Knowledge (XXG) talk:Revision deletion/Archive 4

2632:

need to deal with problems fairly quick and everyone's opinion is the same. I see AN as more of a giant conference table where things are deliberated a bit slower and the issues aren't as urgent. It has always been reserved for ban discussions and like, including things that require the tools. Non-admin don't have access to deleted contribs, for example, but often admin will quote or undelete so non-admin can see and participate. That is why cases about admin accountability should be there as well. ANI can't be gotten rid of, it has taken on the responsibility of WP:WQA, WP:RFCU and the like, and is the best place for behavior problems. Still, AN is better for technical issues and general reviews. They both exist for different reasons. I for incidents, general noticeboard for admin related issues.

1960:, my concerns are similar to Dennis Brown. We don't want a consensus to be built upon what a user recalls seeing. It's possible this could differ from what actually was written. In addition, we have no way of proving that an editor has seen the revision deleted edit(s) short of discussing the text. (Which defeats the purpose of revision deleting a comment anyways.) I'm also afraid that we run the risk of editors "piling on" in support of another fellow contributor, even if they have not reviewed the edit. Finally, this would encourage editors to "be on the lookout" for the contents of revision deleted content before it is removed from public view just so they can be part of the discussion. 4295:

them. I can imagine some scenarios where using that expression could be considered grossly offensive, but I don't think this is one of them. Is it civilised? probably not. Is it universally non-offensive? definitely not. But grossly offensive? hmmm not really. I would add, in some other cases your action might not have generated any complaints, and it would be a stretch to describe this as 'abuse of the tool'. Wrong and not typical, maybe, but not abuse. Can this be improved by clarifying policy? probably not. I suppose we could inject the word 'profanity' into " ordinary incivility, personal attacks or conduct accusations", but I think that's already covered by the current wording. --

3015:"Not as serious" is not the same thing as "equal to a clean block log." The imposition of even a 1-percent de-weighting of my reputation as a Wikipedian is 1-percent more than I deserve for an error of someone else's design. The possibility that I might ever be forced to dig-up diffs or pound-out explanations, when my block log is used as a cudgel in discussion to delegitimize my opinion (as I've seen it done in other cases), is an encumbrance I should not have to bear for someone else's mistake. The fact that this is infinitely permanent - it will be here 2, 5, 10, 30 years from now - compounds each of these molestations by a factor of 2, 5, 10, or 30 times. 5715:

practical difficulties: e.g. in cleaning up after editors later found to have been disruptive, or when tracing the history of an article, or for figuring out why something in the article is the way it is (revdelling removes the link between the edit summaries and the changes made). It can also impede actual attribution for licensing purposes: yes, you can still generate a brute list of everyone who's edited the page, but you can no longer have just the actual list of contributors (if you reuse content after revdel, you'll have to credit everyone from the revdelled part of the history, including vandals and otherwise disruptive users). –

807:. (Note: I probably wrote that text. :)) Attribution is fully satisfied by the list, and rev deletion is even better than the old school method of selective deletion, as it is a simple matter if it is needed to track down a specific bit of text to temporarily remove the rev deletion for investigation. Most importantly, I have more than once found copyrighted content restored to articles after the fact either purposefully or accidentally. Rev deletion is a barrier to this. It would be hard for me to find the link (needle in haystack hard), but several years back I came in to clean a copyright issue and accidentally restored an 1982:. I would like to say yes, that the view of any editor would count, but there are two points against it. First, if you can't see the deleted text, and other deleted text that might be relevant, you're not able to reach a decision. Trying to judge the issue based on memory and without context might not always work. Second, opening the discussion up might mean that those who had posted the text in the first place, or who supported it being posted, could join together to try to force the undeletion. Undeletions and discussions draw even more attention to issues that are often very sensitive. 4322:) that he wasn't concerned with the substance of my edit summary commentary but only used the tool to suppress my use of profanity. I wouldn't describe that as abuse either as it was obviously in good faith, but it was not appropriate. The policy gives leeway to admins to use discretion and err on the side of caution, but censoring profanity is so far outside the scope of how this tool is permitted to be used as to suggest incompetence. I mean, we have repeatedly not got consensus that editors saying "fuck" on Knowledge (XXG) is actionable 2510:

immediately affected by the action (as in, say, a bad block). Challenges of admin actions with no consensus to overturn generally leave them as implemented. As for the particular case that prompted this RfC, I am not convinced the project's biggest dramaboard is a good venue in which to review deletion of sensitive material, for the same reason it's a bad venue for requesting revdel in the first place. I'd probably care less about this if the review happened in a place less well-known for attracting low-quality drama-mongering commentary.

1803:. Characterizing an edit without repeating its content is most certainly not beyond serious editors. The biggest harm would be self-inflicted, editors operating on a faulty memory may mostly damage their own credibility. The harm of repeating deleted material is less in discussion space than in article space, and its likely to be redacted again very shortly. Balancing this potential harm against creating an artificial boundary between editors and admins, where typically DRV has none, I find myself in favour of unrestricted discussion. 297:

reference to a revdel admin, leaves a big loophole for privacy. In addition, when revdel admins have the latitude to refuse requests without proper explanation, they are unaccountable. Perhaps privacy just isn taken seriously or perhaps it's assumed that with sufficient escalation to oversighters or stewards, the issue will be resolved. It is not a reliable system at this point. In fact the more you seek assistance for something like a log deletion, the more you have to reveal private information, such as ip addresses or emails.

1855:. If nothing else, this should be a case of IAR: the intent of the policy is that only admins can comment because only admins can know what was deleted. But that assumption is not true. Following the letter of the policy violates the intent of the policy. Of course, now that we've found the hole in the policy, we should fix it, not limit ourselves to using IAR. I am also astonished at how many of the "no" !votes are claiming that normal users shouldn't comment because they don't have access, when in this case they 2864:

about the potential of abuse through point (c). If a log is completely redacted, no one but administrators will be able to see the blocking admin or the blocking rationale. Hiding this from public view would decrease transparency, which is something that the community has generally discouraged. In addition, I believe it's misunderstood that that log will "disappear". It will only be redacted, which could create more issues than it solves. (e.g. Editors trying to figure out what was redacted and why.)

