12 April 2011 (UTC)reply
Sometimes opposition is raised in related to a candidate's speedy deletion tagging, and sometimes administrators opposing reference specific deleted edits - viewing the deleted contributions of a candidate may be required to verify statements made by administrators opposing on these ground. –
xenotalk 15:55, 12 April 2011 (UTC)reply
I am inclined to oppose this pending an explanation as to why it is necessary for the role. I don't buy the argument in the nomination for a minute, so unless these rights are essential to the CHU or bot flagging aspects of the role, this would seem an unwelcome conflation of unrelated functions. Could bureaucrats comment on how central these tools are to their work currently? Thanks,
Skomorokh 15:56, 12 April 2011 (UTC)reply
I am a bureaucrat; see my above remarks. I wouldn't say that it is central to the role, but I have used it numerous times when determining whether to carry out a rename (e.g. Has this user been _so_ promotional with their prior edits that their rename should be declined and they should simply be blocked?) and I have viewed the deleted contributions of RfA candidates prior to closing the request (e.g. Does this user
really use A1 erroneously far too often, even though the targeted articles may eventually be deleted under more applicable criteria?). –
xenotalk 15:57, 12 April 2011 (UTC)reply
I think all bureaucrat work can and should be done entirely based on open discussion or requests. I doubt that you'd need to check deleted contribs to see if a user excessively engaged in promotion, you just need to check his user talk page. Making further investigation, considering a block, would be an admin job, not bureaucrat job. And when closing an RFA, bureaucrats should consider the opinions provided, not directly investigate the evidence.
Cenarium (
talk) 16:10, 12 April 2011 (UTC)reply
As someone who actually performs the work in question, I disagree - but respect your opinion nonetheless. –
xenotalk 16:13, 12 April 2011 (UTC)reply
Well that's my opinion on how the work should be done, but of course others may have different opinions.
Cenarium (
talk) 16:17, 12 April 2011 (UTC)reply
I agree that the bureaucrats should be guided by the community's comments in closing RfAs (and not their own investigations), but verifying that a statement based on a deleted edit is an accurate representation of the situation (rather than taking it at face value) is still an important task. –
xenotalk 16:18, 12 April 2011 (UTC)reply
Yes it's an important task, but one that can be taken by the users (with view-deleted rights) participating in the RFA. I feel that all of the bureaucrat's reasoning should be based on the public discussion.
Cenarium (
talk) 16:28, 12 April 2011 (UTC)reply
Fair enough - reasonable minds can differ on this. –
xenotalk 16:31, 12 April 2011 (UTC)reply
Please see the oppose and neutral section of
this RfA. If I had closed that RfA, I would have reviewed the deleted contributions of the candidate while determining the outcome of the discussion, in particular, the red links posted by Moonriddengirl in the neutral section. –
xenotalk 16:07, 12 April 2011 (UTC)reply
I think bureaucrats should base their decision on only what is public, the opinions provided by the users, so that all users can follow their reasoning and interpretation of consensus. Investigating deleted evidence on one's own goes beyond the role of bureaucrat, they should refer to the comments made on it.
Cenarium (
talk) 16:14, 12 April 2011 (UTC)reply
Should they take any comment based on a deleted edit at face value, or ensure that the comment is an accurate representation of the situation by viewing the deleted edits themselves? –
xenotalk 16:21, 12 April 2011 (UTC)reply
Replied above to avoid duplicating discussion.
Cenarium (
talk) 16:28, 12 April 2011 (UTC)reply
No one has mentioned it but part of our copyright policy is that although copyright violations are not truly deleted, only a certain small and approved group can see them. We need to carefully consider expanding this ability to another group that may not truly need the ability lest granting access to this material through several processes and to diverse groups dilutes our defense. Adding access to oversighters seems obvious, though.
