The Workshop phase for this case has closed. Any further edits to this page may be reverted by an arbitrator or arbitration clerk without further reason. If you need to make further edits to the page please make a request on the talk page.

Wikipedia Arbitration
About arbitration Talk Arbitration policy (CU/OS) Guide to arbitration Noticeboard + Talk Search archived proceedings Ban appeals Report
Open proceedings All open requests Case requests Clarification and Amendment requests Arbitrator motion requests Enforcement requests
Active sanctions Contentious topics (Log) General sanctions Editor sanctions
Arbitration Committee Discuss + Contact Members Procedures Elections Talk History Clerks + Talk Clerk procedures
Audit Talk Statistics Talk Reports (pre-2016)
Track related changes
t e

Purpose of the workshop

Arbitration case pages exist to assist the Arbitration Committee in arriving at fair, well-informed decisions. The case Workshop exists so that parties to the case, other interested members of the community, and members of the Arbitration Committee can post possible components of the final decision for review and comment by others. Components proposed here may be general principles of site policy and procedure, findings of fact about the dispute, remedies to resolve the dispute, and arrangements for remedy enforcement. These are the four types of proposals that can be included in committee final decisions. There are also sections for analysis of

page. Only Arbitrators and clerks may edit that page, for voting, clarification as well as implementation purposes.

Expected standards of behavior

• You are required to act with appropriate decorum during this case. While grievances must often be aired during a case, you are expected to air them without being
  personal attacks
  , and to respond calmly to allegations against you.
• Accusations of misbehaviour posted in this case must be proven with clear evidence (and otherwise not made at all).

Consequences of inappropriate behavior

• Editors who conduct themselves inappropriately during a case may be sanctioned by an arbitrator or clerk, without warning.
• Sanctions issued by arbitrators or clerks may include being banned from particular case pages or from further participation in the case.
• Editors who ignore sanctions issued by arbitrators or clerks may be blocked from editing.
• Behavior during a case may also be considered by the committee in arriving at a final decision.

Motions and requests by the parties

Template

1)

Comment by Arbitrators:

Comment by parties:

Comment by others:

Template

2)

Comment by Arbitrators:

Comment by parties:

Comment by others:

Proposed temporary injunctions

Template

1)

Comment by Arbitrators:

Comment by parties:

Comment by others:

Template

2)

Comment by Arbitrators:

Comment by parties:

Comment by others:

Questions to the parties

Arbitrators may ask questions of the parties in this section.

Proposed final decision

Proposals by User:Xaosflux

Proposed remedies

Note: All remedies that refer to a period of time, for example to a ban of X months or a revert parole of Y months, are to run concurrently unless otherwise stated.

Administrators reminded (blocks)

1) Administrators are reminded that blocks may be made based on information they are not privy to, and as such these blocks may not be discretionarily lifted. The localized message for the standard unblock form (MediaWiki:Unblockiptext) may be used to reinforce this reminder.

— xaosflux ^Talk 04:06, 31 December 2019 (UTC)[reply]

Comment by Arbitrators:

Comment by parties:

Comment by others:

It is worth stressing that this does not apply only to CU or OS blocks, but to any and all admin actions - for example OTRS/UTRS tickets or IRC. This is why discussing matters with the admin who took the initial action is encouraged. It might be worth having a principle along the lines of "Administrators are expected to ensure they are familiar with all the circumstances surrounding an admin action before modifying or reversing it." Thryduulf (talk) 00:01, 3 January 2020 (UTC)[reply]

@Thryduulf: I would support this in general. However, it's possible that, due to simply the breadth of an issue with a user, the admin may not be familiar with all the circumstances (which is unreasonable). Perhaps "Administrators are expected to make a good-faith effort to ensure they have sufficient familiarity with the circumstances surrounding an admin action before modifying or reversing it." Such phrasing would allow for honest mistakes while condemning hasty actions. Buffs (talk) 19:16, 3 January 2020 (UTC)[reply]

@Buffs: yes, I agree with that. Thryduulf (talk) 00:11, 4 January 2020 (UTC)[reply]

Fine, but doesn't go far enough. There's a specific instance here — reversing a CU block — and ArbCom has been very clear for near a decade. Indeed, nobody reading Template:Checkuserblock-account could be mistaken. Going too broad waters the impact down — maybe sysops need to be reminded of this, maybe just one does — but being milquetoast about makes things less clear, not more. ~ Amory (u • t • c) 02:33, 7 January 2020 (UTC)[reply]

@Amorymeltzer: we've recently updated the the warning at MediaWiki:Unblockiptext, does that strengthen it enough to meet your concern? — xaosflux ^Talk 20:44, 7 January 2020 (UTC)[reply]

Might as well just use that text, though, right? Administrators are reminded that they may not undo any blocks that are labelled as a CheckUser, Oversight, or ArbCom block without consulting a(n) CheckUser/Oversighter or the Arbitration Committee would still be better. I take Thryduulf's point, but I just mean that when I read this proposed remedy I don't think "be careful around a CU block," I think "only do unblocks if I know I'm right." If we're going down this path, being explicit is good. ~ Amory (u • t • c) 21:12, 7 January 2020 (UTC)[reply]

I'm going to put on my Whiner about crappy U/I hat and complain about the unblock form. Yes, it has the MediaWiki:Unblockiptext warning text. That basically says, Sometimes this is dangerous. But, it doesn't tell you that this is one of those times. Special:Unblock should include the block log. I just counted; it took me three clicks to get from Special:Unblock/RoySmith-testing to where I could see the block log. If you want people to do the right thing, make it easy to do the right thing. -- RoySmith (talk) 15:36, 14 January 2020 (UTC)[reply]

I think you can do it in one if you click on "Block user" or "User contributions" in the sidebar, but regardless, this is indeed awful, and doesn't match Special:Block/RoySmith-testing. ~ Amory (u • t • c) 16:49, 15 January 2020 (UTC)[reply]

phab:T242888 seeks to add the block log to the unblock form. — xaosflux ^Talk 17:07, 17 January 2020 (UTC)[reply]

Proposed Principles

Purpose of Wikipedia

The purpose of Wikipedia is to create a high-quality, free-content encyclopedia in an atmosphere of camaraderie and mutual respect among contributors. Contributors whose actions are detrimental to that goal may be asked to refrain from them, even when these actions are undertaken in good faith; and good faith actions, where disruptive, may still be sanctioned.

Comment by Arbitrators:

Comment by parties:

Comment by others:

I think this is supposed to be a proposed principle? Buffs (talk) 19:27, 2 January 2020 (UTC) [reply]

It's those equal signs. Robert McClenon (talk) 21:53, 2 January 2020 (UTC)[reply]

No prob :-) Buffs (talk) 19:07, 3 January 2020 (UTC)[reply]

Administrator conduct (proposed by Robert McClenon)

Administrators are expected to observe a high standard of conduct and retain the trust of the community at all times. Administrators are expected to respond promptly and civilly to queries about their Wikipedia-related conduct and administrator actions and to justify them when needed. Sustained or serious disruption of Wikipedia is incompatible with the expectations and responsibilities of administrators, and consistent or egregious poor judgment may result in the removal of administrator tools.

Comment by Arbitrators:

Comment by parties:

Comment by others:

I quite agree. DES ^(talk)_{DESiegel Contribs} 18:27, 10 January 2020 (UTC)[reply]

Administrator involvement

With few exceptions, editors are expected to not act as administrators in cases where, to a neutral observer, they could reasonably appear involved. Involvement is generally construed very broadly by the community, to include current or past conflicts with an editor (or editors), and disputes on topics, regardless of the nature, age, or outcome of the dispute. While there will always be borderline cases, whenever in doubt, an administrator should draw the situation to the attention of fellow sysops, such as by posting on an appropriate noticeboard, so that other sysops can provide help.

Comment by Arbitrators:

I don't see anything in evidence raising INVOLVED concerns. –xeno^talk 01:15, 17 January 2020 (UTC)[reply]

Comment by parties:

Comment by others:

This is true enough, but I also fail to see where this is relevant to the current case. DES ^(talk)_{DESiegel Contribs} 14:13, 22 January 2020 (UTC)[reply]

Administrator accountability

Administrators are accountable for their actions involving administrator tools, as unexplained administrator actions can demoralize other editors who lack such tools. Administrators who seriously or repeatedly act in a problematic manner, or who have lost the trust or confidence of the community, may be sanctioned or have their administrator rights removed by the arbitration committee. Administrators should be reasonably aware of community standards and expectations when using administrative tools.

Comment by Arbitrators:

Comment by parties:

Comment by others:

Purpose of blocks

5) Blocks serve to protect the project from harm, and reduce likely future problems. They are meted out not as retribution but to protect the project and other users from disruption and inappropriate conduct, and to deter any future possible repetitions of inappropriate conduct. Blocking is one of the most powerful tools that are entrusted to administrators, who should be familiar with the circumstances prior to intervening and are required to be able to justify any block that they issue. In general, once a matter has become "cold" and the risk of present disruption has clearly ended, reopening it by blocking retrospectively is usually not appropriate.

Comment by Arbitrators:

Comment by parties:

Comment by others:

This is straight from policy, but i don't see where it applies to the current case. The CU block that triggered the case has not been alleged to be punitive rather than protective, and i didn't notice complaints of or evidence about improper blocks by RHaworth. DES ^(talk)_{DESiegel Contribs} 14:16, 22 January 2020 (UTC)[reply]

Communication

Due to the collaborative nature of Wikipedia, proper communication is extremely important. All editors are expected to respond to messages intended for them in a timely manner and to constructively discuss controversial issues. This is especially true for administrators in regard to administrative actions. Such expected communication includes: giving appropriate (as guided by Wikipedia's policies and guidelines) warnings prior to, and notification messages following, their actions; using accurate and descriptive edit and administrative action summaries; and responding promptly and fully to all good-faith concerns raised about their administrative actions. Because

, it is required that all communications be civil.

Comment by Arbitrators:

Comment by parties:

Comment by others:

@

WP:IUC. The words "timely" and "constructive" aren't enough in this case (and others), IMHO. Buffs (talk) 19:10, 3 January 2020 (UTC)[reply

]

In ordinary editing matters, a "timely" response is not necessarily expect (and is often not received), but failure to make such a response may result in the editor being deemed to have conceded whatever point was at issue. Similarly, a message asking an admin to attend to some issue on which that admin has not previously acted do not carry an expectation of timely response. Responses to requests for explanatiosn of admin actions already undertaken are different, and should be both timely and civil. DES ^(talk)_{DESiegel Contribs} 18:31, 10 January 2020 (UTC)[reply]

"All editors are expected to respond to messages intended for them in a timely manner" appears to conflict with

WP:NOTCOMPULSORY, with emphasis on "all editors" and "respond". The requirement for most users only seems to be reading such messages, and not to continue doing anything they have been warned about. Silence and an end of disruptive edits is a perfectly fine reaction to a warning, when done by editors without access to advanced tools. ~ ToBeFree (talk) 14:30, 12 January 2020 (UTC)[reply

]

TBF has a point.