2821:- my rationale for this amendment is that I was recently subject to an erroneous block. Prior to this I had a clean block log. My block log is now permanently marked; there is no way to clean, and no way for me to undo, some else's mistake. Even in serious criminal matters it is possible to have an arrest record sealed if the arrest was made in error. It is incomprehensible that a block log error should haunt an otherwise capable Wikipedian for, essentially, eternity. Our own, existing, policies ( 2422:*"Grossly insulting, degrading, or offensive material that has little/no encyclopedic or project value and/or violates our biographies of living people policy. This includes slurs, smears, and grossly offensive material of little or no encyclopedic value, but not mere factual statements, and not "ordinary" incivility, personal attacks or conduct accusations. When attack pages or pages with grossly improper titles are deleted, the page names may also be removed from the delete and page move logs." 268:

there's only a very small remaining number of people to consider - admins who happen to be paying attention to the deletion log. If it's really sensitive enough not to mention it at all, I would try and get it Oversighted directly as quick as possible without intervening RevDel. As Oversight is explicitly stated to be applicable to IP reveals, it does seem contradictory that such a summary is suggested for RevDel (which also implies, to me, that just using RevDel is enough in those cases.) —

1884:

that, and allowing them to would defeat the purpose of Rev Del. This doesn't mean that an editor that did see it can't opine, but if you open it up, then some people are demanding it be undeleted for review (no chance that is going to happen), so the community can decide. Or you get piling on by people who simply claim to have seen it, or are basing their opinions on what other people in the discussion are saying. Rev Del is a very sensitive tool that the WMF legal team

4184:

introduced. If you want a valid policy reason call it RD5/G6: technical housekeeping. I'm fine with also calling it IAR, but that is what having ambiguous policy lines exists for: to allow discretion in using it and to give a policy basis for IAR. Like I said, I'm not particularly sure it's something I would have done in this case, but I also don't see it as a big deal by any means, and I don't see a need to update the policy or make a case of it at ANI.

31: 3395:

deleting in copyvio situations. Possibly it was intended to modify the conditions for use of this item, but as written it does not. Also, if it is going to be changed, why not also fix the last phrase which is also logically mixed up. This whole item provides one authorizing-reason. What the heck does "take precedence over" an authorizing-reason mean? It probably meant taking precedence over the "conditions" part of this item.

4081: 2839:-and yet we prohibit editors accidentally blocked from seeking any alleviation from "punishment and humiliation"; the only amelioration is (sincere) condolences. Adopting this amendment would still allow admins to view the unaltered log, it would simply obfuscate blocks made as a result of errors (covering everything from the cat walked across an admin's keyboard, to whatever) from the public-at-large. 2427:*"Purely disruptive material that is of little or no relevance or merit to the project. This includes allegations, harassment, grossly inappropriate threats or attacks, browser-crashing or malicious HTML or CSS, shock pages, phishing pages, known virus proliferating pages, and links to web pages that disparage or threaten some person or entity and serve no other valid purpose, but not mere spam links." 3785:

violations that are so bad that the public shouldn't see it.RD3 focuses on disruption to the project where the revision still being visible, even if reverted, would cause disruption by still allowing people to view it. Probably the most frequent area that I use it in is harassment of other users, but also includes things such as malicious links, personal attacks, and threats of violence.

1934:

really have a clue. Anyone that knows me knows I'm extremely outspoken about non-admin having equal say and have supported that for years, but if the paper misquotes, gives improper context, etc. then you have a bad situation. There is no substitute for direct access, so this is about a technical limitation, not about trust of someone who can actually see it.

3806:

way that normal delete is used, but only one revision obviously. Or can it only hide revision information so users without (deleterevision) cannot view it? Sorry if it's hard to understand what I'm asking. I just wanted clarification on whether Revision delete can be used to delete a singular revision more or less the way normal delete deletes all revisions. ―

3206:. Many users understand what to do, but some user do not wwish to fufill such request, since another user may have edited the article while the bad content was still there, and that there attribution needs to be kept. The fact is that only the username needs to be attributed, not the diff of the edit. Therefore, I ask for a clearer wording of 409: 2794:

intentional or negligent misuse of the admin toolset, (b) a consensus at ANI has determined was a “bad block,” or, (c) the blocking admin has acknowledged the block was made in error or as a result of some misunderstanding by her/him. Immediately after RevisionDeleting any portion of a block log, an administrator must seek Oversighter review.

2059:

disagree. There are certainly some editors who would attempt to contribute to a discussion when they actually have no idea what's been RevDel'd, and it would take some effort from a closer to determine whether a position was informed. I think a decent closer could definitely do that, though, which I think is where OR and I will disagree. ~

1665:"Actions performed using this tool remain visible in the public logs. They are subject to review by other administrators (who can see redacted material), and to reversal upon clear, wider consensus. As with other administrative tools, good judgment and appropriate use are expected; improper use can lead to sanctions or desysopping." 2787:. Due to its potential, use of the RevisionDelete tool to redact block logs (whether the block log entry is justified or not) or to hide unfavorable actions, posts and/or criticisms, in a manner not covered by these criteria or without the required consensus or Arbcom agreement, will usually be treated as abuse of the tool. 3318:

exercise their discretion whether to carry out the deletion (this proposal doesn't change that, as far as I can tell). But Josve05a is correct that legally, all that's necessary for attribution is the username, and the removal of who-did-what is sometimes necessary to more thoroughly delete copyright infringing content.

5080:

not be a big problem, I could see potentially the Bond rights holders getting upset at WP for even having the plot in the edit history. Is a revdel appropriate to remove that content under the idea of a copyvio issue? It's not "blatant" as I'd read the term but would fall into the intent of why we'd want to hide it. --

853:. If there's a history of, or a significant likelihood of, an editor edit warring to restore copyvio material then revdel would be a reasonable per MRG. Wiki admin time is a limited resource -- there are 117,190 active users and 851 admins -- so non-admin user based solutions are preferable. At the time of this writing 874:

had been removed. I don't believe that rev deletion should be mandatory by any means or that it should be used in every circumstance, but I do believe it's a good idea when the content is substantial (increasing the risks of restoration) or when the removal is trivial (impacting only an edit or two of no substance). --

1489:

Oversight should still review. In other words, just a bit more friendly encouragement or advice that any revdel to remove personal information, regardless of its severity or lack thereof, should be reported to Oversight just to have a final review and that there is no harm/wrong aspect to doing this step. --

4799:- Agree with Thryduulf and you all. This RfC is malformed and should be closed. The comments so far are unanimous in expressing various amounts of incredulity. RevDel is a topic that needs community discussion and consensus prior to a RfC, if any. Personally, I don’t see a need for change in this area. 4739:

been exhausted? Please describe why you think this RfC is necessary: for instance, has an admin been using revdel improperly? Are the present revdel criteria inadequate to cover a recent situation? I know of no circumstances when it is permissible to use revdel outside of the criteria already agreed.