Rmhermen (
talk) 16:55, 12 April 2011 (UTC)reply
Given that the number of non-admin bureaucrats is unlikely to ever exceed the total number of current bureaucrats, the number of current bureaucrats is a tiny fraction of the current administrator corps, and under the current practice those (currently hypothetical) non-admin bureaucrats would be removed from the admin corps, I see it as close to a zero-net-sum item. Right now, every bureaucrat can see deleted revs. If Xeno gets his wish and is allowed to keep being a bureaucrat without the admin bit, that's a net change of... zero.
Jclemens (
talk) 01:01, 13 April 2011 (UTC)reply
The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
Add move-related privileges to Bureaucrat
Clear consensus in favour of implementationHappy‑
melon 10:31, 15 May 2011 (UTC)reply
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
In order to remove the technical limitation that
bureaucrats must also be administrators to properly carry out renames, the move-subpages, suppressredirect, and tboverride rights should be added to the bureaucrat permission group.
Support
Proposed. It is important that a Bureaucrat have the ability to move subpages, to move pages without creating a redirect, and to override the title blacklist, especially when carrying out renames. –
xenotalk 15:13, 12 April 2011 (UTC)reply
Required to make the role "self-sufficient" so sensible to add, even though from a practical perspective it won't be needed. --Errant(
chat!) 15:51, 12 April 2011 (UTC)reply
Necessary for the job.
Cenarium (
talk) 15:52, 12 April 2011 (UTC)reply
Makes sense. Otherwise bureaucrats would either need to be admins themselves or ask for an admin to assist when performing renames.
Jafeluv (
talk) 15:53, 12 April 2011 (UTC)reply
Bureaucrats should have the rights they need to do their jobs; right should be self-contained.
ErikHaugen (
talk |
contribs) 17:19, 12 April 2011 (UTC)reply
Unlike view-deleted privileges, this does affect the core functionality of bureaucrats, and so I support. --
Avi (
talk) 20:04, 12 April 2011 (UTC)reply
Personally, I would never support someone to be a crat without being a sysop first, but at least moving pages is relevant to what a crat does (renames). /
ƒETCHCOMMS/ 21:44, 12 April 2011 (UTC)reply
Seems like it might be unnecessary, but also little harm in it.
Beyond My Ken (
talk) 22:15, 12 April 2011 (UTC)reply
I can think of an editor who would more easily pass RFB than RFA. Might as well make this possible. ϢereSpielChequers 22:17, 12 April 2011 (UTC)reply
Straightforward. Bureaucrats need these privileges in order to rename users.
Ucucha 23:26, 12 April 2011 (UTC)reply
If bureaucrats have the ability to rename users, then it makes sense for them to be able to move those pages, as well. The reason this is not the case is because, in the MediaWiki defaults, sysops normally hold right to rename users. Nobody probably bothered to look at the fact that this right was not given to bureaucrats because it's our community norm that bureaucrats be sysops first when, in actuality, they need not be. –
MuZemike 00:38, 13 April 2011 (UTC)reply
Trust the 'crats on this one. Again, I believe that each toolkit/user right should have the tools required to operate independently from any other permissions.
Risker (
talk) 02:34, 13 April 2011 (UTC)reply
Unlike the privileges related to viewing deleted material above, this set of move-related privileges directly affects the ability of bureaucrats to carry out their core duties following an open consensus discussion.--
Danaman5 (
talk) 02:41, 13 April 2011 (UTC)reply
Definitely - the 'crat ability should be self-sufficient, and all 'crat jobs should be doable in full without any other user rights. This is some times necessary for user renames.
עוד מישהוOd Mishehu 07:03, 13 April 2011 (UTC)reply
Can't see the slightest bit of harm in it, and even if it's not needed today, it may be needed tomorrow. Much better to have something and not need it than to need it and not have it.
Pesky (
talk) 08:04, 13 April 2011 (UTC)reply
I fine with this technical change being done: while it would be rare, one of the bureaucrats may not want to remain as an administrator, or merely want to take a break from being one, but are still happy doing the bureaucrat tasks.