WP:ADMINACCT states that "Administrators are expected to respond promptly and civilly to queries about their Wikipedia-related conduct and administrative actions and to justify them when needed." There is no such requirement for standard editors. I would support rephrasing to include that verbiage from ADMINACCT. Buffs (talk) 18:09, 13 January 2020 (UTC)[reply

]

Civility

fourth pillar of Wikipedia

. It is not optional or merely desirable; it is required of all editors at all times. Editors may be warned for breaches of civility, and may be sanctioned for repeated breaches of civility.

Comment by Arbitrators:

Comment by parties:

Comment by others:

It "shouldn't be necessary" to restate this principle, but sometimes it is necessary to restate some things over and over again. Robert McClenon (talk) 22:27, 3 January 2020 (UTC)[reply]

A thousand times, YES! Buffs (talk) 22:36, 3 January 2020 (UTC)[reply]

The ArbCom should almost always include an appropriate

incivility by the administrator is an important issue. It also touches on the need for administrators to respond civilly to inexperienced or combative editors who may be uncivil, either because they don't know any better, or because they do know better. Robert McClenon (talk) 21:39, 6 January 2020 (UTC)[reply

]

Deletion

In order to maintain the highest quality of content in the encyclopedia, it is necessary to have processes for the deletion of pages from the encyclopedia. Decisions about the deletion of pages must be made deliberately and with collaboration, and with mutual respect for other editors who may have different philosophies about the maintenance and deletion of pages. Speedy deletion, which bypasses discussion, is only intended for unambiguous cases. If there is uncertainty as to whether a page should be speedily deleted, it should not be speedily deleted, but its deletion may be reviewed using a process with community involvement. Administrators are responsible for the decision as to whether to approve a request for speedy deletion, and are required to err on the side of caution, deleting a page speedily only if there is no doubt, and declining the speedy deletion request if there is even slight doubt. Because any editor, experienced or inexperienced, may tag a page for speedy deletion, administrators are required to exercise judgment and responsibility before deleting a page.

Comment by Arbitrators:

Comment by parties:

Comment by others:

Yes, this is important. Thryduulf (talk) 00:12, 4 January 2020 (UTC)[reply]

A little wordy, but something to this effect, for sure. I think it'd be good to also note somewhere that there is a lot of deletion, speedy or otherwise, necessary for the hourly maintenance of this project. As an example, the most recent 500 page deletion log entries stretches back just 7 hours. Whatever the outcome here, I think it worth noting that deletion is incredibly important. ~ Amory (u • t • c) 02:39, 7 January 2020 (UTC)[reply]

Indeed, but it's worth noting that the number of speedy deletions requiring admin judgement is relatively small. I could only find two A7s in the last 500 deletions, for example. Espresso Addict (talk) 03:33, 7 January 2020 (UTC)[reply]

The issues here certainly involve improper deletions when sysop judgment shouldn't have been particularly required, so (pending evidence) I don't know how much we would want to distinguish between "easy" (G13) and "hard" (A7) CSDs. I only meant to note that a handful of sysops (strained metaphor alert) walk through a veritable flood of CSDs every day, and this case is about the method in which one fords that river, not that wading in is problematic. ~ Amory (u • t • c) 21:18, 7 January 2020 (UTC)[reply]

Indeed, this needs stating. Espresso Addict (talk) 03:33, 7 January 2020 (UTC)[reply]

I certainly agree with the above principle. Perhaps wording from this could be combined with wording from my proposal on the same topic below. DES ^(talk)_{DESiegel Contribs} 18:34, 10 January 2020 (UTC)[reply]

Yes, it's surprising that it needs saying, but it does. And that's well explained. Boing! said Zebedee (talk) 18:48, 10 January 2020 (UTC)[reply]

Proposed remedies

Note: All remedies that refer to a period of time, for example to a ban of X months or a revert parole of Y months, are to run concurrently unless otherwise stated.

RHaworth is desysopped

1) RHaworth (talk · contribs · blocks · protections · deletions · page moves · rights · RfA) administrative privileges are revoked. RHaworth may apply to have them reinstated at any time, either through the usual means or by appeal to the Committee.

Comment by Arbitrators:

Per the below comments, the committee hasn't done this is quite some time. The community decides who is given the admin toolset.

talk) 22:13, 4 January 2020 (UTC)[reply

]

Comment by parties:

Comment by others:

ArbCom isn't really charged with making admins - so "appeal to committee" to get sysop access at some indefinite time in the future is at least unusual for a remedy. This could be seem more of an indefinite "suspension" then a removal perhaps - but in that case the community option shouldn't really be open. — xaosflux ^Talk 20:22, 3 January 2020 (UTC)[reply]

Literally a cut & paste of an ArbCom decision: Wikipedia:Requests_for_arbitration/Tango. I'm not against rephrasing. The general point remains. Buffs (talk) 20:46, 3 January 2020 (UTC)[reply]

This used to be common, but for quite a few years now it has fallen out of favour, with successive committees preferring to leave the decision about whether the user has the community's trust to hold the admin tools solely to the community (via RFA). The Tango case was 2008, compare Wikipedia:Arbitration/Requests/Case/Enigmaman#Enigmaman desysopped from 2019. Thryduulf (talk) 00:22, 4 January 2020 (UTC)[reply]

Struck. v1.1 below Buffs (talk) 16:32, 6 January 2020 (UTC)[reply]

RHaworth is desysopped v2

1.1) For misuse of administrative tools and generally failing to meet community expectations and responsibilities as outlined in

.

Comment by Arbitrators:

Without comment on the whether desysop is warranted, I would avoid use of "et al", replacing it instead with whatever was meant to be covered by this phrase. –xeno^talk 18:19, 8 January 2020 (UTC)[reply]

Comment by parties:

Comment by others:

Rephrase per @Xaosflux: now citing a more recent decision. Buffs (talk) 20:46, 3 January 2020 (UTC)[reply]

Definition of et al: "and others". Feel free to add to the list, but the general tenor is that Admins need to follow the rules like everyone else. Buffs (talk) 20:39, 8 January 2020 (UTC)[reply]

WP:CIVIL, please specify this concern. If not, please remove it. ~ ToBeFree (talk) 14:37, 12 January 2020 (UTC)[reply

]

ToBeFree other policies have been violated, in my opinion. I’ll leave it up to the arbitrators to apply the necessary policies or rephrase as they feel is necessary. Buffs (talk) 05:49, 13 January 2020 (UTC)[reply]

I remain unsure whether this satisfies a minimum necessary decision - if nothing else a CSD prohibition would strike out most cases, though obviously I'm aware that if he moved his significant amount of work elsewhere and nothing else changed we'd just be back in a couple of months. Nosebagbear (talk) 19:41, 19 January 2020 (UTC)[reply]

RHaworth is prohibited from speedy deletion admin actions

2)

Comment by Arbitrators:

Comment by parties:

This is obviously in case #1 doesn't pass. Buffs (talk) 19:35, 3 January 2020 (UTC)[reply]

Comment by others:

You're a party now? Should I be searching your deleted contributions for pages you edited that RHaworth correctly deleted?

Also, for an admin with his overall pattern of actions, this isn't appreciably different from a desysopping. —

Cryptic 15:42, 8 January 2020 (UTC)[reply

]

This is obviously in case #1 doesn't pass. Buffs (talk) 19:35, 3 January 2020 (UTC)[reply]

moved. Giving the option that perhaps RHaworth could use his admin powers in a different capacity; this is an alternative for ArbCom to consider in the event that desysopping is not what they deem appropriate. Buffs (talk) 18:26, 9 January 2020 (UTC)[reply]

RHaworth is formally admonished/warned for his uncivil actions

3) RHaworth (talk · contribs · blocks · protections · deletions · page moves · rights · RfA) is formally admonished by ARBCOM for his uncivil actions regarding speedy deletions and those who contest them; this is unacceptable behavior. If such behavior continues, ARBCOM will desysop RHaworth

Comment by Arbitrators:

Comment by parties:

Comment by others:

This is obviously in case #1 and #2 don't pass. This gives ARBCOM the option of something other than the nuclear (or "large strike") options. Buffs (talk) 19:35, 3 January 2020 (UTC)[reply]

Proposed principle

Communication with new editors

Editors who communicate with new editors should always strive to be helpful, friendly and welcoming, bearing in mind that new editors are unlikely to understand the encyclopedia's policies, nor its wikicode mark-up. This is especially important for administrators.

Comment by Arbitrators:

Agree, wording could be adapted from Wikipedia:Please do not bite the newcomers. –xeno^talk 18:16, 8 January 2020 (UTC)[reply]

Comment by parties:

Comment by others:

Something along these lines appears important; feel free to improve wording. Espresso Addict (talk) 02:09, 7 January 2020 (UTC)[reply]

Agree. Deb (talk) 15:39, 17 January 2020 (UTC)[reply]

Proposed principles

Administrator conduct (proposed by Thryduulf)

1) Wikipedia administrators are expected to lead by example and to behave in a respectful, civil manner in their interactions with others. Administrators are expected to follow Wikipedia policies and to perform their duties to the best of their abilities. Occasional mistakes are entirely compatible with adminship; administrators are not expected to be perfect. However, sustained or serious mistakes are incompatible with the status of administrator, and consistently or egregiously poor judgement may result in the removal of administrator status.

Comment by Arbitrators:

Comment by Parties:

Comment by Others:

Mostly copied from the Rama case. While RHaworth hasn't been grossly uncivil, the evidence page does indicate a lack of respect towards especially new users. I also think it's important to stress that occasional mistakes are not problematic. Thryduulf (talk) 17:25, 17 January 2020 (UTC)[reply]

Administrators must act in accordance with policy

2) Administrators have been granted access to advanced tools not available to ordinary editors, this includes the ability to block and unblocked users and to delete and restore pages. These tools must only be used in accordance with the relevant policies and guidelines, these include, but are not limited to, the Wikipedia:Blocking policy, Wikipedia:Deletion policy and Wikipedia:Criteria for speedy deletion.

Comment by Arbitrators:

Comment by Parties:

Comment by Others:

It should go without saying that administrators must follow policy, but not doing so is at the heart of this case. It wouldn't be a bad idea to include something about community trust in here, but I couldn't work out a way to do so succinctly or clearly and couldn't think of a separate principle about it that didn't mostly duplicate this one. Thryduulf (talk) 17:25, 17 January 2020 (UTC)[reply]

Importance of the deletion log

3) To make it as easy as possible for other administrators to verify speedy deletions were correct and for inexperienced editors to understand why a page was deleted, it is important that the deletion log is accurate and gives the reason for the deletion. In most circumstances best practice is that all jargon, including speedy deletion criteria, and any relevant discussions are explained and/or linked in the summary for this reason.