4304:

I do not believe this was an appropriate use of RD2. Revision deletion is to be used sparingly, due to the ability to hide what was done by whom and with what reason. Page deletion is comparatively less intrusive. The bar for RD2, as I've understood it, hinges on the word "grossly". If any reasonable

4229:

Posting just a non-disruptive spam link in an edit summary seems less clear to me. I get spam like that on my blog... unfortunately we don't have a way to clean up the history page (hiding edit summaries from the default history-view while still giving readers a way to see them if they want) the way

3842:

Revision delete can't do that, if I were to revision delete everything in a revision (content, editor and summary) all three would be crossed out. However, "normal" deletion can be used to do this by deleting the whole page then restoring all revisions except the target one. To a non-admin editor the

3551:

No, it doesn't change anything, just the wording of the criteria, since currently it sounds as though you can't do revdels if there are multiple editors involved (multiple revs), but since only names need to "stay" for attribution, the part with "If redacting a revision would remove any contributor's

2863:

I'm quite against this, particularly because of part (b). ANI can receive a selective audience based upon when a post is made, who follows the ANI noticeboard, and how quickly a thread is closed. I'm not comfortable with hiding something from public view based upon popular opinion. I'm also concerned

2278:

The default is to leave as status quo, but more importantly, when it comes to "grossly insulting, degrading, or offensive", you err on the side of caution, which is to leave it deleted. The same is true with all BLP, if in doubt, leave it out. Other polices (like BLP) come into play where you can't

905:

That's all very hypothetical. What's real is the confusion of having a section of a talk page history suddenly go gray with a bunch of strikethroughs appearing -- the usual response is confusion and/or anger that an editor's edits have been removed (a misconception, of course). If the user interface

883:

You explained how a copyright violation was accidentally reinserted into an article. There's no evidence of harm -- did WMF have to expend resources because of it? Did the copyright holder suffer any economic loss? What is losstwith a revdel is loss of transparency -- it makes it impossible for most

354:

It is an interesting topic, and one where opinions will vary globally. In the UK, the UK Information Commissioner might well regard such an outing as a data breach, though that is a marginal interpretation. Certainly it is good manners to inform the individual who has been outed. I would go with good

296:

This issue REALLY, REALLY NEEDS MORE ACCOUNTABILITY. On the presumption that privacy is somehow important, the fact that admins can discover information as requrests are made, and that oversight is not recommended except for total deletion, while revdel and log deletion are gray areas recommended for

5154:

I would not expect any less from Masem. Respectfully disagree with them though. Information about a movie being known before its release doesn't directly impact its commercial line. Case in point Avengers: Endgame became the highest grossing film of all time despite all the information that came out

4700:

The introductory statement in "Criteria for redaction" that currently reads "RevisionDelete should not be used without prior clear consensus for "ordinary" incivility, attacks, or claims of editorial misconduct." is amended to read "RevisionDelete should not be used without prior clear consensus for

4669:

4618:

4452:

It is disheartening that language like this is considered acceptable. We're supposed to be working towards inclusiveness and a non-hostile environment. Use of aggressive and/or profane language is counter to those goals. That being said, it's clear that my revdel of this comment was beyond what's

4129:

I'd say that RD6 ("non-contentious housekeeping") is completely wrong for this purpose. If it's to be done in a case like this, it's done with IAR. So-called "breaking the interface" is not a particularly sound reason for deletion, especially if it contains encyclopaedic content, and I wouldn't want

2984:

I have never done a RevDel on a log entry, but when one is done on a normal article revision, it does not diappear from the history. It is still apparent that there was an entry, but just what it was is hidden. In a block log, that would be more alarming than a block followed by an unblock with edit

2631:

Many decisions at ANI can be made without an admin talking or acting, which is not the case at AN. It is reserved for issues that are primarily admin centered or need the tools. This includes announcements and the like that ANI doesn't get. I think of ANI as more of a fire/police station where we

2314:

If there is no consensus to change it, it can remain. If one admin is passionate about reversing, they can reverse it and then a discussion can ensue. A "no consensus" on that would leave it as the second admin reverted. It is like a speedy deletion followed by a deletion review ———— a no consensus

2058:

I disagree with her on this issue, but I would consider OR one of the strongest advocates of non-admins being able to contribute as much as technically possible. There is a regrettable class divide on Knowledge (XXG), but OR is part of the solution, not the problem. Her criticism is valid, even if I

2036:

The problem is that there's no way to verify a non-admin's memory of a deleted revision or ensure they've seen all the relevant material without discussing it in enough detail to defeat the purpose of the revdel. DRV can solve this by temporarily restoring deleted content if necessary, but obviously

1933:

And I don't mind making exceptions, and will always listen to non-admin familiar anyway (as will most admin), but I think you need to keep the policy limited to those with full access. It is a can of worms that can lead to piling on by people who are just piling on to defend someone when they don't

1479:

Recently, there was an IP editor that had signed their real name and then promptly erased it, which I took as a likely indication they didn't want that revealed, so I RevDel'd it, but in reading this page I was unsure if I should report that to Oversight. I did as at AN and they said I should, and I

311:

What is more, it is highly inefficient and a strain on wikipedia admin resources. Suppose an editor is an inattentive logger-out and frequently mistakenly posts while logged out. Are they to bother and badger admins to constantly fix their privacy accidents? There should be a forum on reforming this

5684:

on the versions that contain the copyright infringement to help avoid inadvertent restoration in the future if the copyrighted content is extensive. Otherwise, so long as the infringing text is removed from the public face of the article, it may not need to be removed/deleted permanently unless the

5648:

When cleaning up copyright violations, it's standard practice to revdel the portion of the history that's between the addition of the copyvio and its removal. But how about when there's a fair number of intervening edits? Should they be revdelled, even if there's many of them and they have resulted

4780:

I've just been summoned here by bot and I'm also struggling to understand why this is an RFC - where is the context? Where is the prior discussion? What is the motivation? Why is change needed? Why is there not an explicit option for no change (the only thing I can support given the lack of answers

3805:

I know that Revision delete can show and hide the contents of the revision, the edit summary of the revision, the name of the user/IP address that made the edit, or any combination of those three. But I wondered whether it would be possible to normally delete a single revision, pretty much the same

2999:

If I looked at a block log and find that reason, blocker etc were crossed off due to log entry redaction, that would still look suspicious, and would raise questions that are not answered. A block log having entries is not as serious as it sounds above. In the case mentioned above, it is more about

2378:

defaulting to reversal of deletion. As stated in the Misuse section of the policy, "Material must be grossly offensive, with little likelihood of significant dissent about its removal. Otherwise it should not be removed." This makes it fairly clear that RevDel is not appropriate where there will be

950:

restored some content that had been selectively deleted to remove a copyright violation - what we used to do routinely. With revision deletion, people can at least see that the content has been removed and where, instead of being confronted with a strange gap in the article's history and the sudden

358:

A sub question is regarding a user who uses a reasonably clear interpretation of their own name. I do this and am easy to find with Google. In such a case how does one define outing me and similar editors, and what level of posted detail would be the trigger for you to consider informing me by (eg)

267:

I was on my phone earlier so couldn't read into the matter easily, but I see now what you mean. I agree that it's a confusing gray area. Obviously the intent is to not draw attention to the sensitive material; but considering that you're already getting the material out of the public eye by RevDel,

5079:

has had the same IP editor add and remove a plot summary - but the film is not set to have a premiere for a month so either this IP is shooting the breeze or they actually actually revealing the plot through an inappropriate means (pirated a copy or broke an NDA, etc.) While other films this would

4758:

This also feels like instruction creep. It's implicit that ordinary vandalism does not require revdel. Conversely, any vandalism that requires revdel is not ordinary: there is some other factor in play, such as a gross BLP violation or personal attack, that clearly indicates revdel is appropriate.