Acalamari 11:39, 13 April 2011 (UTC)reply
Xeno hits the nail on the head. I've thought about this one problem before, actually. --
(ʞɿɐʇ) ɐuɐʞsǝp 14:54, 13 April 2011 (UTC)reply
Could be significant in the context of a rename. The tool we're discussing is easily reversed (like all admin tools) if misused. If we trust someone to promote admins, then by implication we probably trust them to at least handle the related page moves properly. Candidates seen as having a problem in this area would quickly find themselves opposed at RFB.
FT2(
Talk |
email) 10:30, 15 April 2011 (UTC)reply
Agree it is useless. Another solution in search of a problem.
Ruslik_
Zero 19:08, 12 April 2011 (UTC)reply
See my replies to Juliancolton below. –
xenotalk 19:36, 12 April 2011 (UTC)reply
It should be that bureaucrats must also be administrators, thus nullifying the premise. --
SmokeyJoe (
talk) 09:57, 13 April 2011 (UTC)reply
Although people with advanced permissions tend to deny it quite vocally, there is a class system in Wikipedia. Although people tend to accept it but can't agree on how to fix it, RfA is a horrid system at this point. I think Pesky inadvertently responded to you in the support section too, by the way.
Sven ManguardWha? 20:33, 13 April 2011 (UTC)reply
Discussion
Was requested in
bugzilla:25752; a developer requested a clear consensus discussion be held prior to committing the change. –
xenotalk 15:13, 12 April 2011 (UTC)reply
Leaning towards supporting this, but pausing per what I said below.
Ncmvocalist (
talk) 16:23, 12 April 2011 (UTC)reply
Renameuser extension apparently allows bureaucrats to move all user subpages of the renamed user. So, move-subpages userright is not necessary.
Ruslik_
Zero 08:41, 19 April 2011 (UTC)reply
The userright will still be useful to cleanup in case a particular rename hangs. –
xenotalk 14:10, 19 April 2011 (UTC)reply
In this case all restrictions discussed below will apply.
Ruslik_
Zero 07:35, 20 April 2011 (UTC)reply
The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
General discussion
I think these are good proposals and am very much in favor especially for the first two, but on the other hand, is there any reason for not having these "higher" userrights restricted to admins? I don't mean that we should prevent non-admins from applying for checkuser rights, for example, but if an editor can be trusted with checkuser tools, can they not also be trusted with the "regular" mop?
Zakhalesh (
talk) 15:24, 12 April 2011 (UTC)reply
How is adminship not a social requirement for bureaucratship? I admit, I don't get around to RFA much, but in my experience it is very much a social requirement. Unless you meant to say that adminship is not a technical requirement for bureaucratship, in which case you would be correct.--
Danaman5 (
talk) 15:34, 12 April 2011 (UTC)reply
There is no requirement that an editor must be an administrator prior to their standing for bureaucratship, and nothing preventing a bureaucrat from resigning their administrative privileges while retaining their bureaucrat rights. Please do keep in mind that best practices suggest that user groups should be self-sufficient and self-contained, and making the technical changes to the bureaucrat privilege would not prevent the community from opposing candidates at RfB based on their not having successfully completed RfA. –
xenotalk 15:38, 12 April 2011 (UTC)reply
I understand your point, but when you say it is not a "social requirement", it implies to me that editors who apply for bureaucratship without first being an administrator are not opposed based on that fact (i.e. their doing so is accepted by the community), but that does not seem to be the case to me. Perhaps just a case of unclear wording.--
Danaman5 (
talk) 15:41, 12 April 2011 (UTC)reply
That the community has not yet bestowed bureaucratship on an editor who has never been an administrator does not mean that it will not one day occur, and there is no policy-based (i.e. social) requirement that one be an administrator prior to standing for bureaucrat. Furthermore, more than one bureaucrat has indicated a desire to step down from adminship while continuing to serve in their capacity as a bureaucrat. This proposal strictly seeks to make higher-level userrights self-sufficient and self-contained, and does not seek to change the community's attitude towards a non-administrator standing for bureaucratship. –
xenotalk 15:46, 12 April 2011 (UTC)reply
This provides a separable ability for there to exist a 'cat-without-sysop, which seems like a quite useful combination. Even if the community decided that was a bad combination, such a prohibition could be made on a policy basis, so I see no particular harm done by implementing this even in that case. And the community may very well wish to consider making some crats-without-sysop going forward in some cases, e.g., if renameuser backlogs got to be a problem. --
joe deckertalk to me 15:56, 12 April 2011 (UTC)reply
The community may very well wish to do a lot of things; but there's a time and a place - if such a situation arises which would justify the need for this, then I'd prefer to implement it after knowing what the circumstances are and/or what the wiki climate is like at that time.