Comment by Arbitrators:

Comment by Parties:

Comment by Others:

This needs wordsmithing (particularly I don't think it emphasises the need for accuracy enough) but I think its an important principle. An administrators reminded about/administrators encouraged to use best practice remedy might also be worthwhile. Thryduulf (talk) 15:00, 18 January 2020 (UTC)[reply]

@Thryduulf: Are you talking about more than just deleting with the supplied capsule reasons (plus link to any relevant AfD or copyvio url)? Espresso Addict (talk) 17:20, 18 January 2020 (UTC)[reply]

@

G8: Talk page of a deleted page", "Wikipedia:Redirects for discussion/Log/2019 December 2#Pokemon clone closed as delete". I'm talking about summaries like "Blatant vanity" (RHaworth, 19 December), "Page created by a move of a now G5ed DAB created by a banned sockpuppet" (RHaworth, 17 December). In those two cases someone familiar with Wikipedia jargon would probably understand why it had been deleted, but would be impenetrable for a good faith new user wondering why the page was deleted. Thryduulf (talk) 12:14, 19 January 2020 (UTC)[reply

]

Proposed remedies

RHaworth's use of speedy deletion is restricted

1) RHaworth is limited to a maximum of 5 speedy deletions in any 1 hour period (based on the time in the deletion log). Each deletion must explicitly reference at least one speedy deletion criterion.

Any page deleted in contravention of this remedy that is not listed at deletion review may be undeleted by any administrator without discussion. The undeleting administrator must inform RHaworth on their talk page. Any undeletion per this remedy is without prejudice to a XfD nomination by any editor. Pages undeleted (by anybody) are still considered as deleted for the purposes of the maximum 5 speedy deletions in any 1 hour period restriction.

Repeated or egregious violations of this remedy must be notified to the Arbitration Committee at ARCA.

This restriction may be appealed six months after this case is closed or six months after the most recent violation, whichever is later.

Comment by Arbitrators:

I'm afraid that such a narrowly targeted remedy will be difficult to enforce - i.e. cause more drama than it would solve. Restricting an administrators to only 5 summary deletions an hour seems impractical for an administrator to actually do CSD. It would be easier (and close to functionally equivalent) to do a blanket restriction on using the delete tool. Maxim_(talk) 18:22, 8 January 2020 (UTC)[reply]

Perhaps something like “prohibited from deleting more than 25 pages per hour, and from triggering a single deletion event more than once every five minutes” might work. The goal, of course, is to try to ensure each deletion event (e.g. a page + a talk page, or a page+redirects) is given due care and attention. –xeno^talk 20:53, 19 January 2020 (UTC)[reply]

Thryduulf: deletion event would be deleting a set of pages with an immediate relation like talk page of, or redirects to, a deleted page. –xeno^talk 23:02, 19 January 2020 (UTC)[reply]

I've checked my own log, and I have never deleted more than 22 pages a day, and when I reached that number itwas usually clearing up AfD. DGG ( talk ) 22:05, 22 January 2020 (UTC)[reply]

Comment by parties:

Comment by others:

This is intended as an option for consideration as a lesser alternative to full desysopping, with the goal of requiring RHaworth to fully consider every speedy deletion they perform. I'm not sure if it will work (bespoke restrictions do not have a great track record) but at least whether a page has been deleted or not, and whether a deletion summary lists one or more XfD crtieria are objective measures. I'm also frankly unsure if a remedy short of deysopping is justified. The intent is that if an ARCA is brought under this remedy that the committee will desysop unless the violations are found frivolous or otherwise not justified. Wordsmithing is more than welcome. Thryduulf (talk) 15:20, 8 January 2020 (UTC)[reply]

I'd been mulling over something very similar, with respect to both rate and deletion summaries. I'd have suggested something closer to 15 or 20 per hour - I don't know that I've ever spent twelve minutes considering whether to speedy delete a single page - but that sort of number is the kind of tweaking that the arbitrators are perfectly able to handle on their own. Also worth considering whether to include G8 deletions of associated talk and redirect pages as part of the initial speedy deletion.

The wording restriction I'd been mulling included requiring an explicit reference to a speedy deletion criterion - as here - but also a requirement never to use the default deletion summary provided by the speedy deletion tag. This would deal both with logs like the infamous "Wrong Redirect! Not Useful!" on

Cryptic 16:16, 8 January 2020 (UTC)[reply

]

G8's of associated pages is not something I'd thought about, but it does make sense for them not to increment the counter. I can't immediately think of clear way of putting it, given that it would cover talk pages of the deleted page, redirects to the deleted page, and (in some cases, e.g. templates) subpages of the deleted page. Maybe requiring a link to the principle page in the deletion summary - e.g. if deleting

WP:CSD#G8: talk page of deleted Draft:Example". It should only count if RHaworth was the one to delete the primary page - in other cases they should be checking to see that it was correctly deleted and that there was no reason why the dependent page wasn't deleted along with it, but equally if there are three redirects to a page deleted by someone else, the three G8s should count as only 1 of RHaworth's permitted deletions. Which is all getting very complicated, and complicated restrictions are generally poorer than simple ones as there is more chance of inadvertent breach, more chance of incorrect accusations of breaches (in good or bad faith), more chance that intentional breaches will not be spotted and more opportunities for the restriction to be gamed. Thryduulf (talk) 17:50, 8 January 2020 (UTC)[reply

]

I appreciate the sentiment behind this; however, RHaworth has already told me that he has no issue with me restoring any article he deletes, citing

WP:BRD as a reason. [1] I therefore don't think the restriction will work; the first time I decide a deletion from him is out of policy, I will restore it and simply be told "the system works!" Ritchie333 ^{(talk) (cont)} 18:08, 8 January 2020 (UTC)[reply

]

I think the idea behind this is sound, but I don't think it will work in application. Buffs (talk) 17:49, 15 January 2020 (UTC)[reply]

@Xeno: "deletion event" would need to be defined, and that definition would be the crux of whether that restriction would work or not. Thryduulf (talk) 21:54, 19 January 2020 (UTC)[reply]

RHaworth must explain speedy deletions

2) If asked, RHaworth must civilly explain how a page they speedily deleted met the letter and spirit of every speedy deletion criterion listed in the deletion summary. This explanation must be given within a reasonable period of the question being asked (if asked on their user talk page) or being notified of the question (if asked elsewhere).

Comment by Arbitrators:

Isn't this already covered by administrator accountability requirements? –xeno^talk 18:12, 8 January 2020 (UTC)[reply]

Comment by parties:

Comment by others:

Obviously this is only required if they are not desysopped, but the goal is to ensure that (a) they take care with every deletion they do, and (b) give explanations of all deletions when asked. I'm not sure how this would be enforced if breached though. Wordsmithing is more than welcome. Thryduulf (talk) 15:20, 8 January 2020 (UTC)[reply]

I'm pretty sure RHaworth already does this (at least from his point of view!) His talk page has the regular summary "usual clutch of replies" which suggests he does explain all deletions when asked as soon as he is able (albeit not necessarily in a manner that satisfies the target recipient). Ritchie333 ^{(talk) (cont)} 18:16, 8 January 2020 (UTC)[reply]

@Ritchie333: Yes and no, he explains (to a degree) why he deleted it, but doesn't explain how it meets the criteria. Sometimes that's obvious (e.g. a copyright violation) but other times it isn't ("I have the greatest suspicion about your motives", "the article was hopelessly spammy"). This is intended as a way to (semi-objectively) improve the communication. Thryduulf (talk) 21:54, 8 January 2020 (UTC)[reply]

@Ritchie333: fixing the ping. Thryduulf (talk) 21:55, 8 January 2020 (UTC)[reply]

@Xeno: It is covered by how I read the requirements, but apparently isn't by how others read them. I don't know how well my reading aligns with the wider consensus. Thryduulf (talk) 21:54, 8 January 2020 (UTC)[reply]

I've asked for an explanation/clarification and his response was to delete my question; his perplexing archival system makes it tough to find it...I suspect that's intentional. Biographitor had the same response (user was indeed a sock, but no explanation was given in the edit summary and was blocked a week AFTER this request). These aren't the only examples. Buffs (talk) 18:39, 9 January 2020 (UTC)[reply]

Just to clarify, Biographitor's question wasn't deleted, it was moved to another place on the talk page and replied to thus: "Kindly have the decency to wait until someone with no

CoI thinks you are notable and writes about you here." I've also got to say that since RHaworth was told at ANI multiple times that was incivil, I'm not really warming to any remedy below a desysop, but I'm open to persuasion especially if I see RHaworth's right of reply on this. Ritchie333 ^{(talk) (cont)} 19:08, 10 January 2020 (UTC)[reply

]

• The point of speedy deletions is that no explanation is required; reference to the applicable speedy deletion criterion is all the explanation that should be necessary. If you have to explain beyond that, speedy should be declined. Ivanvector (^Talk/_Edits) 20:35, 13 January 2020 (UTC)[reply]
  • Respectfully disagree - that's sufficient for an experienced, good-faith Wikipedia editor, but new users frequently don't understand the criteria in application. They'll need explanations, for example, of why it's not ok to cut and paste from a public (and thus "free") website, why links to press releases aren't an assertion of importance, why it's not ok to repost an article right after AFD deleted it, why a non-free image was deleted for being orphaned despite it being on the draft they just wrote. And that's assuming they see the reason for deletion at all - the capital letter and 1-2 numerals that are all that links to
    Cryptic 21:22, 13 January 2020 (UTC)[reply
    ]
    • Ah, we're talking about explanations to the editor, I didn't parse that. I assumed the intent was to have explanations in the deletion log as justification to others reviewing the deletions. There are templates available with usually-decent canned explanations for different criteria, and Twinkle places them automatically (by default), for example {{db-spam-deleted}}. If you mean that RHaworth should always provide an explanation to the editor whose page he's deleting, I tend to agree, really that should be considered best practice. I don't see if there's evidence that he has not been, although the pattern of new editors going to his page to ask about why their page was deleted seems to indicate not. Ivanvector (^Talk/_Edits) 23:28, 13 January 2020 (UTC)[reply]

Yes. Without reservation. Buffs (talk) 17:45, 15 January 2020 (UTC)[reply]

Administrators reminded (speedy deletion nominations)

3) Administrators are reminded that they must ensure pages nominated for speedy deletion by others meet at least one speedy deletion criterion before deleting. If a page does not meet the criterion under which it was nominated but does meet a different one, it is important that the correct criterion is recorded in the deletion log.

Comment by Arbitrators:

Comment by parties:

Comment by others:

It is not clear that RHaworth has been doing this but it is something that is almost impossible to prove either way. I don't think the issue is likely to be exclusive to RHaworth, and responsibility for pages getting incorrectly deleted this way is shared between the deleting admin and the editor placing the tag. Putting all that together I'm not sure a stronger remedy than this would be practical, especially as the effects should be resolved by other remedies requiring greater care anyway. Thryduulf (talk) 16:24, 15 January 2020 (UTC)[reply]

Regardless of everything else, I'd be fine with something like this (I know some folks find these pointless, but still). I might broaden that first line to include pages not currently nominated; speedy deletion by sysops without prior nomination doesn't appear to be at issue in this case, but it's equally important for the broader corps to mind those situations where it's just one pair of eyes. ~ Amory (u • t • c) 17:00, 15 January 2020 (UTC)[reply]

That's a good point. Every one of RHaworth's speedy deletions I looked at were nominated by someone else, and I don't recall anyone bringing any issues with self-nominated deletions up in evidence, so it's not directly relevant to the case but still very important generally. Thryduulf (talk) 19:00, 15 January 2020 (UTC)[reply]

Absolutely! I think it's pretty clear that RHaworth has repeatedly done this (whether intentionally or due to negligence is irrelevant). Declarations are worth including as ArbCom rulings can be used for a basis of updating our policies, guidelines, and procedures. They are a worthy endeavor. I especially like the use of the word "must". Buffs (talk) 17:26, 15 January 2020 (UTC)[reply]

I think a general reminder like this is a very important outcome, because RHaworth isn't remotely the only admin who pays little attention to the CSD criteria and deletes pages out of process. I saw many in my time but didn't have the energy to try battling against them, and hopefully this case might wake up some more CSD abusers. Boing! said Zebedee (talk) 19:14, 15 January 2020 (UTC)[reply]

I think this proposal is crucial. As others have said, RHaworth is not the only admin who doesn't always give an independent evaluation to pages that are marked for speedy deletion and there are some admins who delete at a far more rapid rate that he does. I'm not minimizing the evidence presented, I just think that this isn't an isolated problem.