4183:

We did this when the long edit summaries were first introduced and people were messing around with them. I wouldn't have done it myself, but I don't think it's bad to have been done. As a whole, the community has gotten a lot more open to revision deletion as a tool than when the feature was first

3394:

Maybe there is a good idea there, but as written this furthers logical scrambling. The whole subject text is to provide one authorizing-reason for deletion, and place conditions on use of that particular reason. The revised text takes out that "condition" wording and inserts some general info on

1883:

This is one of the very rare instances I think adminship is needed to be a part of consensus. If you don't have access to the deleted material, then you don't have a basis for the opinion. Admin can not only "remember seeing it", they can go and review it while it is deleted. Non-admin can't do

1584:

The oversight team has a well-staffed mailing list and OTRS queue with round-the-clock coverage; for this kind of solution to work we would need the same kind of arrangement for revDel-willing administrators. I'd worry it would create Yet Another Backlog. Creation of an administrators mailing list

1507:

As an additional comment, the reply I got back from the OS committee on my RevDel report was very friendly and encouraging about keeping vigiliant on even matters like this one, and that's the type of language I'm not quite seeing on this page in terms of using RevDel in a situation where personal

1415:

When I originally created my account, I included my full name on my user page. This of course is still accessible through the history. I would prefer that my full name not be associated with my Knowledge (XXG) account. Would it be possible to have older versions of my user page removed? Thank you.

873:

I just explained explicitly how harm came from having a copyvio in the edit history. I put it back. I did that - myself - because I didn't know it was there. This was not an edit warring situation. I wasn't fighting anybody. I rolled back the article to an earlier state and restored a copyvio that

131:

In fact, the time and date is logged in the deletion log, including the username of the admin who performed it. Suppression / Oversight deletions do not display the name of the oversighter. If you click on a page's history then "View all logs", you should have the information about who performed a

5714:

My own view is that the current wording is appropriate. Attribution need not be understood in the narrowest terms. After revdel, it's no longer possible to see who wrote what for the text that's been added in the meantime. Such concealment of the provenance of article text can lead to a number of

4443:

Agree with Ivan. I would even go as far as to say that the term "fucked up" is potentially closer to "messed up" than to the term "fuck". I mean, saying "America is fucked up" is obviously and totally different from "fuck you America", right? Anyway, I don't think this needed to be revdelled. And

4280:

because I don't want to escalate this into enforcement (although, I am disappointed that Floquenbeam saw fit to do a run-around of my RD2 by copying the text of the edit comment into the talk page). I'm just looking for clarity as to what's considered acceptable in an edit comment, and when it's

3317:

as a harmless codification of current practice. Even if there are intermediary edits, the problem is that even those intermediary revisions still contain copyright infringing content that needs to be removed. It sometimes comes at the cost of eliminating who-did-what, and administrators can still

2497:

method. BRD defaults to the status quo of whatever existed before the bold action in the event of no consensus. Here, we have bold, discuss, no action, which defaults to the bold use of RevDel. Is it really the status quo for a single editor to use RevDel unilaterally, with a clearly demonstrated

2260:

just about everywhere, "no consensus" means "default to status quo". There's no reason to make this particular, unusual situation an exception. If anything, it's the opposite; if a reasonable subset of admins reviewing the situation think something is "grossly insulting, degrading, or offensive",

4294:

It seems to me the issue is whether saying that something is "fucked" is grossly offensive, or just offensive (or perhaps neither). If someone said that they fucked up an edit, would you be offended, or grossly offended? I would view it as several things, but grossly offensive wouldn't be one of

2208:

makes no reference to revision deletion, and it's a bit of a stretch to say that we're required not only to remove violating content but to revision delete it. The criteria for revision deletion requires material to be "grossly insulting, degrading, or offensive", which does not include most BLP

2070:

My usual chapter-and-verse on this issue is that adminship is essentially a technical role - you have more buttons on your interface - and it should provide no advantage in making essentially social decisions, like closing a discussion. The current question strikes me as a technical matter; only

1028:

It's plenty transparent, I think. You know what happened, and you know why. The content hasn't disappeared and, as Flatscan notes, the logs are clear. You can't see every edit, but they are easily and quickly evaluated by any admin and restored easily if needed. Selective deletion offers none of

3784:

Borrowing this from my OS application, but this is how I explained it then with a few tweaks: RD2 focuses on content that could be grossly insulting or degrading or is a gross BLP violation. The focus is on the content that is objectively offensive to the point of causing harm to persons or BLP

2920:

I can see some argument in support of removing total errors, such as mis-clicks and mis-identification, but also think they are usually cancelled out by the unblock. However negligence and intentional misuse are things which shouldn't usually be hidden (I'm thinking here about scrutiny of admin

2875:

This is more than sufficient to allow the correction of an error. Any reasonable editor should be able to review the block log and see that the block was made in error or was overturned. Finally, I'm also not sure why you are recommending the oversight team to review the revision deletion since

2299:

The open ended nature of the question appears to apply to RD in general. At the very least RD1 should default to delete, per the precautionary principle. Same with contentious BLP material. In particular for the latter, POV pushers can very easily appear in sufficient numbers on a discussion to

1296:

One type of vandalism I routinely see is where somebody writes that some guy is gay or some girl is a bitch. This is particularly common on articles about schools. It is obvious in these cases that some kid is trying to defame somebody, usually a classmate, teacher, or principle. My question is

5120:

the aspect of commercial impact - in this case the plot summary for a highly-anticipated film not yet out in any type of public release - that that would be appropriate to use RevDel to remove those instances of it. We already use RevDel for outright copyvios (large chunks of copyrighted text,

3525:

I'm not at all sure I follow your line of reasoning there, you seem to be saying that despite there being proposed new wording that changes when this criterion would be used, when this criterion could be used will not actually change. If it isn't a change then... what is it and why are we even

2607:

discussion linked in the OP? As far as I can see, It was remarkably free from drama. I have discovered from some experience that a straight yes/no survey cuts down on drama by 90%. Speaking generally: there is no need to copy the revdeled material into ANI for discussing it: I managed to do it

2509:

BRD is fundamentally about editing and is less applicable to admin actions. It may only be a "wheel war" on the third revert, but it's still impolite at best to unilaterally reverse another admin on grounds of "I disagree" rather than clear evidence of error, especially when no other editor is

1698:

1) In cases where non-administrators know enough about the redacted material to evaluate whether revision deletion is appropriate, can they weigh in on a discussion regarding whether revision deletion was appropriate? More specifically, do their contributions to such a discussion contribute to