Ncmvocalist (
talk) 16:13, 12 April 2011 (UTC)reply
Mmmm, it's true that I phrased that in a very "someday" sort of way, and that that part of my argument is premature. However, it sounds to me as if Xeno has considered and may still be considering asking to drop his sysop bit but not his 'crat bit. I don't think it's premature to ask what that would look like. Moreover, Xeno is right as near as I'm aware about one thing, we have no policy that says you can't have that combination of bits. As I'm more of a "that which is not prohibited is allowed" rather than a "that which is not specifically allowed is prohibited" kinda guy, I think it does seem worth asking "what would 'crat-sans-sysop" look like now as a result, whether that discussion happens here or elsewhere. *shrug* --
joe deckertalk to me 16:20, 12 April 2011 (UTC)reply
What about the fact that if view-deleted rights were added to bureaucrat rights, this would make it a bigger deal for non-admins to pass RFB ? Yes it doesn't seem needed to me.
Cenarium (
talk) 16:31, 12 April 2011 (UTC)reply
I think that's a red herring. Each role should have all the required privileges needed to do any of the tasks required by that role, without relying on any other (presumably subordinate or prerequisite) role. If roles are considered "social" prerequisites, then the addition of these privileges will create no new risks, because each of these proposals overlap with what has already been required.
Jclemens (
talk) 16:52, 12 April 2011 (UTC)reply
I don't consider that viewing deleted pages is needed for the bureaucrat role, but never mind.
Cenarium (
talk) 17:01, 12 April 2011 (UTC)reply
Unless I'm dreadfully misreading the proposal, I don't see that there's any real problem to go along with these solutions. People fail RfB for being an admin for only 10 months, yet we're going to set up non-admins to run?
Juliancolton (
talk) 18:58, 12 April 2011 (UTC)reply
The problem is that the community
seems to largely agree that adminship is not a social requirement for certain higher privileges (nor should it be), but it is currently a technical requirement because some rights necessary for the tasks are inherited from the administrative package. Try to focus on the technical aspects of the proposal and set aside your worries that we might have proposed a solution without a problem - even if this were the case, there is no inherent harm that would be caused by the technical change. –
xenotalk 19:07, 12 April 2011 (UTC)reply
But it's more than just a technical issue. Userrights shouldn't be self-sufficient; they should compliment each other depending on what a given user is interested in doing on the project. It's true that there's not a huge risk in implementing these proposed changes, but I guess I just feel it's a lot of confusion for an as-of-yet unverified hypothetical situation.