I always feel better about admins who I see occasionally removing a speedy tag and posting a notice to the tagging editor advising them that the tag was inappropriate in a particular instance...it shows that they are looking at tagged pages carefully. But there are some admins who don't seem to question any pages that are CSD'd. And I've also been chided for removing tags, especially for A7s and G11s, from the tagging editors who are upset at being contradicted so I think it would also be helpful to remind our more active speedy taggers to be more careful when they evaluate pages. Liz ^{Read! Talk!} 05:29, 22 January 2020 (UTC)[reply]

Proposed principles

Deletion is not ordinary editing

1) Page deletion, unlike ordinary editing, can only be reviewed by admins, and even for an admin is far less convenient to review, because deleted pages do not appear in searches in the same way that other pages do. Thus the "many eyes" principle cannot be depended on to correct incorrect or unwise deletions, and BRD is not an appropriate model for deletions, even if the deleting admin freely allows other admins to undelete if they see fit.

Therefore, deletion is only appropriate with consensus, either through an XfD discussion, or through total lack of objection to a PROD, or the advance consensus granted via

. This advance consensus is explicitly limited to the exact terms of the written criteria. Admins do not have consensus to delete pages not meeting any criteria, even pages an admin thinks a waste of time, sure to be deleted after a discussion. Deletion of pages clearly not meeting any criterion is an abuse of the admin permissions. Habitually deleting pages clearly not meeting any criterion is a serious abuse, potentially requiring a sanction. 13:50, 10 January 2020 (UTC)

Comment by Arbitrators:

Comment by parties:

Comment by others:

Very well put. Boing! said Zebedee (talk) 04:44, 10 January 2020 (UTC)[reply]

Agree. Espresso Addict (talk) 04:58, 10 January 2020 (UTC)[reply]

Spot on, and this is very important. Thryduulf (talk) 09:57, 10 January 2020 (UTC)[reply]

• Note, final sentence added after the three comments above had been made. DES ^(talk)_{DESiegel Contribs} 13:52, 10 January 2020 (UTC)[reply]

Agree totally, and RHaworth's claim it's like BRD is perplexing. Ritchie333 ^{(talk) (cont)} 17:26, 10 January 2020 (UTC)[reply]

I wonder if people haven't been taking the two BRD diffs too literally. Taken in context, I read both as blanket permission to reverse his admin actions, without necessarily saying anything about how he goes about deleting things - that is, he's referring only to the RD part. Plenty of admins give similar permission - for example,

Cryptic 23:05, 12 January 2020 (UTC)[reply

]

WP:CSD nor log his exact reasons, since if anyone else finds a problem they can jsut undo it. If I ahve correctly understood the intent of his claim, this is a problem, because deletions are not easy to monitor, even thsoe of an admin with only a moderate deletion count, such as myself. Given the numbers of deletions made by this admin (now detailed on the evidence page) meaningful review of even a small fraction is impossible, and thus the community must depend on a serious effort at meticulous care. RHaworth has not, to the best of my understanding, claimed to deliver such care, and the evidence suggests that he at least sometimes does not. If all he means is that he per-accepts undeletion by ohers that is different, but I don't think his remarks will bear that interpretation.DES ^(talk)_{DESiegel Contribs} 07:52, 13 January 2020 (UTC)[reply

]

In [https://en.wikipedia.org/w/index.php?title=User_talk:Tavix&diff=prev&oldid=898343312 thid edit} when RAH writes So to all admins I say: if you see some bold action that I have done and that you disagree with, simply revert it - don't tell me - just revert it he is saying not only that his permission is not needed, but that no public report or discussion is needed because the system has worked, someone else has corrected any problem. Of course this leaves out of account the cases (and I beloive there to be many) where no one has reviewed, no one has undeleted, and the system has therefore not worked. The moment anyone brings one up, he says "so revert it, no problem". DES ^(talk)_{DESiegel Contribs} 08:05, 13 January 2020 (UTC)[reply]

Failure to include this in whatever conclusion ArbCom comes to would be a travesty. Well said! Buffs (talk) 18:04, 13 January 2020 (UTC)[reply]

I would add that such a ruling by ArbCom should be immediately incorporated into our policies/guidelines. Regardless of any current vagueness in our "rules", RHaworth has violated so many other principles that, IMHO, any ignorance or disagreement on this subject is the result of willfully ignoring community feedback as well as current guidelines and policies. Buffs (talk) 17:34, 15 January 2020 (UTC)[reply]

Well put. It could be that despite RHaworth's willingness to have his actions challenged, it has not occurred to him nor been explained as succintly why it is so much more challenging to review deletions versus other kinds of bold-but-visible actions. Ivanvector (^Talk/_Edits) 20:38, 13 January 2020 (UTC)[reply]

Agreed. The community has created a very broad power for admins to delete things without discussion. But that is limited by the expected strictness of the criteria for deletion. And because any non-discussion-based deletion (so not at xfD) has minimal documentation, it's really hard (basically impossible) for a non-admin to review. So the admin needs to be following the rules or they should be stopped. Hobit (talk) 00:02, 16 January 2020 (UTC)[reply]

Explaining admin actions

2) Administrators are expected to explain their actions on request, as per

and indeed should hold themselves to a higher standard of civility than is required of editors in ordinary editing collaboration, precisely because administrative actions, particularly deletion and blocking, are important, can have significant effects, and often can only be addressed by administrators, unlike editing discussion where any editor can join a discussion and offer a possibly helpful view.

Administrators are expected to explain their actions on request, as per

and indeed should hold themselves to a higher standard of civility than is required of editors in ordinary editing collaboration. This is expected precisely because administrative actions, particularly deletion and blocking, are important, can have significant effects, and often can only be addressed by administrators. In this respect admin actions are unlike editing discussion where any editor can join a discussion and offer a possibly helpful view.

When the editor asking for an explanation is, or might well be, a relative newcomer, unfamiliar with Wikipedia policy and practice, it is an admin's role to explain what is going on, or to refer the editor to existing explanations, or to places, such as the

is not.

Persons not willing to explain admin actions in a civil and helpful manner should not undertake them. Habitual violation of this expectation may lead to sanctions. DES ^(talk)_{DESiegel Contribs} 18:22, 10 January 2020 (UTC)[reply]

Comment by Arbitrators:

Comment by parties:

Comment by others:

This is really several Principles in one but want to highlight "Asking for enough information to clearly identify, say, a deleted page is reasonable -- insisting that it be formatted as a proper

wiki-link is not." as an important issue in interactions I've observed. Best, Barkeep49 (talk) 05:06, 13 January 2020 (UTC)[reply

]

Barkeep49, This cold be split up into multiple principles, and of course the Arbs are free to edit proposals as they see fit. But all of these are expectations of admins when responding to questioins about admin actiosn, and IMO they all go together, each part making the others stronger and more dignificant, and violatiosn of several parts being more of a problem than violatiosn of any one part alone. DES ^(talk)_{DESiegel Contribs} 08:09, 13 January 2020 (UTC)[reply]

Yeah I don't disagree with anything you write and wanted to make sure this wasn't lost amidst everything going on in this page. Best, Barkeep49 (talk) 17:12, 13 January 2020 (UTC)[reply]

In general, yes, this should be included, but the first "sentence" has so many clauses, I'd break it into 2 or 3 sentences rather than such a long run-on sentence. Buffs (talk) 17:37, 15 January 2020 (UTC)[reply]

Now revised in accord with yoru suggestion, Buffs DES ^(talk)_{DESiegel Contribs} 01:42, 22 January 2020 (UTC)[reply]

Proper use of OS and CU blocks

3) Checkusers and Oversighters are empowered to make blocks based on private information, which cannot be undone by non-functionaries without consulting a CU or OS editor. This greatly restricts the normal unblock process. Therefore no block should be so designated unless non-public information was actually used in the decision to block. Where the reason for blocking is apparent from public on-wiki information (such as diffs and log entries) and private info does not significantly add to the reason for a block, even if it was checked, it is wrong for a block to be designated as a CU or OS block, and doing so may in future subject a functionary to sanction, up to loss of status. DES ^(talk)_{DESiegel Contribs} 01:11, 22 January 2020 (UTC)[reply]

Comment by Arbitrators:

Comment by parties:

Comment by others:

Apparently some CUs think that any b;lock by a CU, even if it did not depend on private info, may or should be designated as a CU block. That should be clearly stated to be against policy. DES ^(talk)_{DESiegel Contribs} 01:11, 22 January 2020 (UTC)[reply]

None of what you have written in the last sentence above has foundation in policy but since you have claimed that is against policy then please link to it and quote it. It is standard practice to run checks on accounts and then mark them as checkuser blocks based on that private info.
 — Berean Hunter (talk) 03:23, 22 January 2020 (UTC)[reply]

In

CheckUser#CheckUser_blocks it is said that heckusers can block accounts based on technical (checkuser) evidence. They will make clear in the block log summary that they have blocked as a "checkuser action". It seems to me that this pretty clearly implies that blocks not "based on technical (checkuser) evidence." should not be marked as checkuser blocks. I admit that I had thought that there was a clearer statement of this on that page, and perhaps I will propose adding one. But since ARBCOM both appoints and removes CUs, and can do so "for cause", while the exact possible causes have been left unenumerated, that unlike in most areas, ARBCOM seems empowered to make policy on this issue. Still my statement above may have been stronger than is justified by the current policy, although in my opinion it should be clear policy. DES ^(talk)_{DESiegel Contribs} 04:01, 22 January 2020 (UTC)[reply

]

My understanding (which may be incorrect) is that in the situation which triggered this arbitration case, an editor admitted to having used a second account improperly. A CU used technical evidence to confirm this, and to check for possible other socks, and found none. The editor was blocked, and the block was marked as a CU block, b ut the block did not in fact depend on any technical evidence, it was fully supported by the editor's admission and the editing history. Nor were any additional facts relating to the severity or frequency of the abusive editing revealed by any technical evidence, and any admin who read through the editing history was in fact in possession of all the evidence needed to properly evaluate the block and pass on an unblock request, but could not know that this was true, because of the CU block designation. Assuming for purposes of argument that all this was true in a hypothetical case (whether this matches the facts in the current case or not), I think a CU designation would have been improper, and should subject the CU to admonishment at least, in a future case with these facts. If it has been common practice to mark a block as a CU block in such circumstances, that practice is wrong and should change, IMO. DES ^(talk)_{DESiegel Contribs} 04:01, 22 January 2020 (UTC)[reply]