1488:

report it to the oversight committee to review. While that is sorta said in the language presently, there's a lot of emphasis on the revealing of personal information as forms of personal attacks or the like, and not for what I'd consider more a mundane mistake as this IP had done, even though

2793:

and an editor who has been blocked in error may have the erroneous block and unblock lines of the block log RevisionDeleted on request. “Blocked in error” does not mean a block that is overturned or shortened on appeal, but, rather, a block that either (a) Arbcom has determined constituted an

4373:

As I said in my linked reply I accept the criticism as to the nature of my edit summary and I'm not looking for it to be restored or for any other admin action here, and I was happy to have just thanked RoySmith for his action and comment and moved on. It's my talk page, I regularly reply to

3734:

Most people don't have the resources or interest to deal with the hassle of running scripts and storing / comparing multiple copies of Knowledge (XXG). So at the moment, the risk of RevDel content being revealed is low. And given 80% of RevDels are removals of obvious libel and plagiarism, a

918:

for the real, non-theoretical impact revdeling has on editors. (That specific oversight, was, unfortunately necessary because it was an outing issue). Couple that with inconsistency in application -- sometimes copyvios revdel'd, sometimes they're not -- makes it appear to be a capricious and

2964:

I'm not out for revenge. I'm out for restorative justice. I would like my clean block log restored; an admin being warned or admonished is irrelevant to me. There's no reason I should, though no fault of my own, have to live with someone's mistake for the next 50 years when there is an easy

2472:

This RfC is not meant to address the specific situation at ANI, which is why I didn't include specific criteria that are relevant to that discussion. This RfC is more broadly trying to determine how appeals should take place - whether non-admins should be involved and whether the default is

5105:

The edits seem to originate from a registered user, so you could maybe discuss it with them. They claim it's an official plot from a private viewing. I don't know what 'official plot' means, but I think you'd have to assume there's at least some assumption of confidence. I'd view this as a

2071:

people with the right interface buttons can see the material. If participation is open to anyone who happened to see it pre-revdel, the effect is that the closer will rely to a greater than usual degree on their personal familiarity with the non-admins to judge their contribution. That's

1528:

seems pretty clear as written -- specifically says it the blue box "It does not matter whether the privacy breaching material was posted by the user themselves" and then "an administrator may provisionally delete the information from public view to minimize harm, then promptly contact an

1888:

you have adminship to access. Admin were specifically selected to be trusted to do this, subject to review by other admin, and accountable for how they use it, all the way up to Arb. Other actions should and are reviewable by the community, but you can't review what you can't see.

4156:

It may have been running off of some screens, requiring scrolling to see it all, but while perhaps annoying and/or not visually appealing, it is still functional. I don't think RD should be used for this, in this policy or IAR. I would think the underlying issue is fixable with CSS

3513:

This has been established practice for over a decade, way before revision deletion was activated, using the much cruder selective deletion method. RevDel made it easier to ensure that the list of contributors would be preserved. This proposal here changes nothing to that practice.

3600:

Could simplify it even further: "Blatant copyright violations. In order to maintain attribution, only the content of the offending revisions should be hidden, not the usernames or IP addresses associated with them. Best practices for copyrighted text removal can be found at ..."

2075:

if the concern is some kind of perceived social inequality. (A relevant question: exactly the same logic could be applied to say that anyone who saw suppressible material should be able to weigh in if the suppression is questioned. What if anything differentiates the two cases?)

248:

The case that concerned me is all fixed - accidentally revealed IP, I rev-delled it using the drop-down reason, and emailed Oversight who dealt with it. But then I re-read the policy page here which says, effectively, "don't use the drop-down reason", and wondered what reason I

2656:

As for the original ANI thread, I find it rather unsatisfying, actually, although not in a way that would really be addressed by a change of venue. I see a conversation among mostly men about an action taken by a female admin to redact material about a female harassment victim.

895:

Having copyvio content in our articles creates potential legal liability for our reusers, potential loss of property to copyright holders, damage to our credibility and wastes the time of editors who polish and build on content that we can't keep. That's plenty of "harm" to me.

4509:

Any log entry should not allude to the fact that the revision deleted material is potentially suppressible, this would invite scrutiny which is to be avoided. You could use RD3, or RD5 (deletion reasons 13 and 14 apply, and perhaps others - often 3 depending on the content).

4670:"ordinary" incivility, attacks, or claims of editorial misconduct. RevisionDelete may be used to redact "ordinary" vandalism at the discretion of administrators, but the author's username should not be redacted unless the author's username violates Knowledge (XXG) policy." 2732:. This proposal asks for a rather significant change to the revision deletion policy that requires a broader range of input than was presented here. If anyone wants to further pursue the proposal, I recommend publicizing a second RfC on central community noticeboards like 5693:

What I take away from these is that revision deletion here is advisable, but – under normal circumstances – not strictly necessary, and that it should be avoided if it's going to get in the way of attribution for the content added by the intervening edits. However, after

5702:

Now, if this recent example is representative of current practices, and it is indeed the case that the wording of RD1 is not relevant anymore, then we'd really need to start thinking about changing the policy and the documentation. There was a proposal to this effect

5533:

Have you considered implementing this with, for example, the "sysop-show" CSS class? You could hide the template from readers, but admins wouldn't have to edit the page to be able to view the template. Or you could pick another user level from those available at

4139:

Well, that language mirrors G6, which would definitely be applicable in cases like this if you were dealing with a page that disrupted the encyclopedia from a technical perspective and wasn't particularly vital (think someone's 1MB user subpage full of binary.)

857:

has entries older than a month waiting for attention, further evidence relying on the admin subcommunity, rather than the much larger editor community, to address copyright problems is not optimal. Additionally, and perhaps more importantly, research has shown

4225:

Unambiguous advertising should be covered by c.5, as a reason to delete. The phrase "but not mere spam links" in c.3 seems to be saying that adding a external link to a page is not purely disruptive. Editors may want a chance to discuss whether the link is

5505:

revision deletions, even oversightable ones, in cooperation with a private "warning/no-save-allowed" edit filter that, when activated, would send an alert to the IRC channel. It would mean oversighters would need visibility to that edit filter's log though.

5115:

I missed that, but I guess what this means more broadly that if those of us with revdel have a reasonably strong concern that, under US fair use law, that some material added by editors that would otherwise tick off all the other boxes for a fair use defense

4118:

where an edit summary was removed because it broke the interface, making it difficult to read on some devices. I'm wondering, should we have a specific allowance for "breaks the interface", or does that just fit comfortably in "non-contentious housekeeping"?