Juliancolton (
talk) 19:25, 12 April 2011 (UTC)reply
I don't make these proposals because I like arguing with users who think the solutions I am proposing have no problems to go along with them. In 2010, I actually intended to relinquish my administrative privileges after
bugzilla:25752 was fulfilled, and I know of at least one other bureaucrat who has considered doing so as well. –
xenotalk 19:31, 12 April 2011 (UTC)reply
Are you hesitantly apologizing, or indicating confusion at my response? –
xenotalk 19:38, 12 April 2011 (UTC)reply
"Userrights shouldn't be self-sufficient; they should compliment each other depending on what a given user is interested in doing on the project" <-- Does this mean you think that the privileges found in 'rollback', 'reviewer', 'file mover', 'autopatrolled', & 'account creator' should be removed from the administrative rights package and sought individually by administrators? If not, is there a reason that userrights redundancy is desirable in the administrative package but not other higher privileges? –
xenotalk 19:51, 12 April 2011 (UTC)reply
To be honest, I really don't care either way with most of these. I'm not (and never have been), a big "bugzilla" reader anyway. It occurs to me that either you "trust" someone, or you don't. I know the whole Arb/Crat/Admin/Editor thing isn't supposed to be hierarchical in structure, but in practice it often works out that way. Although, I can see where "Crat" is more of a horizontal position to "Admin" in function. Just don't recall ever seeing some of these rights being assigned to users who aren't members of a particular group. — Ched :
? 22:58, 12 April 2011 (UTC)reply
Comment Xeno, it sounds like you really want to go the whole hog and allow turning each permission bit on and off separately. There is some logic to that at lower user levels (like rollback or reviewer), which are mostly about competence rather than trust. At the upper levels, you're proposing to separate out what was a few roles activated by well-established processes, into a bit vector with a configuration space of exponential size in the number of bits, with individual bits being potentially given out on an ad hoc basis. Switching from roles to bits is potentially pretty drastic and I think should take considerable mulling over before implementation. I'd be ok sticking with roles (adding new ones like "auditor" if necessary), and allowing role-holders (who got the roles through normal processes) voluntarily relinquish bits that they received as part of their roles. (But if want bureaucrat-like duties without sysop bits, why not run for Steward?).
If I understand SlimVirgin and Cenarium's concerns, it's something like a
code is law worry that anything that any capability enabled by the software will be used and then entrenched, if it delivers the past of least resistance to some ephemeral goal once or twice. Looking at it this way, and with the increased complexity of the current RFC, I'm uncomfortable with having people acquire the bits through less formal processes than what we have now, therefore have reservations about putting the technical framework in place for it. I suppose this is a reversal from what I was saying earlier.
69.111.194.167 (
talk) (new address) 00:54, 13 April 2011 (UTC)reply
No... that is not really what I am asking for here, though I think that is what MZMcBride is
suggesting on the talk page. I don't really follow your "bit vector" example, but I don't think that is what I am asking for either. I am asking for a few user groups to receive permissions they require for their role - permissions which are usually inherited by virtue of most higher permission holders also being administrators. While this has worked until now (since it is rare for a non-administrator to seek higher permissions, and rarer still for them to be granted), there is no harm in remedying a situation that has largely come about by organic growth (the higher permissions were apparently put together quickly on an urgent need basis). If I relinquished my administrator rights, I could still rename users, but I wouldn't be able to move their subpages during the rename. That is just a technical glitch that needs to be overcome, and that is what this proposal seeks to remedy. (As for your bracketed suggestion - Stewards are not permitted to act on their local wiki, and indeed are not to carry out renames with their steward hat on projects with bureaucrats.)
At the risk of sounding repetitive: Not everyone wants to be an administrator, and not everyone wants to be an administrator forever. Further, 'administrator' is not some kind of "super user" or prerequisite for advanced privileges, nor does the RFA process or the flipping of the administrator bit suddenly make a user any more or less trustworthy than they were prior to them receiving administrative privileges. –
xenotalk 14:06, 13 April 2011 (UTC)reply
The abusefilter right seems to me to be a very obvious example of why the present system is absurd. Almost all of the current admins were "promoted" before those filters existed, and hardly any understand how to write regular expressions. The present system of packaging user rights is simply lazy, and a historical artefact.