This is not relevant to this case. Given a non-CU/non-OS cannot know whether private information was used (even if there is (also) public evidence) they cannot know whether a block marked as a CU or OS block was marked correctly or not. Accordingly they must treat it as though it is valid and either leave it be or consult with someone with the appropriate permission. If there is a need to clarify/amend anything about the placing of CU or OS blocks or examine whether any are being placed incorrectly then that needs to be done separately to this case. Thryduulf (talk) 09:58, 22 January 2020 (UTC)[reply]

It is not relevant to assessing the actions of RHaworth, no. RHaworth was required to take the notice of a CU block at face value, or to ask a CU to clarify or confirm if there was doubt. But I was under the impression that the actions of the initially blocking CU were also in scope for this case. This proposed principle is aimed at that CU, and at similar future actions. DES ^(talk)_{DESiegel Contribs} 14:33, 22 January 2020 (UTC)[reply]

@DESiegel: Only RHaworth and TonyBalioni (the filter) are parties to this case, and the scope is explicitly given at Wikipedia talk:Arbitration/Requests/Case/RHaworth/Evidence#Scope? by SQL who quoted the instructions given to the clerks as Scope will be restricted to the admin actions of RHaworth and associated behaviour... Thryduulf (talk) 15:15, 22 January 2020 (UTC)[reply]

Very well, Thryduulf struck as out-of-scope. I still think this should be addressed, but this is apparently not the time and place. DES ^(talk)_{DESiegel Contribs} 15:46, 22 January 2020 (UTC)[reply]

"an editor admitted to having used a second account improperly" The CU block was placed before they admitted the other account in their unblock request. The block did depend on technical evidence as checks were run. The initial complaint about the cu block was concerning the timeline of the accounts but that has been shown as being without merit. What you are stating in the above is a new and novel complaint. Regarding the justification for running the checks as well as placing the cu blocks, it has not been shown that this was outside of policy. I agree with Thryduulf and believe this is out of the scope of this case.
 — Berean Hunter (talk) 15:19, 22 January 2020 (UTC)[reply]

I had not folloewed the comments on the initial block carefully Berean Hunter, which is why I explicitly made this a hypothetical. If the CU check was proper in this case, good. From comments made in this case I still get the impression that some CUs mark blocks as CU blocks when technical evidence has been checked, but ads nothing to the publicly available evidence. But it seems this was not such a case. DES ^(talk)_{DESiegel Contribs} 15:46, 22 January 2020 (UTC)[reply]

Proposed findings of fact

RHaworth frequently makes deletions not supported by any of the

WP:CSD

1) RHaworth has, on many occasions, deleted pages for "vanity" or 'not notable" or similar reasons which might well support a deletion at an AfD (or other XfD), but which do not fit within any of the listed criteria at

]

Comment by Arbitrators:

Comment by parties:

Comment by others:

RHaworth frequently makes deletions not clearly explained in the log

2) RHaworth has, on many occasions, deleted pages with vague and general reasons in the deletion log. These pages may have in fact been subject to one or another of the

]

Comment by Arbitrators:

Are you able to provide a few examples? –xeno^talk 12:51, 22 January 2020 (UTC)[reply]

Comment by parties:

Comment by others:

From the evidence: "Wrong Redirect! Not Useful!" (

Cryptic 14:22, 22 January 2020 (UTC), 14:47, 22 January 2020 (UTC)[reply

]

Thank you. –xeno^talk 14:28, 22 January 2020 (UTC)[reply]

• "Wikipedia is not for eulogies" (was tagged G11, but could've been salvaged by cutting the worst sections; actually a G12 from [2], but I doubt anyone checked)
• "new articcle [[J Narasimha Swamy]] created" (plainly invalid, per
  WP:RDRAFT
  )
• "Content is not in English" (
  WP:NOTCSD
  #16. I didn't machine translate to check for other criteria)
• "This is not a FP on en.wiki ([[WP:TW|TW]])" (arguably vandalism, but I wouldn't care to try)
• "housekeeping ([[WP:TW|TW]])" (equivalent to no deletion comment at all)
• "no context" (not only is that an A-series criterion, there was ample context in the draft title)

Also

Cryptic 14:41, 22 January 2020 (UTC)[reply

]

Proposed remedies

Note: All remedies that refer to a period of time, for example to a ban of X months or a revert parole of Y months, are to run concurrently unless otherwise stated.

RHaworth must clearly log and explain deletions

1A) All deletions by RHaworth must have log entries that include a specific speedy deletion reason with a link to the proper section of the

WP:CSD

or a link to the relevant deletion discussion.

1B) RHaworth must notify the original creator, and/or at least one major contributor, on such user's user talk page, of each deletion and the reasons for it, in addtion to the deletion log. This may be done by template, or by any other reasonable method, but must include a clear explanation, and a link to the proper criteria or discussion, for each such deletion.

1C) RHaworth must, on request, explain clearly and in a

civil

manner, how any deletion qualifies under the relevant CSD or other deletion policy.

Comment by Arbitrators:

Doesn’t 1B create an increased and possibly unattainable standard? Aren’t creators generally notified by the tagger? Would RHaworth be expected to notify creators of abusive pages (would run counter to wp:deny...) –xeno^talk 12:48, 22 January 2020 (UTC)[reply]

Comment by parties:

Comment by others:

This is a variant on remedies proposed by others since there is no time for a back-and-forth with the other proposers. i think it is important to insist on both a proper log and a proper explanation when requested, and this ties to my proposed findings of fact. DES ^(talk)_{DESiegel Contribs} 01:46, 22 January 2020 (UTC)[reply]

Obviously the arbs may mix and match ideas from this with ones from other proposals. DES ^(talk)_{DESiegel Contribs} 01:47, 22 January 2020 (UTC)[reply]

There needs to be some limited exceptions to the requirement to explain - G5 deletions and non-speedy deletions being obvious ones, similarly anyone who has requested RHaworth not post on their talk page (I don't know if anyone has, but the Committee should not overrule their wishes if there are). Blocked editors in at least some circumstances too. G7 and U1 do not need exceptions, but something like "I've deleted $page as you requested, if you want it back see

WP:REFUND" would be fine. Also, deletion of multiple related pages by the same author/contributor can be covered in one notice. Thryduulf (talk) 10:15, 22 January 2020 (UTC)[reply

]

I personally would do zero deletions if under a 1B restriction. Twinkle might make it tolerable - I don't know, it conflicts with my own javascript - but even if it does, it's repugnant to effectively require the use of a fragile add-on to perform basic editing or administrative functions. Mitigated somewhat by the fact that RHaworth does use Twinkle. 1A is a reasonable restriction; 1C is already required of all administrators by

Cryptic 15:00, 22 January 2020 (UTC)[reply

]

RHaworth must not make more than 20 deletions in any 1 hour period

2) RHaworth must not make more than 20 deletions of any sort in any 1 hour period. For purposes of this remedy, deletion of a page, its talk page, and any sub pages will be considerd a single deletion. This remedy may be appealed after a period of 6 months.

Comment by Arbitrators:

Comment by parties:

Comment by others:

Redirects to the deleted page should also be considered part of that deletion. I would support a limit this high only in conjunction with a requirement to notify of every speedy deletion (excluding G5 deletions) - three minutes per deletion is still quite a lot. Thryduulf (talk) 10:09, 22 January 2020 (UTC)[reply]

Proposed principles

Blocks based on private information

1) A small number of trusted administrators, known as functionaries, are granted special, limited access to non-public information for various purposes, including to protect editors from serious disruption, to suppress harmful information about editors and other living persons, to manage and prevent disruptive actions by malicious users, and to review the actions of other functionaries. The functionaries have the authority to block accounts based on this non-public information, and will clearly indicate when such a block is enacted most commonly through a notation in the block log. Regular administrators cannot review non-public information, and so they must not undo the clearly-marked actions of a functionary without consulting with that functionary or with other functionaries capable of accessing the same non-public information, even if they genuinely believe that such a block is improper.

Comment by Arbitrators:

Comment by parties:

Comment by others:

Adding this because nobody else has yet, basically. It seemed to have come up in the case request that several users believe that an administrator who believes they have adequately deduced the meaning behind all our cryptic findings and tags can in good faith skip the step of consulting with the functionaries. That is a dangerous interpretation especially when it comes to oversight blocks, and I would like Arbcom to make a strong statement that that is not the case. My wording could probably use assistance. Ivanvector (^Talk/_Edits) 21:22, 13 January 2020 (UTC)[reply]

I agree completely with what you are saying here, and something along these lines is needed. However, your wording definitely needs improvement (I'm a functionary and it took me a few attempts to parse the last sentence!) - I'll have a go at improving it in a day or so if nobody beats me too it. Thryduulf (talk) 23:11, 13 January 2020 (UTC)[reply]

Absolutely. Applies to everyone, not just RHaworth. Buffs (talk) 17:44, 15 January 2020 (UTC)[reply]

TonyBallioni's #CheckUser and Oversight blocks (below) is a much better phrasing of what I intended here, so I have struck this proposal. Ivanvector (^Talk/_Edits) 17:49, 21 January 2020 (UTC)[reply]

Wikipedia has no firm rules

2) The

" policy. The principle invites users to ignore written rules which prevent them from improving the encyclopedia in good faith. However, the pillar cautions against reckless ignorance of the written rules.

Comment by Arbitrators:

Comment by parties:

Comment by others:

In a nutshell: IAR is not a

get out of jail free card. Ivanvector (^Talk/_Edits) 21:22, 13 January 2020 (UTC)[reply

]

A few points

• The 5th pillar states "Be bold but not reckless in updating articles". I do not believe "deletion" of an entire article would fall under an article update. Nor do I see this as an an accurate summary of a portion of 5P5.
• Our deletion policies clearly annotate "Pages can be deleted without any discussion if they meet one of the criteria for speedy deletion. Speedy deletion is meant to remove pages that are so obviously inappropriate for Wikipedia that they have no chance of surviving a deletion discussion. Speedy deletion should not be used except in the most obvious cases."
• RHaworth describes his administrative actions as in line with
  WP:BRD
  , but BRD only discusses editing articles. It mentions nothing about administrative actions.
• If we're even going to entertain the concept of
  WP:IAR
  justifying these actions and treatment of noobs, they should be rare, not ongoing for 4+ years with numerous actions.
• RHaworth isn't accused of being ignorant of the rules. He full well knows them. Despite objections and complaints, he refuses to appreciably change his behavior. I can find no attempt in the past 4 years to attempt to change the rules on his behalf. This isn't a one-off case of
  WP:IAR
  or "whoopsies". It's a long-term problem.

Accordingly, I would rephrase the last sentence: "However, the pillar cautions against reckless actions." Buffs (talk) 22:29, 13 January 2020 (UTC)[reply]

• I oppose including this in any form. As others have said above, both IAR and BRD apply, by their terms, only to editing articles, not to deletion, or other administrative policies (such as blocking). Including this would imply the contrary, even with the limitations suggested. DES ^(talk)_{DESiegel Contribs} 00:49, 22 January 2020 (UTC)[reply]

Proposed remedies

RHaworth must explain all deletions

1) Whenever RHaworth deletes any page, he must make a reasonable effort to notify the page creator and/or any significant contributors on their user talk pages. Notification may be through the placement of an appropriate template included in

) which notifies users automatically, or through a customized civil message on the user's talk page.