2872:

Very short blocks may be used to record, for example, an apology or acknowledgement of mistake in the block log in the event of a wrongful or accidental block, if the original block has expired. (If it has not, the message may be recorded in the unblocking

1217:

That is rather... circuitous. Since RD5 says "It is important that the underlying reason for deletion be made clear in the log summary," wouldn't it be obvious that the policy being referred to by the reason "RD5 - Valid deletion under policy" is RD4? :-/

3295:

Wording of criteria should be clear to both all parties, without assuming that they know the details of the licensing rules besides a basic understanding of what constitutes permissible/impermissible copying. See my even simpler wording suggestion below.

2935:

I'd suggest that all bad blocks result in a 24-hour block for the blocking admin to ensure creation of a standing record that allows "scrutiny of admin actions" but I can imagine the absolute howls of protest and horrified undulations that would produce.

1379:(thereby outing her), besides blocking the account and redacting the username from every single one of his edits? What if someone insists, for example, on adding a bunch of categories (eg. Sock puppet of such and such user) to the account's userpage? -- 3871:

alone. "Delete and undelete specific revisions of pages (deleterevision)" There was a time where I believed that Revision delete was just a combination of delete and undelete, but only focused on one revision at a time. But that was over a year ago. ―

5698:

it became apparent that revdel is – at least by some – used more extensively. It was also pointed out that revision deletion doesn't hide the names of the editors, so strictly speaking it doesn't get in the way of attribution for licensing purposes.

1564:

To request RevDel, on the other hand, a concerned editor is advised to hunt down an administrator who may be active from a long list of administrators, or to learn how to use IRC. But don't post it on a noticeboard. No, that might be too convenient.

2157:

2) If there is no consensus for or against revision deletion, should the redacted material remain deleted? In other words, is the burden of showing consensus on the administrator who used RevDel or the editor(s) disagreeing with the use of RevDel?

1711:, non-administrators should be able to contribute to such discussions when they can do so from an informed perspective. Adminship is about tools, not a leadership role, and the broader community should have a role in examining the use of tools 5663:

that can be redacted without removing attribution to non-infringing contributors. If redacting a revision would remove any contributor's attribution, this criterion cannot be used. Best practices for copyrighted text removal can be found at

4937:

I don't see why you think this is so trivial. I was under the impression that the community, when originally authorizing revdel, thought it important to prohibit its use for ordinary vandalism. That seems like something we ought to respect.

2602:

as anyone else: however, I have no more confidence in Arbs or Admins as compared to the people who lurk there. For better or worse, that is the place where non-admins weigh in regularly. As to concerns about drama: does anyone object to the

2473:

restoration or removal. In other words, does the RevDeling administrator have the burden of defending their usage of RevDel (rather than other editors having the burden to challenge that usage) and who does he or she have to defend it to? ~

553:

4763:

is called for—and again, that's outside the realm of ordinary. I don't see why any change to the guidelines is necessary; I agree with Redrose64 that the proposer of this RfC has not demonstrated the need for the formal commenting process.

338:

When a Knowledge (XXG) registered user's identity is outed, do we notify them (by email if possible) that it happened, and that it has been revdel'd? IMHO, we should. I wonder if that should be a suggestion on the WP:REVDEL page. Thoughts?

3447:

The assertion that all that is needed is a username in the revision history doesn't add up. If there is more than one edit, then who did what is lost when the revisions are deleted. Simply removing copyvios is usually sufficient anyway.

4886:

I was also wondering about this. I recently noticed the revdel of what I had assumed was "ordinary" vandalism (on wheels). Looking at the examples listed for "RD3: Purely disruptive material", they all seem much more severe than this.

2586:

grossly offensive BLP violations, which is desirable. ANI has a huge traffic including many users who don't care to weigh in on this sort of discussion. It makes little sense to talk about revision deleted material in front of them. ~

951:

disappearance of some of their own edits. Revision deletion is the best of both worlds - it minimizes the potential for real and tangible harm while keeping transparency. That said, consistent standards of application would be nice. --

5430:

There is no good way to get material quickly revision deleted without drawing attention to the problem. Waiting an hour for oversight is probably ok but apart from IRC you can find recently active admins by looking, for example, at

4393:) and this one was more of the same, though perhaps a bit more pointed than some others. I'll refrain from using aggressive language in my edit summaries in the future. That should be the only action that comes out of all this. 2653:-biggest dramaboard ;) If "matters pertaining to sensitive material shouldn't be discussed in the most public and drama-filled venue short of Jimbo's talkpage" doesn't count as a "principled objection", I don't know what would. 960:

You've lost me. Have you seen what a rev del looks like from a not admin account? It's not at all transparent. I still see the harm as hypothetical. Could you provide an actual example? (Six minutes after my previous post this

391:

I see no harm in being bold and adding a note about it to the project pages. "Admin that revdel in relation to persons that can be contacted by email should consider doing so as good manners and at at their discretion." type

3000:

showing a change of mind by the admin. The log entries show that, and the unblock shows that the block was reconsidered. The block was announced on the talk page as well. But a user is permitted to remove that announcement.

1483:

My suggestion is that there should be wording in the oversight section that suggests that even if you made a revdel in doubt if oversight was necessary, you should be encouraged to be proactive to revdel personal material

4478:

There are occasional cases where I find oversight-able material, which in line with both policy and good sense, I RevDel first as a pro tem measure, before hunting down an Oversighter for both review and full oversight.

1297:

whether is specific type of vandalism qualifies for the use of RevDelete. Does it count as "Grossly insulting, degrading, or offensive material" or as "'ordinary' incivility, personal attacks or conduct accusations"?

1628:*Q1: Consensus on whether or not non-administrators can participate in discussions involving RevDel'd content could not be determined because consensus does not exist in favor of or against allowing or disallowing. 1996:. Usually I'm on the opposite side of this fence, but the thing is, I do not trust Wikipedians as a group to keep their traps shut when they don't know what they're talking about. At least with admins you know they 4701:"ordinary" incivility, attacks, or claims of editorial misconduct. RevisionDelete may be used to redact "ordinary" vandalism, including the usernames of authors of vandalism, at the discretion of administrators." 2869:

The portion of the blocking policy that you cited does not apply here. That policy discourages admins from making short duration blocks (e.g. 1 minute) to leave a record of a warning. Right below that it states,

1917:

This is specifically talking about text anyone can see. The text in question was a passage from a Washington Post article. Anyone can read it, once a link is provided (nobody, on any side, objected to the link).

3342:- if someone makes intermediary edits elsewhere in the article between the insertion and deletion of the violating text, those revisions should not be deleted. This change in policy would mean they are deleted. 2224:

for BLP issues, rather consensus is required for undeletion. Further, the criterion you quote (criterion 2) is not the only criterion used for BLP violations. Criteria 3, 4 and 5 are also frequently applicable.

4050:

I was answering with IRC in case anyone else looked at the header :) A template would attract more attention. An OTRS feed is what we have for OS, but I’m not sure it’d be ideal here (too much to clean up.)

1228:

I don't think it would be obvious. Look at a few different admin deletion logs and you will see that adherence to the rules of edit summaries ranges from proper to not so proper... I think it will blend.