MalleusFatuorum 02:24, 14 April 2011 (UTC)reply
Xeno, I guess I wonder what you think "administrator rights" means, in terms of relinquishing them. Do you mean just the technical capabilities of +sysop that aren't in +bureaucrat? For example, say you're not on arbcom, you're still a bureaucrat, and you've relinquished "administrator rights". There is an AE request about user XYZ's editing in some articles under discretionary sanctions. Can you close the request by placing user XYZ under a topic ban, notifying him on his talk page, and updating the appropriate logs? That doesn't use any sysop buttons, but it's an action reserved for "uninvolved admins". So I'm wondering if you'd consider yourself still authorized to do that. Me, I think adminship is no longer about "buttons" and RFA has a higher bar than before, because admins have greater authority than before, in areas like AE/ANI/etc. We never hear about "rouge admins" any more, since the out-of-process "rouge" stuff they used to do to keep the site running, has since morphed into recognized and accepted processes. But that makes adminship a much "bigger deal" than before. As I see it, the big deal is authority, not buttons.
69.111.194.167 (
talk) 03:45, 14 April 2011 (UTC)reply
I think that is somewhat off-topic (a philosophical discussion about what it means to be an administrator probably belongs elsewhere) - nevertheless - no, if I turned in my administrative privileges, I wouldn't close a request at AE because it calls for uninvolved administrators to close requests. –
xenotalk 12:20, 14 April 2011 (UTC)reply
Comment. This proposal, even if adopted, will not make bureaucrats self-sufficient. They still will not be able to move user css and js subpages without 'editusercss' and 'edituserjs' userrights. They are also will not be able to move move-protected userpages without 'protect' userright.
Ruslik_
Zero 19:01, 14 April 2011 (UTC)reply
Hadn't thought about that - thank you for pointing that out. This is probably something that could be dealt with "under the hood", so to speak - rather than adding those rights to the proposal. –
xenotalk 19:37, 14 April 2011 (UTC)reply
It would be a good idea to include them in this, but now it is a hassle to add after so many have voted. Ruslik0, are there any more rights that would be used? What about the user editnotices?
Graeme Bartlett (
talk) 21:45, 14 April 2011 (UTC)reply
I wonder if a separate RfC would be appropriate that posed the question more directly: "Should admins who become bureaucrats be allowed to later relinquish their adminship but retain the technical abilities needed to fully function as a bureaucrat?" I think the current proposal assumes the answer is yes, and while it might be the case that the community agrees with this (I would support it personally), I think the lack of context and the tradition that bureaucrats have always been admins first is leading a lot of people to assume this proposal has no practical point given that all 'crats are (currently) admins.
28bytes (
talk) 00:04, 15 April 2011 (UTC)reply
How could they miss that context with me clamouring about it all over the page? ;> I think the simplest solution to the problem highlighted by Ruslik would be to ask the developers to make the RenameUser interface move all the target user's pages regardless of protection status or whether they end in .js or .css. –
xenotalk 13:03, 15 April 2011 (UTC)reply
Of course, the RenameUser interface can be made to move js/css files as well as to move subpages, not to create redirects etc. However, this also would make the proposal under discussion meaningless. The truth is that without those userrights that I mentioned above your dream of being a purified bureaucrat will remain just a dream. It seems impossible to be an effective bureaucrat without being also a sysop.
Ruslik_
Zero 19:29, 15 April 2011 (UTC)reply
How so? The discussion seems to confirm that users do not have an issue with bureaucrats continuing to rename users while not being sysops. As such, the software should be fixed to allow this. (In any case, the majority of users who seek rename do not have custom js or css and do not have moveprotected pages, based on my observations since you've mentioned this.) –
xenotalk 19:34, 15 April 2011 (UTC)reply
Every time you are renaming an account you will need to check whether there exist js/css subpages (this is simple) and whether some subpages are move-protected (more difficult). Otherwise you may end up scattering various subpages over two accounts. The system likely will not display any warnings. And if you find them you will need to ask somebody else to do the job. So, the life of the hypothetical purified bureaucrat will not be easy.