Comment by Arbitrators:

As I understand it, RHaworth is generally responding to tags placed by other editors and the creators were generally notified by the taggers those 76 pages mentioned. Would RHaworth be expected to leave a follow-up explanation? –xeno^talk 14:18, 22 January 2020 (UTC)[reply]

Comment by parties:

Comment by others:

This is a different version of Thryduulf's #RHaworth must explain speedy deletions, but rather than compelling an explanation only if asked, this compels RHaworth to provide an explanation by default. On 20 December 2019, RHaworth deleted 76 pages, and provided user talk page explanations for none of them; I haven't checked to see how many of those actions resulted in inquiries on his user talk page. This proposed remedy is intended to allow him to delete pages but to get him in the habit of explaining to the primarily new users whose pages are being deleted why they're being deleted, without them having to go to his talk page where the replies are often terse. A side effect is this slows him down. I believe the wording is loose enough to allow for the occasional exception, for example notifying a blocked sockpuppet that their article was G5 deleted is not a productive use of one's time, but I don't want to get into listing every possible exception to a remedy here. Ivanvector (^Talk/_Edits) 16:21, 20 January 2020 (UTC)[reply]

@Xeno: While the nominator should inform the creator this does not always happen, so RHaworth (or other deleting administrator) should provide that if it is missing. I'd say an explanation should be given if the page was deleted for a different reason to the nomination. e.g. Finesse & Synquis (the first deletion I analysed in my evidence) the nominator moved a page created by someone else and nominated the resulting redirect for speed deletion under criterion G7 (which explicitly does not apply to that situation) without informing the creator of the move or deleting nomination. RHaworth deleted this as G7 per the tag, but it would have been a valid R2 speedy deletion.
In cases where the nomination was correct and everybody correctly informed already then a simple "I have deleted the page for the reason given above" or similar would fulfil this requirement (imo) if there is no exception for such circumstances. I still prefer to limit this to on request (as I and DESiegel have proposed) though as this removes the need for any of this complexity. Thryduulf (talk) 15:29, 22 January 2020 (UTC)[reply]

Hm, was the initial move in order? –xeno^talk 15:56, 22 January 2020 (UTC)[reply]

@Xeno: It was moving a newly created very short stub article from the article namespace to the draft namespace - see Draft:Finesse & Synquis and its history (only the redirect was deleted; based on the article notability is questionable but it was not an A7 candidate). I believe this is allowed and while notifying the creator of the move and new location is absolutely best practice but is not required. I'm open minded about whether this sort of move should be allowed without discussion, but I believe if it is then notification should be required in almost all circumstances. RHaworth's only involvement was the deletion though so going further into these moves is something for a different venue (please do ping me to any such discussion though). Thryduulf (talk) 16:58, 22 January 2020 (UTC)[reply]

My understanding is that current policy permits such moves by any editor if done in good faith, but some editors have suggested that some degree of consensus should be required. Certianly a number of experienced editors make such moves without discussion fairly routinely. Since the redirect will either be suppressed or promptly deleted as a cross-namespace redir, without a notification the original editor has no easy way to find the page. None of which seems to be relevant to RHaworth's actions in this instance. Thryduulf is right about that, I think. DES ^(talk)_{DESiegel Contribs} 17:27, 22 January 2020 (UTC)[reply]

Draftifying such as this has become a very common practice (rather to my dismay). It tends to get done by a script which automatically notifies the creator. The burden of notification seems to me to fall on the draftifying editor, rather than the redirect-deleting admin, in this case. Espresso Addict (talk) 20:56, 22 January 2020 (UTC)[reply]

Admins are already required (by

WP:ADMINACCT

) to explain any action when asked, with loss of rights available as a remedy for non-compliance, so (IMO) Thryduulf and DESiegel's proposal is redundant. If evidence shows that RHaworth is habitually not providing explanations when asked then he should be at least formally admonished, and if as has been alleged he is purposely not providing explanations to certain classes of editors, he should be desysopped for conduct unbecoming. But xeno, you have a good point that we normally expect taggers to notify when tagging an article for deletion, and I hadn't thought of that. Requiring RHaworth to first determine if notification has occurred and then to notify only if not seems like an unreasonable burden to me, but Thryduulf's suggestion of saying he has deleted for the reason already given (or "per discussion" if that condition applies) seems like a good solution.

As for draftifying, I think it's out-of-scope as moving a page to draft is not deletion, and most importantly the page remains visible. If it should be a requirement to notify a page author that their page has been moved to draft space (and I agree it should, especially because the {{

deleted in draftification) then that should be a global requirement, not a restriction for one administrator. Ivanvector (^Talk/_Edits) 14:23, 23 January 2020 (UTC)[reply

]

IMHO, RHaworth has already been adequately informed by the community, but

refuses to change. Even in this case, we've openly asked for an explanation. Via proxy, he seems to have no explanation/justification despite multiple queries. I concur that THIS proposal is redundant. Buffs (talk) 21:39, 23 January 2020 (UTC)[reply

]

Proposed principles

CheckUser and Oversight blocks

1) Because they have access to private information unavailable to other administrators, CheckUsers and Oversighters may sometimes block users in these capacities. The use of these templates is designed to ensure that the individuals reviewing the blocks have full access to all relevant information. The Arbitration Committee has consistently held that administrators who unblock users who have been CheckUser or Oversight blocked without the consent of a CheckUser or Oversighter may be desysopped.

Comment by Arbitrators:

Comment by parties:

Threw in OS blocks here because I think it makes sense just to lump it into one principle as this is likely going to be cited in some policy at some point, and the same principle applies: blocks placed by functionaries in their capacity as an operator of advanced permissions cannot be overturned by a sysop who does not have access to the information. TonyBallioni (talk) 05:18, 17 January 2020 (UTC)[reply]

Comment by others:

Better wording of what I said. I made a very minor spelling correction (desysoped -> desysopped). Ivanvector (^Talk/_Edits) 15:25, 17 January 2020 (UTC)[reply]

Quite true, but IMO requires the companion principle I have proposed of not designating a CUor OS block when private info was not in fact relied on. DES ^(talk)_{DESiegel Contribs} 01:51, 22 January 2020 (UTC)[reply]

Per my comments there, I disagree with that proposal. Administrators who do not have CU/OS permission must always treat blocks marked as such as being correctly marked, so this proposal needs no companion. Thryduulf (talk) 10:04, 22 January 2020 (UTC)[reply]

Administrators and policy knowledge

2) While administrators are not expected to know every facet of policy, they are expected to generally be aware of the policies and guidelines surrounding use of their advanced permissions. In particular, administrators should be aware of the policies on blocking and deletion as actions related to these technical abilities are monitored closely by many in the community.

Comment by Arbitrators:

Comment by parties:

Fine with some wordsmithing, but expressing the basic principle that administrators should know how the blocking and deletion policy works as these are the two most controversial uses of the tools, and with the addition of protection are basically the "core" or the technical sysop toolkit. TonyBallioni (talk) 05:18, 17 January 2020 (UTC)[reply]

Comment by others:

This is important. It's also important to note that if another admin points out that you got part of policy wrong that you either accept that you were wrong and don't repeat the error, or discuss the matter and in future act according to the consensus of that discussion. What you don't do is ignore the consensus of half a dozen or more ANI threads all agreeing that your interpretation of policy is wrong. Thryduulf (talk) 11:47, 17 January 2020 (UTC)[reply]

^^^ This! Buffs (talk) 16:34, 17 January 2020 (UTC)[reply]

Proposed findings of fact

RHaworth removed a CheckUser block inappropriately

1) RHaworth unblocked Velanatti, who at the time was subject to a CheckUser block, without discussing the block with a CheckUser and obtaining their consent to unblock.

Comment by Arbitrators:

Comment by parties:

This needs to be a finding of fact. It's a bright line rule that virtually every administrator knows. Even if there is just an admonishment, the fact that he removed a CU block is critical, because otherwise no one would have bothered to file the case. The deletions show a pattern of ignoring policy. The CU unblock shows him ignoring policy on one of the only brightlines for administrators. TonyBallioni (talk) 05:18, 17 January 2020 (UTC)[reply]

Comment by others:

Agreed. This was the impetus for bringing the case and was the straw that broke the camels back regarding admin actions generally. Thryduulf (talk) 11:48, 17 January 2020 (UTC)[reply]

RHaworth has misrepresented suppression

2) RHaworth had previously contacted the Oversight team regarding suppression, and had been informed of the correct methods of inquiry, both via email from an arbitrator and off-wiki by a member of the oversight team. Despite this, he has misrepresented the policy to new users and others.

Comment by Arbitrators:

Comment by parties:

Basically see Vanamonde93's evidence. I'll also do analysis in a bit showing that he specifically inquired about the user in the BN thread after giving them bad advice, and didn't go back to correct it after being told in person what to do, even though he was editing before the BN thread was made.

Comment by others:

This is important to note. Thryduulf (talk) 11:49, 17 January 2020 (UTC)[reply]

Proposed remedies

Note: All remedies that refer to a period of time, for example to a ban of X months or a revert parole of Y months, are to run concurrently unless otherwise stated.

RHaworth desysoped

1) RHaworth is desysoped for conduct unbecoming an administrator. He may regain administrator access at any time via a successful request for adminship.

Comment by Arbitrators:

Comment by parties:

Keep it simple and just say conduct unbecoming. Yeah, there are about 1000 policy violations, whether inadvertent or intentional, but regardless it boils down to the fact that he's not behaving in a way we would expect admins to behave in. I'm going to argue it's the only outcome that will work because every other restriction here is basically a boutique sanction just for him, and boutique sanctions tend not to work. TonyBallioni (talk) 05:18, 17 January 2020 (UTC)[reply]

Comment by others:

I think that, if we keep it too simple without mentioning explicitly WHY he is desysopped, it will dilute the effect. I think we should specify which policies were violated and rise to the level of desysopping. Buffs (talk) 16:37, 17 January 2020 (UTC)[reply]

I agree with Buffs that explicit is best here. There are things that RHaworth has done that are poor but not desysopworthy, there are things that rise to that level only when repeated time and again, and there are things that are bright line transgressions on their own. It benefits nobody to conflate them. Thryduulf (talk) 17:02, 17 January 2020 (UTC)[reply]

I agree with Thryduulf's statement There are things that RHaworth has done that are poor but not desysopworthy .... I think there's room for improvement, but I think that would be true for most, if not all, of us. — Preceding unsigned comment added by Ched (talk • contribs) 22:42, 17 January 2020 (UTC)[reply]

I agree with Thryduulf and Buffs here. If we are going to desysop, we should make it very clear just what actions justify that step, and what would or would not be likely to result in similar measures applied to others in future. DES ^(talk)_{DESiegel Contribs} 01:55, 22 January 2020 (UTC)[reply]

Administrators reminded (lifting OS and CU blocks)

2) Administrators are reminded that they must consult with a CheckUser or Oversighter before lifting blocks labeled CheckUser or Oversight blocks. Unblocking accounts marked this way without the consent of a CheckUser or Oversighter may lead to removal of administrator access.