219:

I don't know if you're specifically looking for a pre-Oversight reason (that's not a power I have), but I see the quoted reason from the drop-down quite frequently when encountering deleted revisions. —

3111:

I agree with this change. It doesn't broaden existing policy. It doesn't matter if someone is recognisable after two seconds or ten minutes, if they going to be recognised it's going to get deleted. --

5556:

attention from people who watch edit logs, but for super-sensitive things, well, that's what IRC and email are for. I was not aware of the "sysop-show" CSS class. That makes this much easier to do.

2379:"significant dissent" toward its use. It makes perfect sense to follow through by undoing the revision deletion when "significant dissent" is encountered and a "no consensus" outcome is reached. ~ 1786:. The only advantage admins have, in this case, is their access to the information. Just like any other area of editing here, anyone who knows what they're talking about may express their opinion. 94: 859: 5402:". Also if you don't want everyone in the channel (mostly admins, idling not allowed for non-admins) to see the edit(s) then you can just ask "any admin around?" and then PM whoever replies. -- 1679: 4413:, I'm sure RoySmith didn't have bad intent so that's about it. The general sentiment here also makes it apparent that usage of profanity ≠ eligible for RD2. I think we can close this up now. -- 2170:

this policy is actually irrelevant to most contentious cases. The problem is, that a significant number of contentious revision deletions are on BLP grounds and that policy is quite clear that

4854:

Summoned by bot. What is this about? What is trying to be achieved? Why are we being presented with changes to text in such a way that it's a big headache to tell what's changed? What a mess?

1602:

Transparency? What transparency? We're talking about things that should be hidden from the edit history. Currently, editors are recommended to use IRC. Transparency. You really crack me up. —

3076:

a specific target to the right reader. For example a smear could target a person known locally by a nickname or other allusion that no Knowledge (XXG) administrator has heard of, but that is

2037:

that doesn't work if the material is sensitive (and most disputed revdels will be, basically, subjective differences of opinion on how sensitive something is). It's a matter of practicality.

1080:"Consistent standards of application" supports dealing with the copyvios in a manner not requiring sysop tools due to the vastly greater number of non sysop editors available for wiki world. 2831:

Blocks should not be used solely for the purpose of recording warnings or other negative events in a user's block log. The practice, typically involving very short blocks, is often seen as

626:. My understanding is that attribution issues are only created by revdel if the username is hidden, the edit is not, and the edit contains copyrightable information – the case described by 4647:

Summoned by bot, but this RfC could use more context. It sounds like it's asking "should we follow the criteria for redaction or not?" What is the reason here? What defines "ordinary"? —

5491:

This is just a mock-up for now: To use the mock-up version the administrator would have to remove the "quiet" parameter then preview the page to see the revisions that needed deleting.

3462:"Who did what" is not needed for proper attribution according to the license. And no, just removing copyio from 'live verions' is not sufficient, that' why we have this criteria, due to 938:

harm here. If people are upset because they're confused, the best response would seem to be to improve messaging or education so that they're not confused anymore. Beyond that, while we

170:, plus there is no indication which edit each of those four actions refers to. Can someone please explain? Or perhaps this question should be posed anew down at the bottom of this page. 5218:

What's the best off-wiki way to get the attention of logged-in administrators so they can quickly revision-delete pending Oversight in cases that need it without drawing attention (see

2876:

suppression would not be utilized here. If you wanted the oversight team to serve as a "review board" you would need a larger RfC as that would expand the remit of the oversight group.

2173:

the burden of proof is on those who wish to retain, restore, or undelete the disputed material ... If it is to be restored without significant change, consensus must be obtained first.

4923:

Making a fuss about LTAs is very unhelpful. Knowledge (XXG) is not a bureaucracy and we are not going to write down precise rules about anything. Find something else to worry about.

4092: 5400:

What's the best off-wiki way to get the attention of logged-in administrators so they can quickly revision-delete pending Oversight in cases that need it without drawing attention

4956:

is the best course of action. Are you really sure you want to take time to generate further argument around an LTA? Is that a worthwhile contribution? What I said is factual, see

3499:

I don't agree that the benefit of doing this is worth the loss of the revision history if there were multiple, substantive intervening edits, regardless of what the license says.

2921:

actions rather than the blocked editor). I'm absolutely not in favour of "a consensus at ANI has determined was a “bad block", since consensus on community boards, especially at

3867:

Ah, right. I just wanted to confirm whether or not Revision delete was able to do that. Since it's not exactly clear what Revision delete is capable of, if one were to look at

5726: 5552:

The motivation was to avoid having to flag down sysops on their own talk pages or via email, while keeping things as low-key as possible. Obviously, this would still attract

980: 608: 5215:

At least twice in the last month I've emailed Oversight. It took over an hour in both cases. That was okay in those particular instances but in some cases it's too long.

116:

Is the time and date when the revision was deleted is documented, and if so, why isn't the time and the username of the admin which deleted it isn't mentioned next to it?

3769:

or fill a page with obscenities targeted at a specific person, won't that be purely disruptive material too? Or is the criterion's scope larger than I'm understanding?

3686:

The result of Revision deletion/Archive 4 is weird grey lines on history pages that no one can click on. They should have a mouseover added explaining what is going on.

2332:- RevDel is, in some ways, like speedy deletion. Just like if speedy deleted content doesn't get a consensus to be restored it stays deleted, same is true about RevDel. 5673: 1838:

if they know they can contribute, and if their recollection or what they say is wrong, then others that can really confirm, or really know can weigh in and contradict.

801: 3413:

The "take precedence over" i part of the exisitng criteria, and is not suggested to be added or changed. Feel free to tart another RfC, or an "option 2" to remove it.

2246:. Most of these deletions are BLP issues, so the default has to be that they remain deleted unless a consensus of admins emerges that the deletion wasn't appropriate. 1142:. The current phase of the review is examining the committee's procedures concerning advanced permissions (and the appointment and regulation of permissions holders). 5587:

Is there a "Here's how to RevDel" anywhere? This page is so long that I'm having a hard time figuring out just the nuts and bolts of what it does and doesn't do. :D

3540:

I believe the OP wants to codify the existing practice in a different language. I don't believe the proposal really changes anything in a long established practice.

3138:

Support this common-sense improvement that removes an absurdity. If no administrator has heard of the nickname or allusion, how would any of them know whether it's

1057:

Remarkable language, English - words can be rich with all kinds of meaning. :D Transparent also means "Obvious; readily apparent; easy to see or understand." (says

906:

the software presented to non-privileged editors wasn't so horribly misleading, I'd be more supportive of the revdel'ing of copyvios, but it's not. See discussions

734:

Are usernames ever hidden when using RD1? If RD1 and removing attribution rarely overlap, I think that it would be better to reduce the mention in RD1 and make the

4210:

has a lengthy product advertisement as its edit summary. This project page says that RevDel is for "not mere spam". What's the best way to deal with such cases?