Ruslik_
Zero 19:49, 15 April 2011 (UTC)reply
Yes, which is why the RenameUser extension should be fixed to not be interdependent on other userrights. –
xenotalk 20:02, 15 April 2011 (UTC)reply
The likely result is WONTFIX.
Ruslik_
Zero 13:12, 16 April 2011 (UTC)reply
For what reason? –
xenotalk 13:16, 16 April 2011 (UTC)reply
Developers have a lot of more pressing issues to deal with. This is simply a non-issue from the technical point of view.
Ruslik_
Zero 17:06, 16 April 2011 (UTC)reply
I don't follow your logic; but as Happy-melon has helpfully explained, this section of the discussion is moot because the RenameUser extension already largely operates how it ought (i.e. independently of lower-order userrights). –
xenotalk 19:06, 18 April 2011 (UTC)reply
The 'editusercss', 'edituserjs' and 'protect' permissions are not tested for in the RenameUser extension, only that the user has the 'move' permission. To use the terminology of this thread, it has already been 'fixed'. If it hadn't, it would have been a one-line change to one boolean parameter to make it so, and would have been something I'd very happily have done. Developers are, in general, volunteers and work on whatever interests them. Members of other projects should generally not try to either second-guess, or dictate, what those interests are or should be.
Happy‑
melon 10:02, 17 April 2011 (UTC)reply
I appear to be wrong. However, I do not understand why it checks 'suppressredirect' but not other three permissions? Is there any logic here?
Ruslik_
Zero 16:02, 17 April 2011 (UTC)reply
I guess because it's a whole feature in itself, not a partial feature; if you have 'suppressredirect' it shows the checkbox and lets you do it; if you don't it doesn't even let you try. Similarly the option to move pages is only available if you have the 'move' permission. Incidentally, the 'move-subpages' permission is not checked either, so I don't really know why that's included in this discussion (presumably innocent ignorance in the same way).
Happy‑
melon 17:32, 17 April 2011 (UTC)reply
I does not appear to check $wgMaximumMovedPages as well. So, any number of subpages can be moved. It also does not check if the user is blocked. So, blocked bureaucrats can still rename users.
Ruslik_
Zero 06:32, 18 April 2011 (UTC)reply
Hmm... not limiting the number of pages moved is correct, I would say. Not checking blocked status is definitely not; fixed that in
r86369.
Happy‑
melon 23:47, 18 April 2011 (UTC)reply
Ok, I noticed that in title.php isValidMoveOperation checks whether the target title is protected. This check is not influenced by $auth.
Ruslik_
Zero 07:27, 19 April 2011 (UTC)reply
Thank you for explaining that, Happy-melon. Yes, I (and Ruslik, it seems) presumed that because 'suppressredirect' was checked, the other rights would be checked as well. –
xenotalk 19:01, 18 April 2011 (UTC)reply
Wouldn't it be better/easier to grant the
researcher right to all crats/CUs/OS non-admins instead of adding viewdeleted privileges to these user groups, in order to prevent software's internal redundancy (in view of the fact that a lot of administrators are also bureaucrats and/or checkusers)?
Ruy Pugliesi◥ 04:11, 1 May 2011 (UTC)reply
Short answer, Researcher is Foundation-administered, not Arbcom-administered.
Jclemens (
talk) 04:30, 1 May 2011 (UTC)reply
Ok. We could implement a similar "ArbCom -administered" permission or authorize "researcher flag" being administered in this way, instead of adding more privileges to crats and checkusers. For example: on Portuguese Wikipedia, there is the
eliminator, which allows non-admins to view deleted contributions and also delete/restore pages.
Ruy Pugliesi◥ 20:37, 1 May 2011 (UTC)reply
From the better-late-than-never department: researcher does not allow one to view deleted revisions (
Special:UserGroupRights). –
xenotalk 15:32, 7 September 2011 (UTC)reply