Comment by Arbitrators:

Comment by parties:

So no one says they didn't know again. Having a remedy posted at AN might make people pay attentions. Or at least serve as a reminder. TonyBallioni (talk) 05:18, 17 January 2020 (UTC)[reply]

Buffs, it’s a version of the most litigated word in English (“shall”) problem. Some people use “should” to mean “must”. I’m not a lawyer, but the advice I’ve been given in every business and contracts law class I’ve taken is just to use “must” to mean “must” so there’s no argument. Also pinging Ivanvector on this since his interpretation of “should” is different than mine. I read it in current policy to mean “must” and I think a statement here would help clarify. TonyBallioni (talk) 18:16, 18 January 2020 (UTC)[reply]

Comment by others:

Yes, every administrator should already know this, but a reminder cannot hurt. Thryduulf (talk) 11:51, 17 January 2020 (UTC)[reply]

Absolutely. Moreover, it is worth an official ruling from ArbCom so it can be incorporated completely into policy (as noted elsewhere, portions of our current guidelines and policies still say "should" when it really should be "must"). Buffs (talk) 16:38, 17 January 2020 (UTC)[reply]

@TonyBallioni: the context in which I am most familiar with this issue in is the Highway Code (I'd guess the same will be true of a good proportion of UK editors), where "must"/"must not" designate rules based in criminal law while "do/don't" and "should/should not" are advisory (Introduction to the Highway Code: Wording of the Highway Code). Obviously in a Wikipedia context doing/not doing something is not a criminal offence but changing policies so that "should" is consistently advisory and "must" is consistently mandatory would certainly be something I'd support. Thryduulf (talk) 11:53, 19 January 2020 (UTC)[reply]

(responding to ping) Generally agree with Thryduulf, as usual. In my interpretation, "should not" means proceed with caution and be prepared to defend your action; "must not" means do not proceed or you will face consequences. The question is whether the policy says "only undo checkuser/oversight blocks if you have a good reason" or "never undo checkuser/oversight blocks for any reason". If that did not need to be stated more clearly, we would not be here. Ivanvector (^Talk/_Edits) 19:46, 19 January 2020 (UTC)[reply]

It's definitely a question somewhat outside the scope of this ArbCom matter, but I think it's worth reviewing our policies and figure out where "must" and "should" should be swapped. Without a clear framework, we're going to confuse people. I spent over a couple of decades in aviation. Must=required/mandated. Should=standard or desired actions. Let's be clear. Buffs (talk) 17:26, 20 January 2020 (UTC)[reply]

Proposed remedies

Note: All remedies that refer to a period of time, for example to a ban of X months or a revert parole of Y months, are to run concurrently unless otherwise stated.

No CSD without prior tag

1) RHaworth is not to speedy delete any page that does not have a CSD tag on it placed by another user.

This restriction does not apply to

.

This restriction may appealed 6 months after the closure of this case, and every 3 months thereafter.

Comment by Arbitrators:

Comment by parties:

Basically what Thryduulf said. My entire point on arguing for a desysop has been that with both the level of what’s happened and RH’s general activity as an admin, there really isn’t anything that will work short of a desysop. Boutique sanctions are difficult to enforce, and for prolific admins, they’re bound to be violated at some point even if just on accident. Anything less than a desysop would functionally amount to one: either in scope or in a few months. TonyBallioni (talk) 18:29, 18 January 2020 (UTC)[reply]

Comment by others:

This is designed to ensure two pairs of eyes on everything - currently admins can delete without this (a discussion on it closed as no consensus recently), but as an intermediate step I think it reduces certain risks. Obviously he'd still be entitled to tag pages like any other user. Nosebagbear (talk) 11:37, 18 January 2020 (UTC)[reply]

The issue isn't RHaworth deleting things that haven't been tagged, he rarely does this. The issue is that when someone else tags something incorrectly he deletes anyway. Thryduulf (talk) 12:29, 18 January 2020 (UTC)[reply]

This doesn't really address the problem, as Thryduulf said. RHaworth already does tag articles that have not been tagged, rather than deleting them. It's an inefficient use of administrative overhead to apply sanctions that don't address a problem. Ivanvector (^Talk/_Edits) 15:37, 20 January 2020 (UTC)[reply]

I oppose this. In addition to what Thryduulf and Ivanvector have said, a "second pair of eyes" is not always one that actually understands how deletion on Wikipedia works. I don't think I even need to provide examples of incorrect CSD taggings by newbies. Glades12 (talk) 18:29, 20 January 2020 (UTC)[reply]

Oppose There's no evidence this is an issue. If it is, provide/cite such evidence...to do so, you'll need to petition to re-open the Evidence page. Buffs (talk) 23:08, 20 January 2020 (UTC)[reply]

Oppose this This would not be particularly helpful. Many, perhaps most, of the more egregious speedy deletions listed in nthe evidence 9and others that I should have added to the evidence) were apparent rubberstamping of bad tags, not single-handed deletions. The problem was the lack of independent verification tha the speedy criteria did in fact apply, and the whole BRD-based idea that bad or IAR deletions don't matter, because someone else can undo them. DES ^(talk)_{DESiegel Contribs} 00:54, 22 January 2020 (UTC)[reply]

Proposed principles

Template

1) {text of Proposed principle}

Comment by Arbitrators:

Comment by parties:

Comment by others:

Template

2) {text of Proposed principle}

Comment by Arbitrators:

Comment by parties:

Comment by others:

Proposed findings of fact

Template

1) {text of proposed finding of fact}

Comment by Arbitrators:

Comment by parties:

Comment by others:

Template

2) {text of proposed finding of fact}

Comment by Arbitrators:

Comment by parties:

Comment by others:

Proposed remedies

Note: All remedies that refer to a period of time, for example to a ban of X months or a revert parole of Y months, are to run concurrently unless otherwise stated.

Template

1) {text of proposed remedy}

Comment by Arbitrators:

Comment by parties:

Comment by others:

Template

2) {text of proposed remedy}

Comment by Arbitrators:

Comment by parties:

Comment by others:

Proposed enforcement

Template

1) {text of proposed enforcement}

Comment by Arbitrators:

Comment by parties:

Comment by others:

Template

2) {text of proposed enforcement}

Comment by Arbitrators:

Comment by parties:

Comment by others:

Place here items of evidence (with diffs) and detailed analysis

Response to evidence presented by Thryduulf

As Thryduulf must be aware, deleting clearly unencyclopedic drafts tagged G11 is an activity engaged in by nearly all admins who work on the speedy deletion of drafts. I don't argue that it is in process, but it seems to fall under the spirit of IAR. Espresso Addict (talk) 00:38, 7 January 2020 (UTC)[reply]

Comment by Arbitrators:

Comment by parties:

Comment by others:

IAR cannot apply to speedy deletion by very definition. The speedy deletion criteria are the only occasions where an admin may delete something without explicit consensus to do so. IAR is only for actions where the outcome uncontroversially improves the encyclopaedia. Every deletion that is done outside of the process is inherently controversial. If other admins are deleting these sorts of drafts as G11 then they need to stop doing so immediately because they are abusing the tools. Thryduulf (talk) 08:48, 7 January 2020 (UTC)[reply]

Agree with Thryduulf. IAR should not be applied to CSD, otherwise the criteria would not be so rigidly defined. Boing! said Zebedee (talk) 06:55, 8 January 2020 (UTC)[reply]

Also, IAR is intended for uncommon situations that were not envisaged by the rule you are ignoring. If you ever find yourself having (or even wanting) to ignore a rule on a regular basis then either you or the rule is wrong and one or the other needs changing. Thryduulf (talk) 14:33, 8 January 2020 (UTC)[reply]

Yes, exactly. Boing! said Zebedee (talk) 14:51, 8 January 2020 (UTC)[reply]

I've always had the impression that rules do get bent at CSD not infrequently, but the deletion log would have be scrutinized to be more certain (which is a mammoth task). I would argue there's a difference between making up reasons for deletion on the spot versus bending the rules. If we were to pass a principle along the lines of, "all summary deletion outside of a strict reading of CSD is bad", and it was consistently respected, my gut feeling is that XfD and PROD would probably be overwhelmed. Maxim_(talk) 18:32, 8 January 2020 (UTC)[reply]

Only briefly, as the CSD criteria would be adjusted to cover the types of things everybody agrees should always be deleted, but this will not happen without evidence that a new/expanded criterion is actually needed (see

WP:NEWCSD point 3). Thryduulf (talk) 21:46, 8 January 2020 (UTC)[reply

]

It's inevitable that some judgment will happen in edge cases where some might think an article fits a criterion and others won't (though in my view, if it's an edge case then it should be rejected as not unambiguously a CSD, but that just raises the question of where the edge is). The obvious example is A7, where what constitutes a claim of importance can be debatable (though, as far as I understand it, A7 is intended to deal quickly with things like "So-and-so is the prettiest girl in my class", articles about people's amateur bands, etc). But it's worse than just a bit of rule bending at the margins. In my time I saw admins frequently totally ignoring the CSD criteria and deleting pretty much indiscriminately (sometimes with near automated speed), and that's mainly why I gave up working CSD - it's disheartening to work diligently on something trusted to us by the community (with carefully worked out rules) while watching others trampling all over it through either gross incompetence or "I know better" arrogance. In this case, for example, we've seen A7 applied to categories of articles where it explicitly does not apply, and A-series criteria applied outside article space where again they explicitly do not apply. As Thryduulf suggests, if the criteria need expanding, that should be by community consensus and not by IAR. And if this case acts as a reminder to admins that they don't have the right to abuse the CSD criteria, and perhaps lead to community-based review of them if necessary, then I think that would be a good thing. As an aside, I used to reject a lot of CSD requests and then watchlist them for a while, and very few went to PROD or XFD (at least in the time I was watching). Boing! said Zebedee (talk) 04:33, 9 January 2020 (UTC)[reply]

I don't delete these, nor the very similar case of unencylopedic drafts tagged U5. I'll readily admit to declining to untag them, however, and leaving them to be deleted by less meticulous admins - and in particular, admins willing to deal with upset newbies without having the recourse of being able to point at the letter of policy. —

Cryptic 16:38, 8 January 2020 (UTC).[reply

]

I'm commenting on what I have observed to happen in the field. Arbitrators who do not work in deletion should be aware that out-of-process deletions of clearly unencyclopedic drafts are not unusual. Thryduulf is undoubtedly technically correct to say that IAR should not apply to common actions, but when faced with a draft along the lines of "EA is cool, LOL!" I am not sure how the good of the encyclopedia is served by declining a deletion request on it. The content is useless; the creator is unlikely to learn how to produce encyclopedic content; and the draft system is already so clogged up that decent abandoned drafts on notable topics must be swept out unread with the dross. I am certainly guilty of failing to decline them. Whether or not we need a new speedy criterion (I have argued strongly that we do) is not a topic for this forum. Espresso Addict (talk) 23:42, 9 January 2020 (UTC)[reply]

The benefit to the encyclopaedia comes from administrators not abusing their tools. We are given the ability to delete (almost) any page at any time, but we have the right to do so only in strictly limited circumstances. If it is acceptable for you to delete a page is "clearly unencyclopaedic" without reference to either consensus or speedy deletion criterion then why would it not be acceptable for me to speedy delete Abu al-Jud, Draft:Garchomp or User:Espresso Addict as unencyclopaedic? None of them meet any CSD criteria either. Thryduulf (talk) 23:54, 9 January 2020 (UTC)[reply]

Well, to take your rhetorical question seriously, Abu al-Jud is clearly encyclopedic; Draft:Garchomp is possibly encyclopedic (I don't follow the notability of Pokemon characters) and certainly could be redirected to a list; and User:Espresso Addict is a good-faith attempt of a long-term, if intermittent, good-faith, if imperfect, contributor to document their actions, interests & responsibilities. And I assume this is a rhetorical "you"? For clarity, I don't recall whether I've personally ever deleted such drafts. I used to get a lot of abuse for declining A7s; I've become tougher over the years at ignoring it. Espresso Addict (talk) 00:10, 10 January 2020 (UTC)[reply]

I don't find the question to be rhetorical at all. If I were an admin, why not

WP:IAR

and delete these pages because I feel that the rules are getting in the way of improving the encyclopedia? The problem here is that IAR is intended to be the rare exception. I could see it legitimately potentially applying to any ONE of those pages (only in the sense that it's a possibility; to be clear, I'm not advocating it). If that were done once and the community said "nah", as an admin, I'd say "ok" and move on.

The two problems with this line of reasoning is that speedy deletions, by definition, should never fall under IAR and that they happen so frequently with RHaworth. First these rules are designed to back up admins and give them the framework to rapidly remove nonsense/unhelpful additions via deletion when no knowledgable reasonable person would contest such a removal. If it would be contested, it should be nominated for deletion. Ergo, if it doesn't fit the rules of XfD, it should be nominated for deletion. When the community gave Admins this authority, it was with that understanding. By applying IAR, an admin is going against consensus. I'm equally puzzled by justification of

WP:BRD

. By definition, BRD doesn't apply to administrative actions that cannot be done.

Second is the CLEAR frequency of such an application AND repeated errors. "I'm sorry. My mistake. I'll learn" for 4 years with no apparent changes in behavior leads me to believe that none of those two-word statements are actually true. They are there to placate the community. Buffs (talk) 18:02, 13 January 2020 (UTC)[reply]

Third, I have yet to find an attempt by RHaworth to change the rules. If he'd said "I think that guideline/policy is wrong" and tried to change it, I'd be much more sympathetic to such justification. Such an attempt would at least show that he's trying to work within the system to make an improvement. Given that I cannot find such an attempt, I see no other conclusion than he's decided the rules broadly don't apply to him (especially

WP:CIVIL)...a far more dangerous problem for an admin. Buffs (talk) 16:44, 17 January 2020 (UTC)[reply

]

• I want to say that I fully agree with Thryduulf's point above. Deletion is not like normal editing, and BRD cannot be applied in the same way that it can with normal editing. Therefore deletion is only legitimate when there is consensus. Either consensus at an XfD discussion, consensus by non-response to a PROD, or advance consensus via the CSD. But that advance consensus is explicitly limited to the stated criteria, construed strictly. Thus an IAR speedy is never valid. When I see a page tagged for IAR that does not fit any criterion, I decline it. I find that I decline something close to half of all tagged A7s I see, and more than half of all U5s. And when I notice that some other admin has deleted something that IMO clearly does not fit any of the criteria, I will ask that admin to restore, and if the admin does not do so, I will often take the matter to DRV, where invalid speedys are often overturned, although not always. Some speedys take longer than others, but I could never do the kind of review I think is required and delete at the rates it has been said in the evidence that RHaworth has achieved in deletion. Now maybe he can review thoroughly at that pace, but I am doubtful. Also, when I decline a speedy I normally notify the tagger and apply {{oldcsd}} to the talk page of the relevant article of page. Thus each decline requires three edits at least, and some copying of content. Again, this means I can't go though Category:CSD at the pace some achieve, and I have done more deletions than some admins, although nowhere near as many as others. including RHaworth. But I do sometimes completely clear the category. DES ^(talk)_{DESiegel Contribs} 02:57, 10 January 2020 (UTC)[reply]
  • I just want to chime in with an agreement that CSD is a horrible place for IAR. If there is a large class of articles that commonly get deleted at CSD that don't meet the criteria, we need to either change the rules or stop doing the deletions. Hobit (talk) 23:58, 15 January 2020 (UTC)[reply]
• See, for example, Draft:Fura Chhamzi Sherpa, deleted G11 by Deb just now. Espresso Addict (talk) 15:15, 17 January 2020 (UTC)[reply]
  • I would have declined that as a speedy deletion. It needed much work copyediting to improve the English and I've not evaluated notability, but that was an attempt at a neutral biography, not exclusively promotional. Thryduulf (talk) 15:22, 17 January 2020 (UTC)[reply]
    • In my opinion, that particular draft had every different hallmark of promotional intent, even if you leave aside the fact that it was apparently created by the subject. I don't delete unless I'm satisfied that there is an intention to promote. Deb (talk) 15:29, 17 January 2020 (UTC)[reply]
      • WP:G11 says nothing about intent to promote. The criterion says: "This applies to pages that are exclusively promotional and would need to be fundamentally rewritten to serve as encyclopaedia articles, rather than advertisements." and "Note: Any article that describes its subject from a neutral point of view does not qualify for this criterion." (bolding in the original). Whether the intent was to promote or not, it was mostly neutrally written and was not an advert." Thryduulf (talk) 15:37, 17 January 2020 (UTC)[reply
        ]
        • I can't agree with that assessment. There was pretty much nothing neutral about it. The article is a brazen advertisement for the creator's business. When I say that I don't delete unless I'm satisfied that there is an intention to promote, I mean that I take this into account, not that it's my sole criterion. If a contributor simply strays over the line because they don't understand the rules, my normal response would be to seek to sort out the wording. In the case of this article, I considered it irretrievable. Frankly, if someone claims to have died and been miraculously brought back to life, without a single citation to back it up, that seems to me to be obvious self-promotion. At the end of the article, she even openly asks for donations. Deb (talk) 16:21, 17 January 2020 (UTC)[reply]
      • I can't see the deleted page so I can offer no opinion on whether it should have been deleted, but I have to agree (and stress) that G11 is not about perceived intent, it is only about content that is unambiguously promotional. Boing! said Zebedee (talk) 13:10, 18 January 2020 (UTC)[reply]
• I also agree with Thryduulf's notion here about BRD and deletion. Unlike a regular edit to a page, a speedy deletion can essentially only immediately be reverted if the admin behind the action agrees to undelete, and that is quite rare for RHaworth. If they disagree, the page has to go through a week-long discussion where consensus may be reached for the deletion to be overturned. In other words, "bold-revert-discuss" turns into "bold-discuss-maybe revert". Glades12 (talk) 08:03, 19 January 2020 (UTC)[reply]

I looked at

AGF best.~ On the other hand, an adequately skilled UPE might create a nice article with the sole attention of promoting the subject and collecting a fee.-- Deepfriedokra

Re:"IMHO, it fairly reeked of promotional content. I could not see a way to fix it without starting over from scratch." If that's the case, then it sounds like a textbook G11.

Re: "Now as to intent, do not know how to assess for intent." Exactly, and for speedy deletion we're not supposed to. Boing! said Zebedee (talk) 09:19, 22 January 2020 (UTC)[reply]

MSGJ: RHaworth sometimes deletes pages too quickly to be able to review them properly

MSGJ notes that RHaworth had deleted 49 pages in a single minute, but they are mostly G13, stale drafts, for which consensus appears to exist to delete if they meet the unedited for six months criterion. The remainder are G8, page dependent on a deleted page, which appear to be talk pages of deleted drafts or redirects to deleted drafts. Espresso Addict (talk) 07:04, 14 January 2020 (UTC)[reply]

Comment by Arbitrators:

justfor comparison, the fastest I've ever gone is 4 a minute, even when dealing with the most obvious sort of material. DGG ( talk ) 22:08, 22 January 2020 (UTC)[reply]

Comment by parties:

Comment by others:

Looking at a single minute in isolation isn't really indicative, what matters is what came before and after. For example if the preceding half hour was spent checking that all the pages met the G13 criteria and identifying the talkpages and redirects to them then batch deleting them in a single minute, followed by another period with no deletions is not evidence of not taking care. What does indicate problems is spending an average of a minute (or less) examining individual nominations over the course of half an hour or longer (see my evidence). Thryduulf (talk) 12:13, 14 January 2020 (UTC)[reply]

I would just comment that it sometimes doesn't take more than a few seconds to recognise an advert. It does often depend on the individual reviewer's reading speed. Although RHaworth often deletes articles that I wouldn't, personally, have deleted, I don't think it follows that his swiftness in doing so necessarily indicates a lack of diligence. Deb (talk) 15:36, 17 January 2020 (UTC)[reply]

When you look at the deletions highlighted by Thryduulf, it seems pretty clear to me that he's not taking the necessary time to review. i.e. Clearing the backlog seems more important to him than doing the job correctly. When you're spending less than a minute over an extended period of time, you're spending most of the time going through the deletion process and none/almost none in reviewing. Buffs (talk) 16:50, 17 January 2020 (UTC)[reply]

It depends a lot on what the page is. One can delete a talk page of a deleted article (G8) in the time it takes to press delete, having -- of course -- previously reviewed its contents before deleting the article. If a lot more than a minute is required to assess any individual page, then perhaps it's not an obvious candidate for speedy deletion. Espresso Addict (talk) 17:25, 17 January 2020 (UTC)[reply]

In some cases it can take more than a minute e.g. to understand the situation, to determine whether a claim is credible, to verify whether the user was evading a block at the time of the edit, whether it is a copyvio of the given source (in the latter case especially if there is limited paraphrasing) but in general if you cannot determine whether a single page meets the criteria within a couple of minutes then in most cases it isn't going to be one of "the most obvious cases" and so not eligible for speedy deletion. In other cases it really can take seconds - talk pages of redirects deleted at RfD are a good example - it's rare for them to have more than just a WikiProject banner. This is not something that can be judged from a single instance, but rather taking into account all the actions of a period of time. Thryduulf (talk) 17:36, 17 January 2020 (UTC)[reply]

Re DGG's speed: I've just checked and I don't average more than 2/minute even for user-requested deletion of user pages with no other edits, and no talk pages to check, which is about the simplest situation I encounter. Espresso Addict (talk) 22:54, 22 January 2020 (UTC)[reply]

On 31 Dec 2019 I deleted some 20 pages, 10 of them within an hour.On 11 Dec 2019 I deleted some 24 pages, 16 within the first hour of that day. on 27 Nov 2019 I deleted some 34 pages, 19 within the course of an hour. I did 20 Deletions on 23 nov 2019. I did 23 deletions on 21 Nov 2019, 18 within 1 hour. I think i am a fairly careful reviewer of pages tagged for CSD, and i am sure I rejected some during those periods. This just a check from the most recent section of my deletion log, I haven't tried to run a query to find my maximum frequency. But 20 an hour seems like something that a careful admin could achieve without being the equivalent of begin banned from deletion. See the log] DES ^(talk)_{DESiegel Contribs} 23:15, 22 January 2020 (UTC)[reply]