884:

editors to verify the content was in fact, "blatant," and, if an accompanying citation was removed, makes it difficult to use the content in a non copyvio manner.

962: 5020: 4731:

Why has this somewhat-complicated RfC been set up without, so far as I can see, any prior discussion of (or even a link to) the incident (and presumably, there

3843:

target revision would have disappeared from the page history, with just the deletion log entries as evidence of what was done. Hope that answers your question?

804:

which at the time of the adoption of that text included instructions for selective deletion of articles and attribution by list on the article's talk page. See

2900:

more or less per Mike V. Also redacting bad blocks would remove from public view evidence that might later need to be used against a malfeasant administrator.

2493:

In response to your support for no action above, I would actually call no action the exception in this situation. Normally, we resolve disagreements using the

597: 81: 69: 64: 59: 5610: 3968:

How do I request a RevDel for any of the other criteria ("Grossly insulting, degrading, or offensive material")? The template only seems to cover CopyVio.

1568:

Why don't you ask one of the oversighters how to set up a mailbox similar to theirs, and make it as easy to request RevDel as it is to request Oversight? —

638:). It is possible that RD1 use may intersect with instances of hidden usernames, but those edits should have been reverted and revdel'd completely already. 5494:

If there is interest, I'll make a general-purpose "revision delete" template for cases where oversight is not needed and where there are no other privacy,

2018:

Are these know-nothings the same people who elected you and elected all the admins? I also notice a rather striking "class divide" among the answers here.

4974:

Could you be more specific? I agree that it's good to follow the principle rather than the letter of policy, but we should try to align them if we can.

1585:

would also be subject to concerns about reducing transparency. Perhaps an RFC to gather opinions on the idea? (The technical implementation is easy.) –

751: 4350:

because of the use of profanity. And even if it were a personal attack, circle back to the "not for removing ordinary incivility" clause. There are

4028:

I can buy the "No Noticeboard" reasoning, but does this not strike as a bonkers way of doing things - not having templates or a merged email set-up?

2582:

Sorry, non-admins was bad word choice. I meant "inexperienced users". In other words, it keeps the visibility down a bit if the discussion involves

676:

I am having difficulty seeing how revision-only revdel can "remove ... attribution". Consider this hypothetical series of edits to a movie article:

3938:

It might have been more appropriate to call it revision hiding. Also it can hide log entries (perhaps a vandal moved a page to a disruptive name).

3228:

Redacting a revision should not remove any contributor's attribution. Usernames and IP addresses should not be hidden, only the revisions themself.

2783:

for grossly improper content, and is not permitted for ordinary matters; the community needs to be able to review users' block logs and other logs

635: 4351: 2184:

is the only consistent option. In truth, of course, the outcome of this part of the RFC is irrelevant as BLP is always going to take precedence.

5014: 4272:

for background). I was surprised when I got pushback to this from several uninvolved editors, so I'm seeking clarification of proper usage of

1762:: Obvious interpretation is obvious. The restriction to administrators is only there because only they can normally see the revdeled material. 1249:

implies that it can be done through consensus, but it doesn't say how the process can be initiated nor where it should happen. I've looked at

654: 5530:

What specifically is the motivation here--to allow flagging down sysops for revdel requests without defacing articles for readers? I like it!

2097:

Before you start a discussion about a "class divide" here, you should go check out the gender divide in the ANI case and consider it.--v/r -

4088: 5049:

with a spamuser template. It was declined which is the first time that has ever happened to me. You can see the discussion about that here

1675:"Material must be grossly offensive, with little likelihood of significant dissent about its removal. Otherwise it should not be removed." 1380: 996: 1766:

clearly says that only clear-cut cases with little likelihood of dissent should be revdeled. A community discussion can indicate dissent.

708:

s edit and RD1s (revision only, not username or edit summary) the three edits where the copyvio was present. Is it necessary to attribute

3965:

I've requested lots of RevDels for copyvio issues in the past - originally with the template, then via use of Enterprisey's RevDel tool.

2532:, where there's a lower traffic among non-admins and a higher traffic among admins. This is also in line with similar processes, such as 202:

avoid obvious suggestive terms in the reason (e.g. don't use "RD4", "oversight", "private material", "hiding IP of logged out user" etc.)

3033:

such a thing. That being said, I believe this will largely be a technical issue, which is something I'm not fit to get involved with. --

1608: 1574: 3701:

Why is a mouseover needed, when the "weird grey lines" are things like the text "Edit summary deleted" that have been struck through? —

47: 17: 5106:

rights/copyright issue, and think at least as a precaution, it's fair game for revdel RD1 (or maybe RD5 if you're being creative). --

4591:

Under what circumstances may revdel be used to delete "ordinary" vandalism that does not otherwise specifically fall under one of the

5474: 4319:) where I quoted all of the places in the policy where it specifies the tool is not for removing "ordinary" incivility. He replied ( 2623: 2574: 2457: 2400: 2025: 1925: 1773: 4549: 661:'s analysis is correct. To meet the attribution requirements, "X, Y and Z wrote this article" is sufficient. The clumsy wording on 4399: 3230:

Best practices for copyrighted text removal can be found at WP:Copyright problems and should take precedence over this criterion.

3218:

Best practices for copyrighted text removal can be found at WP:Copyright problems and should take precedence over this criterion.

377:. I suppose if it's personal info that's revealing enough for an admin to revdel, it's worth notifying the user by email, IMO. -- 3738:

But to the point: I have no moral qualms with this warning doubling as a set of instructions, but I'm curious if anyone does. --

3661:"To avoid the Streisand effect, there is no dedicated on-wiki forum for requesting revision deletion under other circumstances." 1061:- I can never remember the IWL for that.) When content is rev-deleted, it's obvious. When it's selectively deleted, it's not. -- 5695: 5057:'s reasoning but, as they suggested getting a second opinion I wanted to double check on this situation. Thanks for your time. 4269: 3915: 3854: 3672: 3610: 3305: 2358: 2133: 1631:

Q2: If no consensus exists on whether or not RevDel'd content should remain redacted, then it will default to: remain deleted.

988: 5665: 4378:

bot notifications about topics describing the decline of United States society with political and/or sarcastic comments (e.g.

3971:

Tracking down one of the admins seems a bit tough, since I don't know who is online at any given moment and I don't use IRC.

854: 555: 5038: 3888: 3822: 3569: 3483: 3430: 3377: 3250: 3148: 1437: 5681: 5382: 5320: 5315:

has a note at the top saying "The fastest way to request oversight is to email the oversight team." if that means anything.

5286: 5262: 5190: 5160: 4592: 4506: 4343: 4340: 4337: 4334: 4331: 2415: 2172: 1730: 1525: 1376: 1246: 1189: 739: 627: 543: 416: 3187: